<%BANNER%>

The power of food labels

Permanent Link: http://ufdc.ufl.edu/UFE0041946/00001

Material Information

Title: The power of food labels Marketing environmental impacts and animal welfare on meat labels as gains versus nonlosses and the influence on attitudes and voting intentions
Physical Description: 1 online resource (162 p.)
Language: english
Creator: Abrams, Katherine
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: animal, aversion, chicken, environment, focus, food, framing, labels, loss, meat, organic, regulatory, welfare
Agricultural Education and Communication -- Dissertations, Academic -- UF
Genre: Agricultural Education and Communication thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Consumers receive information about how their food is (or is not) produced on a regular basis through the labels they see in the grocery store. Production labeling claims like eco-friendly, cage-free, and no hormones offer information about the product they are on and about the conventionally produced products that do not carry these claims. The theories of loss aversion and regulatory focus suggest that messages, such as food production claims, can be framed as gains or nonlosses and have different persuasive effects, but the theories contradict each other. This study used an experimental design with a convenience sample of 660 college students to examine how consumers? attitudes toward food products are affected by gain- and nonloss-framed production labeling claims about animal welfare and environmental impact and whether this on-package marketing can also affect intent to support an animal welfare ballot initiative. The results did not reveal different attitudinal effects between gain- and nonloss-framed production claims as predicted by loss aversion and regulatory focus theories; however, the presence of the production claims did significantly reduce positive attitudes toward the product without claims. Exposure to the production claims increased positive attitudes toward the product they were on, but these attitudes did not translate into intentions to support the animal welfare ballot initiative. Over 75% of the sample indicated they intended to support the policy regardless of the treatment. This study attempted to frame nonlosses and gains equivalently, but qualitatively. The results suggest that in the absence of numbers or quantifiable information, the biases of loss aversion, framing effects, and regulatory focus fit effect are minimized. Regardless of how production claims were framed, it is clear that they are a source of information affecting consumers? attitudes towards conventional agriculture products and perhaps even the production system. Agricultural communicators should not underestimate the effects that food marketing and advertising can have on consumers? attitudes toward conventional agriculture and its products, and consider these effects in addition to messages put forth by activist groups and mass media.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Katherine Abrams.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Irani, Tracy A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041946:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041946/00001

Material Information

Title: The power of food labels Marketing environmental impacts and animal welfare on meat labels as gains versus nonlosses and the influence on attitudes and voting intentions
Physical Description: 1 online resource (162 p.)
Language: english
Creator: Abrams, Katherine
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: animal, aversion, chicken, environment, focus, food, framing, labels, loss, meat, organic, regulatory, welfare
Agricultural Education and Communication -- Dissertations, Academic -- UF
Genre: Agricultural Education and Communication thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Consumers receive information about how their food is (or is not) produced on a regular basis through the labels they see in the grocery store. Production labeling claims like eco-friendly, cage-free, and no hormones offer information about the product they are on and about the conventionally produced products that do not carry these claims. The theories of loss aversion and regulatory focus suggest that messages, such as food production claims, can be framed as gains or nonlosses and have different persuasive effects, but the theories contradict each other. This study used an experimental design with a convenience sample of 660 college students to examine how consumers? attitudes toward food products are affected by gain- and nonloss-framed production labeling claims about animal welfare and environmental impact and whether this on-package marketing can also affect intent to support an animal welfare ballot initiative. The results did not reveal different attitudinal effects between gain- and nonloss-framed production claims as predicted by loss aversion and regulatory focus theories; however, the presence of the production claims did significantly reduce positive attitudes toward the product without claims. Exposure to the production claims increased positive attitudes toward the product they were on, but these attitudes did not translate into intentions to support the animal welfare ballot initiative. Over 75% of the sample indicated they intended to support the policy regardless of the treatment. This study attempted to frame nonlosses and gains equivalently, but qualitatively. The results suggest that in the absence of numbers or quantifiable information, the biases of loss aversion, framing effects, and regulatory focus fit effect are minimized. Regardless of how production claims were framed, it is clear that they are a source of information affecting consumers? attitudes towards conventional agriculture products and perhaps even the production system. Agricultural communicators should not underestimate the effects that food marketing and advertising can have on consumers? attitudes toward conventional agriculture and its products, and consider these effects in addition to messages put forth by activist groups and mass media.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Katherine Abrams.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Irani, Tracy A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041946:00001


This item has the following downloads:


Full Text





THE POWER OF FOOD LABELS: MARKETING ENVIRONMENTAL IMPACTS AND
ANIMAL WELFARE ON MEAT LABELS AS GAINS VERSUS NONLOSSES AND THE
INFLUENCE ON ATTITUDES AND VOTING INTENTIONS




















By

KATHERINE M. ABRAMS


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2010

































2010 Katherine M. Abrams





























To all of my boys, Brian, Wrigley, and Copper









ACKNOWLEDGMENTS

This doctoral dissertation may signal the completion of my 24 years of formal

education, but the journey led by curiosity and learning will continue for the rest of my

life. That journey is inspired by many people in my life who I'd like thank for their

support, friendship, guidance, and love.

My dissertation committee members were assets in my development as a

researcher and scholarly writer. I thank Lyle Brenner for teaching me about some of the

most fascinating consumer psychology theories. His class and dissertation guidance

helped position me well for broader fields in communication and increased my

confidence in experimental methodology. Paul Monaghan imparted on to me his

impeccable insight into the complexities of conservation behavior. I hope we can work

together in the future because his community-based social marketing research

fascinates me. Ricky Telg was a source of positive encouragement throughout my

program at the University of Florida in roles as a teacher and researcher. He helped me

maintain my sanity through some of the toughest dilemmas and inspired me to be a

strong leader in any academic role I may take on.

I have many faculty in the Department of Agricultural Education and

Communication to thank. Glenn Israel was one of the best teachers I have ever had, in

his class and through his help on my research. Brian Myers was always willing to

answer my methodological questions when I needed a quick answer. Ed Osborne was

caring and supportive throughout my program and helped me secure my first job.

Hannah Carter was a faculty member who I could be completely open and honest with

and count on for great words of encouragement and advice.









Courtney Meyers has and continues to be a positive role model, friend, and

colleague in agricultural communications. She inspired me to continue on for my PhD

and to be a teacher and researcher. I could always count on her as a listener, adviser,

research buddy, conference hotel roommate, and amazing friend. I envision us

remaining close for our entire lives, helping each other grow professionally all the while

remaining good friends.

Tracy Irani has been my academic and career adviser and constant teacher

throughout my graduate program. She is one of the main reasons I have been

successful and will continue to take on academic challenges. Her willingness to share

research opportunities with me gave me an excellent learning experience that will

continue throughout my career in academia. I can only hope to be as successful and

well-respected as she is in our field.

I'd like to thank Lauri Baker for her friendship and guidance as well. She helped

me get my dissertation idea on paper and was always willing to answer my questions or

verify my thinking. She has been a great workout buddy and friend. We not only work

great together, but share a strong bond that will last a lifetime.

Meredith Cochie, where has she been all my life? I have her to thank for an

amazing friendship, laughter, adventures, gripe sessions, sarcasm, and confidence.

She has a lot of faith in me, and I in her. I know we will continue to be best friends and

colleagues, and maybe one day we'll share that backyard and academic department

again. We can only hope, but god save the community!

My family (Mom, Dad, and Scott) has always been supportive of my goals and life

decisions. Even though I may seem very independent in a lot that I do, I will always









need them for advice and support. I know how proud they are of my accomplishments,

and they are the ones I have always been trying to impress my entire life. My newer

family, Brian's family, is a welcome addition in my life. I cannot thank them enough for

"adopting" me with open arms, love, and support.

Last, but certainly not least, I thank my husband, Brian. Thank you for your

confidence, encouragement, and pride in me throughout my graduate education. We

are about to embark on a crazy journey together as a young couple in our first house, in

a new place, and in new jobs. No matter what this first step leads to, I couldn't imagine

doing it without you. I am so lucky to have you in my life as my husband and best friend.









TABLE OF CONTENTS

page

A C KNOW LEDG M ENTS ....... .. ...................................... ... ................... ............... 4

LIST OF TABLES .......... ..... ............................................. 10

LIST O F FIG URES........................................... ............... 12

LIST OF DEFINITIONS ................ .......... ......... ......... 13

ABSTRACT ............... .. ............................... ......... 14

CHAPTER

1 INTRODUCTION ................ .......... .......... ......... 16

Types of Food Labeling .......... ...... ................ ......... 19
O organic and Natural Foods M market ................ ........ ........ ............ .. .............. 21
Consumers' Considerations in Purchasing Food with Credence Attributes ............ 23
Motivators of "Green" Food Consumerism ........... .............. ...... ....... ...... .. 26
Personal Health .......... ... .. .......... ....................... 26
Environmentalism .................. ...... .......... ...................... 27
Animal Welfare ............. ................... .... ........ 28
Political Actions Affecting Meat Production ........... ............................... ....... 29
Loss Aversion ....................... ... ................... 33
Regulatory Focus Framework.................................... ......... 35
Purpose and Objectives......................... ....... ......... ............... 36

2 LITERATURE REVIEW ............... ..... ................ ......... 39

Cognitive Biases in the Construction of Choice ............ ............. ................ 39
Prospect Theory .......... ......................... .... .............................. 41
Loss Aversion: A Bias and Construct in Prospect Theory................................. 44
Framing Effects........................................ .......... 47
Risky Choice Framing Effects ........... .............................. ..... ......... 49
G oal Fram ing Effects......... ....... .................... ................ ............... 49
Regulatory Focus Theory.......................................... ................ ............... 51
Sources of Regulatory Focus ........................... .................. 52
Regulatory Focus in Consumer Decisions............................ ............... 53
Regulatory fit effect ............... .......... ............_.... .... 53
Moderators of regulatory focus effects ......... ................................... 55
Implications for Loss Aversion .......... ............... ....... ................56
Summary of Regulatory Focus Theory.................... .......... ...................... 57
Measuring Framing Effects through Attitudes........................ ...................... 57
Context of the Theoretical Research .................................. ... .............. ............... 59
S u m m a ry .......... .................... ........................... .......................................... 6 1









3 METHODOLOGY ................ ......... .......... ........ 66

Research Design ......................................... .............................. ....... ......... 67
Controlling Threats to Internal and External Validity...................... ......... 68
Subjects................................................ 70
Independent Variables ................................ .............................. 72
Regulatory Focus ............ ......... ................... ......... 72
Pretesting of Message Stimuli ............. ................................................ 73
Dependent Variables ........... ......... ............................. 75
A ttitudinal M measures ................................. ......... .............................. 75
Voting Behavior ......... .......................................................................... 76
Attribute Variables ................ ...... .......... ......... 76
Instrumentation ................ .... ......... ..... .......... 77
Instrument Content........................................... ............... 77
P ilo t T e s t ............................................ .... ........................................ 7 9
P ro c e d u re .................................................................................................. 8 0
D ata A na lysis .......................................... .................................. 8 1

4 R ESU LTS .............. ........... ................................................... ....... 87

D escriptive A nalysis..................................... ............... 87
D e m o g ra p h ic s ............... ................ ............... ...... .............. ........................ 8 7
Attention to and Purchase Frequency of Meat/Poultry with Production
C la im s .......................................... .................. ............... 8 9
Scale Reliabilities...................... ......... 90
Regulatory Focus Scales.............. ........ .................... 90
A ttitu d e S ca le s ................. ... ........................ .................... ......................... 9 1
Descriptive Analysis of Variables of Interest ................................................ 91
Regulatory Focus ........... ......... ................... ......... 91
Attitude Toward Product ................... ............................. 92
Voting Intention ............. .... .... ........... ....................... 92
Manipulation Checks ................................................... 92
Tests of Hypotheses ................ ........ ......... ............... 93
Post Hoc Analyses ............... ........... ................... ... ............ 96
Attention To and Purchase of Products With Production Claims................... 97
Community Upbringing and Livestock Background ......................... ..... 97
Political Affiliation and Voting Intention............................... ............... 98
Other Non-Significant Relationships....................................... 99

5 CONCLUSION.. ............................................................ .............. 110

Overview ..................... ...................... ............... 110
Key Findings ................ ......... ................... 111
Im plications................................. ............... 113
Theoretical ............................. ............... 113
P ra ctica l.............................................................................................. 1 17
Limitations.................... ....................... 120









Recommendations ................. ....... .................... 122
For Future Research and Theory ...... ...................... .............. 122
For P practitioners ........... ... ................ ........... ............... ........... 124
C o n c lu s io n s .............. ...... ........... ......................................................................... 1 2 5

APPENDIX

A VERBAL PRE-NOTIFICATION ........................... .................... 127

B FIRST CONTACT E-MAIL SENT TO SUBJECTS............................................. 128

C FIRST AND SECOND REMINDER E-MAIL SENT TO SUBJECTS ................ 129

D THIRD REMINDER E-MAIL SENT TO SUBJECTS......................................... 130

E IN STR U M ENT ............. ........ ... ................ .......... ............... ......... 131

LIST OF REFERENCES .......... ............ ......... ................ ............... 150

BIOGRAPHICAL SKETCH ............... ............ ... ......................... 162









LIST OF TABLES


Table page

3-1 Incomplete factorial research design ............................................................. 82

3-2 Regulatory focus questionnaire (Higgins et al., 2001)...... ......... ...... ........ 82

3-3 Results of nominal group assessment................................. ..... 83

3-4 Hedonic/utilitarian sources of attitude scale (Batra & Ahtola, 1991) and
researcher-developed measures........._.............. ............ ............... 84

3-5 Experiment treatment groups ....................... ................... ........ ....... 85

4-1 Attention to selected production labeling claims on meat/poultry .................... 100

4-2 Purchase frequency of meat/poultry with select production labeling claims..... 100

4-3 Promotion focus scale inter-item consistency statistics.............. ............ 100

4-4 Prevention focus scale inter-item consistency statistics.............. ............ 100

4-5 Total attitude scale inter-item consistency statistics............... ...... ....... 101

4-6 Regulatory focus questionnaire descriptive statistics ............... ................. 101

4-7 Attitude toward product (product specific* + general attitude) ........... .......... 102

4-8 Attitude toward product grand means among treatment groups.................... 103

4-9 Effects of labeling claim frame on attitudes toward product with claims........... 103

4-10 Planned comparisons t-test for differences between treatment groups on
attitude toward product with claims.............................. ............... 103

4-11 Effects of labeling claim frame on attitudes toward product with claims........... 104

4-12 Planned comparisons t-test for differences between treatment groups on
attitude toward product without claims................................ 104

4-13 Independent samples t-test for differences between subjects exposed to
production claims and subjects exposed to general product claims ................ 104

4-14 Pearson product moment correlations between grocery shopping behavior
and attitude toward product with production claims .................... ... ........... 104

4-15 Pearson product moment correlations between grocery shopping behavior
and attitude toward product without production claims................. .............. 105









4-16 Mean attitude toward products between community upbringing ................. 105

4-17 Effects of community upbringing on attitude toward product without claims..... 105

4-18 Post hoc comparisons between community upbringing................ ............... 105









LIST OF FIGURES


Figure page

1-1 Cover of Time magazine on August 21, 2009 .............................................. 38

1-2 Humane Farm Animal Care's label ........ ............................. .... ........... 38

2-1 Value function as proposed by prospect theory ......... .. ....... .............. 62

2-2 Value function under prospect theory with reference to gains/non-gains and
losses/non-losses ........ ............. ............. .. .. ............ ............... 62

2-3 Risky choice framing paradigm............. .............. ........... ............... 63

2-4 Goal fram ing paradigm ................................................................................ 63

2-5 Website evaluations as a function of situational prime and regulatory focus...... 64

2-6 Regulatory focus theory conceptual model ................................................. 65

3-1 Operational framework for the current study ................ ............. ............... 86

4-1 Subjects' weekly meat consumption................ ............ ...... ......... 106

4-2 Means between the attitudes toward product with claims in each treatment
g ro u p ........ ............. .. .. ......... .. .. ......... ..................................... 1 0 7

4-3 Means between the attitudes toward product without claims in each
treatm ent group. ...................... ........... .. .. ............ ...... ......... 108

4-4 Means between the attitudes toward product without claims in each
treatm ent group. ...................... ........... .. .. ............ ...... ......... 109









LIST OF DEFINITIONS


Credence Attribute






Factory Farm



Organic Agriculture







Production Claim


Promotion Focus




Prevention Focus




Regulatory Focus


Quality attributes of a product that cannot be assessed by the
consumer before or after use and affect products' perceived quality
only so much as consumers' trust in the claims; for example, a label
on a pork product indicating "raised under environmentally friendly
practices" would be considered a credence attribute (Darbi & Karni,
1973).

A term proliferated by animal and environmental activists to
describe conventional livestock production and concentrated animal
feeding operations.

"A type of agriculture that promotes the use of renewable resources
and management of biological cycles to enhance biological
diversity, without the use of genetically modified organisms, or
synthetic pesticides, herbicides, or fertilizers. Organic livestock
production promotes concern for animal welfare, without the use of
synthetic foodstuffs, growth hormones, or antibiotics." (Eicher,
2003)

A claim referring to how a food product was produced before
slaughter or harvest.

Refers to a cognitive mechanism in which people view their goals
as accomplishments, hopes, and aspirations (ideals or maximal
goals), and are sensitive to the presence or absence of positive
outcomes, or gains and non-gains.

Refers to a cognitive mechanism in which people are more
concerned with safety, responsibilities, and obligations (oughts or
minimal goals), and are sensitive to the absence or presence of
negative outcomes, or non-losses and losses.

A cognitive style/mechanism that regulates how people attend to
information.









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

THE POWER OF FOOD LABELS: MARKETING ENVIRONMENTAL IMPACTS AND
ANIMAL WELFARE ON MEAT LABELS AS GAINS VERSUS NONLOSSES AND THE
INFLUENCE ON ATTITUDES AND VOTING INTENTIONS

By

Katherine M. Abrams

August 2010

Chair: Tracy Irani
Major: Agricultural Education and Communication

Consumers receive information about how their food is (or is not) produced on a

regular basis through the labels they see in the grocery store. Production labeling

claims like eco-friendly, cage-free, and no hormones offer information about the product

they are on and about the conventionally produced products that do not carry these

claims. The theories of loss aversion and regulatory focus suggest that messages, such

as food production claims, can be framed as gains or nonlosses and have different

persuasive effects, but the theories contradict each other. This study used an

experimental design with a convenience sample of 660 college students to examine

how consumers' attitudes toward food products are affected by gain- and nonloss-

framed production labeling claims about animal welfare and environmental impact and

whether this on-package marketing can also affect intent to support an animal welfare

ballot initiative.

The results did not reveal different attitudinal effects between gain- and nonloss-

framed production claims as predicted by loss aversion and regulatory focus theories;

however, the presence of the production claims did significantly reduce positive









attitudes toward the product without claims. Exposure to the production claims

increased positive attitudes toward the product they were on, but these attitudes did not

translate into intentions to support the animal welfare ballot initiative. Over 75% of the

sample indicated they intended to support the policy regardless of the treatment.

This study attempted to frame nonlosses and gains equivalently, but qualitatively.

The results suggest that in the absence of numbers or quantifiable information, the

biases of loss aversion, framing effects, and regulatory focus fit effect are minimized.

Regardless of how production claims were framed, it is clear that they are a source of

information affecting consumers' attitudes towards conventional agriculture products

and perhaps even the production system. Agricultural communicators should not

underestimate the effects that food marketing and advertising can have on consumers'

attitudes toward conventional agriculture and its products, and consider these effects in

addition to messages put forth by activist groups and mass media.









CHAPTER 1
INTRODUCTION

Now that I know how supermarket meat is made, I regard eating it as a somewhat
risky proposition ...so I don't buy industrial meat (Michael Pollan, 2004, T6).

What the well-known author of The Omnivore's Dilemma has to say about meat

from conventional agriculture -that eating it is a risky proposition- is a viewpoint

growing in popularity (DeGregori, 2003). This stems from uncertainty about how animal

agriculture production practices, such as administering subtherapeutic antibiotics,

confining livestock in crates or cages, and concentrating animals in large numbers,

might affect human health, animal welfare, and the environment (DeGregori, 2003;

Hughner, McDonagh, Prothero, Shultz, & Stanton, 2007). What makes Michael Pollan

different from most people is that he has visited many livestock production facilities

while doing research for his books and Washington Post articles, whereas the majority

of Americans have probably only seen a snapshot of animal agriculture from their car

windows at 65 miles per hour.

Science literacy and communication research consistently shows that most people

get their science information-which includes agricultural sciences-from news and

entertainment media (Nisbet et al., 2002). Many U.S. consumers are likely getting their

perspectives on the United States agricultural system from journalists and authors like

Michael Pollan, from talk shows like Oprah and Ellen, and from the news media. These

perspectives from mass media are not typically favorable toward conventional

agriculture as evidenced in movies like Food, Inc., Fast Food Nation, and King Corn;

and in books like The Omnivore's Dilemma and Chew on This. A recent cover story in

Time magazine entitled "Getting Real About the High Price of Cheap Food" (see Figure

1-1) represents what is typical in this type of coverage. These mass-mediated channels









portray large, conventional farms as having negative characteristics that are detrimental

to human health, the environment, and animal welfare.

The United States' agricultural system has intensified over time as a result of

technological and market forces, urban/suburban sprawl, and a decrease in interest of

farming as an occupation (Fitzgerald, 2003; Sassenrath et al., 2008). Livestock

production, in particular, is highly associated with trends toward greater farm

concentration and corporate industrialization, due in part to urban encroachment,

government policies, and the geographic availability of feed (Lobao & Meyer, 2001;

Morrison, Nehring, Banker, & Somwaru, 2004). These external pressures have led to

greater human input and control of food animals' lives from conception to slaughter.

The intensification of production was noted by activist groups who began referring

to large, conventional farms as "factory farms." According to the Oxford English

Dictionary, the first documented use of the term is attributed to a journal of economics in

1890, but started to proliferate in publications in the 1960s (Factory, 1989). Views of this

technologically advanced production system have changed since the 1950s. During that

decade, Americans viewed products of such a production system more favorably for the

system's ability to provide convenience foods that saved the housewife time and money

(Levenstein, 2003). The 1950s were a time in which consumers were focused on the

"relentless pursuit of convenience" (Levenstein, 2003, p. 101). This drove food

producers and manufacturers to develop additives to aid in processing and preserving,

to develop growth additives for animal feed, and to create concentrated animal feeding

operations (again, referred to as "factory farms" by activists). These were accepted

innovations in agriculture because it meant cheaper and more convenient foods, like TV









dinners and meat with every meal (Hooker, 1981; Levenstein, 2003). "Perhaps it was

natural that, in an era when Americans brimmed with confidence in the superiority of

their political, economic, military, and even cultural institutions, they should feel similarly

about their food and those who produced it" (Levenstein, 2003, p. 118).

Acceptance and confidence in large, conventional agriculture production waned in

the late 1960s. In 1969 and 1970, "calls for a return to natural foods resonated far from

the hippie enclaves, striking sympathetic chords among the kind of thoughtful middle-

class Americans" (Levenstein, 2003, p. 195). In 2002, the market for organically

produced meat and produce-which is viewed as more natural (Abrams, Meyers, &

Irani, 2009)-increased dramatically with the creation of the United States Department

of Agriculture (USDA) certified organic label. Today, people tend to view natural foods

more favorably than those produced with human or technological intervention (Rozin et

al., 2004).

While U.S. consumers' preference for natural and organic foods continues to grow

stronger, agriculture is still intensifying production practices with selective breeding,

medical and feed technologies, and advanced mechanical systems (Fitzgerald, 2003).

With these continued changes has come increased concern about the healthiness and

safety of meat, poultry, eggs, and dairy for human consumption; environmental impacts;

and animals' welfare in such an agricultural system. These concerns have steadily risen

in the U.S., as demonstrated through changes in legislation (states banning cages and

crates in hogs, layers, and veal), food labeling (certified organic label in 2002, naturally

raised claim in 2009), and growth of the market for products promoting the absence of

those perceived risks (Greene et al., 2009).









Types of Food Labeling

A variety of labeling claims are used by food marketers to differentiate products

and communicate quality and value to consumers. Among the mix of labels are those

addressing consumer concerns regarding food production practices. Some examples

include the organic label, natural or all-natural, free range, humanely raised, eco-

friendly, no hormones, and no antibiotics. All of these are practices and labeling claims

are voluntary. Production labeling claims, which refer to how the food was produced

pre-harvest or how the animal was raised, are regulated by the USDA Food Safety and

Inspection Service (FSIS). This entity "develops and provides labeling guidance,

policies and inspection methods and administers programs to protect consumers from

misbranded and economically adulterated meat, poultry, and egg products [to] ensure

that all labels are truthful and not misleading" (USDA FSIS, T1, 2009). Processing

labeling claims, which refer to how the food is altered post-harvest or post-slaughter

(i.e., additives, preservatives, coloring), are also regulated by the USDA FSIS.

Perhaps one of the more extensive set of policies and inspection programs

created by the USDA is the National Organic Program. An increase in consumers'

interest and confusion about organic products during the 1990s led to the institution of

the USDA National Organic Program in October 2002 (California Institute for Rural

Studies, 2005). These standards were established to assure consumers that so-labeled

products are produced, processed, and certified to meet the consistent national organic

regulations (National Organic Program, 2002). The standards provide a set of

guidelines for food to be labeled "organic" that affect the growing, handling, and

processing of the food. For organic meat production, the standards prohibit the use of

antibiotics and growth hormones, require animals to be fed 100% organic feed, and









require animals to have access to outdoors and access to pasture for ruminants. The

organic label is considered a certified label because an inspector visits the farm yearly

and on an unannounced basis to certify that the farm's practices are meeting the USDA

organic standards.

The separate certification process is what makes the organic label different from

most production claims. The USDA FSIS has created some claims with specific

guidelines, including free range, free roaming, natural, no hormones, and no antibiotics.

Other claims producers and marketers want to put on their products-which may be

entirely different or variants on those claims (i.e., raised outdoors as opposed to free

range)-may be submitted for approval as well. Production claims are upheld by USDA

FSIS policies, but are approved differently than those that are certified, like the organic

label. The producer or marketer submits the claim to FSIS, along with supporting

documentation (operational protocol, affidavits, and testimonials), and the request to

use the labeling claim is either approved or denied; however, the operation is never

physically inspected before approval or denial (USDA FSIS, n.d.). When USDA

inspectors conduct an annual random inspection of the entire operation for adherence

to required practices, they will check to ensure the operational protocol meets what they

agreed upon to qualify for the voluntary production claim.

The USDA is not the only entity involved in labeling claim creations and operation

inspections. Third-party organizations (i.e., Certified Angus Beef and America Grassfed

Certified) that operate a product or service certification system can be approved by the

USDA International Organization for Standardization Guide 65 (ISO Guide 65) Program

to certify operations for meeting the third party's voluntary production or product









standards to qualify for their label. The ISO Guide 65 Program ensures that third-party

certification programs are applying their standards in a consistent and reliable manner

(USDA Agricultural Marketing Service, 2008). One example of such a program is

Humane Farm Animal Care, which provides certification for their Certified Humane label

(see Figure 1-2). Their standards are highly specific and vary by species (layer hens,

broilers, dairy cows, etc.) and apply from birth through slaughter. In general, this label

means the animal had ample space, shelter, and gentle handling to limit stress; no

hormones or antibiotics in their feed; were not kept in cages, crates or tie stalls; and

were free to engage in natural behaviors such as dust bathing for chickens and rooting

for pigs (Humane Farm Animal Care, n.d.).

Many consumers and researchers alike tend to lump all food with production

and/or processing labeling claims into the category of "organic and natural foods" for

ease of communication, even though the USDA has specific definitions of the terms

"organic" and "natural." More recently, when researchers discuss organic food in the

United States, they are referring to food produced according to the National Organic

Program standards with the USDA label, but the term "natural" in reference to food still

tends to be a catchall term referring to both the processing of the meat and how the

livestock were raised (Abrams et al., 2009; USDA FSIS, 1999). This is important to aid

in the interpretation of the research regarding the market for such foods.

Organic and Natural Foods Market

The $21.1 billion organic food industry in the United States (Greene et al., 2009) is

growing due to consumer concerns about food safety, particularly regarding pesticides,

antibiotics, growth hormones, and genetic modification (Hwang, Roe, & Teisl, 2005).

The organic label distinguishes foods as free of those perceived risks, while other









products attempting to appeal to these consumer concerns use production and

processing claims such as all-natural, no antibiotics, no hormones, and free range.

These products often have a "no" labeling theme or communicate essentially the same

message by saying the livestock or poultry were produced without one of these

perceived risks (Abrams et al., 2009). In July 2009, a large market research company

report indicated that organic food will be the fastest growing food trend over the next

decade with a growth rate of 41% (The NPD Group, Inc., 2009). Organic meat and

poultry is one of the fastest growing segments of the organic food market and is

predicted to grow 27% annually through 2010 (Storck, 2008). A nationwide survey

conducted by the American Meat Institute (AMI) & Food Marketing Institute (FMI) (2008)

found 19% of shoppers had purchased organic or natural meats in the past three

months; however, many were not sure if the meat they purchased was organic or

natural.

The rapid growth of the market for these types of products suggests that a large

segment of consumers have come to value production and processing attributes of the

food they buy (Caswell, 1998; Thompson & Troester, 2002). These value perceptions

have led to increased market diversity as more niche products emerge to fulfill

consumers' needs. Absent of brand differentiation, consumers were willing to accept

more uncertainties and a lack of understanding about production characteristics of food

products to a greater degree (Levenstein, 2003). Consumers now have a heightened

awareness of food production and health factors as a result of recent food scares (e.g.,

2008 beef recall due to downer cattle, 2009 salmonella in peanuts), increased media









coverage of these topics (Craven & Johnson, 1999), and societal shifts in values

(Caswell & Mojduszka, 1996).

Consumers' Considerations in Purchasing Food with Credence Attributes

Today, a consumer may pick up a generic package of fresh chicken and notice the

one next to it advertising no antibiotics, no hormones, and free range. Consumers with

different preferences, including different risk preferences, will rationally choose different

bundles of attributes in foods. Consumers will buy products that give them the most

value in terms of costs and benefits, as long as they are able to accurately judge the

quality attributes (Caswell & Mojduszka, 1996). Quality cues at the point of purchase

are most often extrinsic qualities, such as brand, labeling, and price. During handling

(i.e., cooking, preparing) or consumption is when quality cues are intrinsic attributes

(Ziethaml, 1988). Consumers make assumptions about intrinsic values based on

information from extrinsic cues (Olson, 1978). For example, an organic label on a bag of

potato chips may generate the belief that "it is probably healthier for me." Consumers

often rely on extrinsic attributes in initial purchase situations when they cannot evaluate

the relevant intrinsic attributes of a product or when evaluation of intrinsic cues requires

more input than the consumer perceives is worthwhile (Ziethaml, 1988).

In most cases, production and processing claims are quality attributes that cannot

be assessed by the consumer before or after use. Darbi and Karni (1973) call these

credence attributes. Whether or not credence attributes signal high quality depends on

consumers' trust of the claims. If consumers do not trust production or processing

claims, then they do not signal good quality. For example, a claim on a package of beef

indicating "from cattle raised under environmentally friendly practices" would be

considered a credence attribute; the claim, therefore, is referred to as a credence









attribute. Unless the consumer goes to the farm where that cow came from to determine

the environmentally friendliness of the practices, he or she will have to trust the claim to

derive utility from it. Several studies have found themes of consumer distrust in

credence claims (Abrams et al., 2009; Bruns0, Fjord, & Grunert, 2002; Padel & Foster,

2005; Yiridoe, Bonti-Ankomah, & Martin, 2005).

Despite some consumer distrust of the labeling claims, credence attributes are

becoming more important in the set of considerations consumers make when trying to

determine food quality because the perceived benefits outweigh the trustworthiness

element. The total food quality model developed by Bruns0, Fjord, and Grunert (2002)

states that consumers evaluate expected quality on four levels: taste, health,

convenience, and process. Process refers to how the animal was raised and also how

the meat was processed (presence of additives, preservatives, etc.) and it is the most

relevant component of the model to this study. In the model, "quality is not an aim in

itself, but is desired because it helps satisfy purchase motives or values. The model

therefore includes motive or value fulfillment, i.e. how food products contribute to the

achievement of desired consequences and values" (Bruns0 et al., 2002, p. 9-10). Bech-

Larson, Grunert, and Poulson (2001) conducted a study assessing consumer choice of

food products marketing different health attributes, such as Omega 3's, as simply

present in the product or combined with the specific health benefits of that attribute.

They found that when food products are marketed based on these credence attributes,

quality perception becomes a function of communication effectiveness. The

effectiveness of that communication depends on the credibility consumers assign to it,

consumers' motivation to process the information, and their ability to understand it









(Bech-Larson et al., 2001; Bruns0 et al., 2002). Thus, what the label communicates and

how it fits with the shopper's goals is critical to persuading him or her to purchase the

product.

Previous research has shown consumers tend to prefer organic foods, foods that

were produced in a way that attenuates perceived risks, and foods that appeal to certain

value sets (Loureiro, McCluskey, & Mittelhammer, 2005; Yiridoe et al., 2005); however,

price is usually the primary barrier between attitudes and purchasing behavior (Padel &

Foster, 2005). If organic meat was the same price as conventional meat, the large

majority of consumers (95.3%) would purchase it (AMI & FMI, 2008). Outside of price,

convenience, and habit, when purchasing these types of meat products, several levels

of outcomes are considered; among them are personal health, the environment, and

animal welfare (Yiridoe et al., 2005). The latter two are ethical considerations. The

environment could also be a health consideration, depending on individual

environmental values. Ethical considerations, such as the confinement of livestock in

crates and cages, are gaining importance but typically rank below health and meat

safety risk perceptions; however, pork and poultry come up most often when consumers

perceive risks to animal welfare (Verbeke & Viaene, 1999).

One recent survey found consumers purchase organic and natural meat for a

variety of reasons, the top three being: 1) positive long-term personal health effects

(47.2%), 2) better nutritional value (47.2%), and 3) better health and treatment of the

animal (40.4%). The reduced environmental impact was ranked sixth with nearly 31% of

consumers indicating it as a motivation to buy organic or natural meat (AMI & FMI,

2008). Surveys and polls like this do not shed light on the intricacies of the consumer









decision-making process because researchers determine how the choice items are

chosen and written. Looking at the response items in the AMI and FMI (2008) study,

"positive long-term personal health effects" is framed in a way that suggests this goal for

eating organic and natural meat is an attempt to approach a positive outcome for their

health. The response item "reduced environmental impact" is framed in a way that

suggests the goal is to avoid a negative outcome for the environment. The way

researchers choose to frame their questions and response items can affect how

respondents answer them (Levin, Schneider, & Gaeth, 1998).

Motivators of "Green" Food Consumerism

"The environmental ethic that gained worldwide prominence with Earth Day 1990

placed emphasis on individual responsibility for personal health and social action on

environmental quality and animal welfare" (Yiridoe et al., 2005, p. 196). In the midst of a

strong environmental movement (Dunlap & Mertig, 1992; Gottlieb, 2005), a health foods

craze (Dubisch, 2004; Nestle, 2007), and a powerful animal rights movement (Rollin,

1990, 2003), meat seems to represent a consumer commodity and issue through which

people can demonstrate their values and goals for their health, the environment, and

food animals with little direct involvement in a movement or overt campaign.

Personal Health

First and foremost, food safety is the top concern fueling people's positive

attitudes toward organic and natural foods. Consumers want to be assured their food is

safe, and organic food is often equated with safer food. Perceptions of food safety and

risk typically relate to concern about food production technologies. In the United States,

concern is highest for pesticides and hormones, followed by antibiotics, genetic

modification, and irradiation (Hwang, Roe, & Teisl, 2005). Labels and claims are used









by marketers to appeal to those concerns. Food safety is a concern because unsafe

food could potentially negatively affect a consumer's personal health (Bruns0, Fjord, &

Grunert, 2002). In other words, personal health concerns can be a function of food

safety or nutrition.

Environmentalism

Eighty-three percent of Americans would agree that global warming is a serious

problem and 81% feel it is their responsibility to reduce the impacts of global warming

(Yale Center of Environmental Law and Policy's Environmental Attitudes and Behavior

Project, 2007). Environmental sentiments have been on the rise, but clearly, not all

Americans hold the same levels of environmentalism.

Researchers have attempted to clarify different value orientations toward the

environment. Kempton, Boster, and Hartley (1995) found "environmental values are

already intertwined with core American values, such as religion and parental

responsibility" (p. 13). Kempton et al. (1995) found environmentalism is built upon

cultural models of how nature works and how humanity interacts with it, and is

motivated by environmental values. Americans tend to idealize the environmentalism of

simpler times and desire to return to that more natural way of life. Environmental values

include humanity's utilitarian need for nature, obligations to future generations, the

spiritual or religious value of nature, and for some, the rights of nature in and of itself

(Kempton et al., 1995).

Because most Americans feel some sense of responsibility to the environment,

marketers have begun to recognize the need of environmental or green marketing

(Sheth & Paravatiyar, 1995; Grant, 2008).









Green issues and marketing can work against each other. One wants you to
consume less, the other more. One rejects consumerism, the other fuels it. But
they aren't always opposed. Marketing can help 'sell' new lifestyle ideas. It's a
much-needed function today, when we all need to act fast to mitigate the effects of
climate change (Grant, 2008, Chapter 1: T1)

Purchasing meat with credence attribute marketing claims regarding the environment

(i.e., environmentally-friendly, good for the environment) is a relatively simple behavior

for consumers to reinforce environmental values. While consumers generally have

positive attitudes toward such foods, the difficulty in persuading people to purchase

them is that they often are priced at a premium and consumers' are hesitant to believe

their purchase will have an impact (Vermier & Verbeke, 2006). Marketing can help close

the gap between attitudes and behavior if the right messages are used. Green

marketing often sounds like a good idea, but if it is not based on good intentions and a

deep understanding of consumer decision making, then it will not work (Grant, 2008).

Animal Welfare

Animals are often seen as a part of nature or at least similar to the natural

environment, especially in how people view their purpose. Like nature, animals have

some intrinsic value, but generally a utilitarian value, especially when it comes to

livestock. In the United States, people desire some protection of farm animals, whether

that be based on their intrinsic or utilitarian values. Munro (2005) distinguished between

animal welfare and animal rights. Animal rights refers to the idea that animals, like

humans, have innate rights and interests that should not be compromised for human

benefit. Animal rightists do not believe humans should farm animals at all and often

promote vegetarianism and veganism (Munro, 2005). Animal welfare represents a

balance between human and animal interests and refers to the idea that animals should









not be treated cruelly or in a way detrimental to their health and well-being (Munro,

2005).

The notion of animal rights is often seen as too extreme for most Americans;

however, most support the notion of animal welfare (Garner, 1993). Support or activism

in animal welfare and animal agriculture issues (among others) can occur at many

levels, from participation in an animal protection group to private behavior such as

consumption choices (Seguin, Pelletier, & Hunsley, 1998). In the sphere of individual

behavior, a consumer will likely choose a product associated with improved animal

welfare or production if they somehow feel responsible and/or that their choices will

make a difference (Blandford, Bureau, Fulponi, & Henson, 2002; Vermeir & Verbeke,

2006).

Political Actions Affecting Meat Production

If some consumers are not already "voting with their dollar" to voice support for

alternative livestock production practices, they are supporting state legislation in the

voting booth on initiatives advocated by the Humane Society of the United States and

other well-funded opponents of conventional practices. In Florida, Arizona, and

California, voters have overwhelmingly supported a policy banning common methods of

animal confinement for pregnant pigs, egg-laying hens, and veal calves.

The animal agriculture industry tends to blame animal agriculture opponents, such

as People for the Ethical Treatment of Animals (PETA) and the Humane Society of the

United States (HSUS), for misleading consumers, voters, policymakers, and the media

on issues regarding animal welfare, the healthiness of meat products, and the

environmental impacts of conventional practices (Crowell, 2009; Downing, 2009;

Gabbett, 2008; Goodwin & Rhoades, 2009; Smith, 2009). The HSUS Factory Farms









campaign website has 31 secondary research reports on the industry's detriments to

animal welfare, eight on environmental impacts, and 13 on human health that it widely

distributes to policymakers and corporations (HSUS, 2008; E. Williams, personal

communication, December 4, 2008). These reports are not necessarily misleading but

show that these organizations are attempting to implicate animal agriculture in

detrimentally affecting human health, the environment, and animal welfare. As with any

controversial topic, each side in the debate carefully selects sources and evidence that

supports their perspective on the issue. The HSUS is known for campaigning heavily for

animal agriculture industry reform, using emotional appeals and more persuasive

message strategies than the industry groups like Farm Bureau and the Animal

Agriculture Alliance (Abrams & Meyers, 2009; Goodwin & Rhoades, 2009).

Answering whether the public's support of policy initiatives on livestock care is

evidence of the animal agriculture opponent groups' successful campaigning or

Americans' evolving value-systems regarding livestock production would be like

answering the chicken or the egg conundrum. It is likely that animal agriculture

opponents are more successful as a direct result of changing values and less familiarity

with farming, especially livestock production. In the midst of a strong environmental

movement (Dunlap & Mertig, 1992; Gottlieb, 2005), a health foods craze (Dubisch,

2004; Nestle, 2007), and a powerful animal rights movement (Rollin, 1990, 2003), meat

and livestock production seem to represent a consumer commodity and issue through

which people can demonstrate their values and goals for their health, the environment,

and food animals. "The environmental ethic that gained worldwide prominence with









Earth Day 1990 placed emphasis on individual responsibility for personal health and

social action on environmental quality and animal welfare" (Yiridoe et al., 2005, p. 196).

Within the industry, segments and individuals regard organic agriculture as

another foe of the conventional industry (Obach, 2007) because organic products are

often touted as better in many dimensions, including taste, nutritional value, and

sustainability (Organic Trade Association, 2008). However, whether organic food

actually delivers on these desires and beliefs is controversial and the subject of a

scientifically inconclusive debate (Obach, 2007). A review of 162 studies conducted

over 50 years found that organic food had no nutritional or health benefits over

conventional food (Dangour et al., 2009). A USDA publication reviewing several studies

comparing organic to non-organic agriculture production did find that, generally (with a

few exceptions), organic agriculture has several environmental advantages in "a)

maintaining or building soil quality, b) lessening ground and surface water

contamination, c) reducing greenhouse gas emissions, d) encouraging biodiversity, e)

conserving water and energy resources, and, f) recycling waste" (Gold, 2010, "Find Out

More. Issues and References, number 3," T 1).

Despite the scientific debate, consumers have come to believe in the superiority of

organic and more naturally produced foods. The Harris Poll found that more than three-

quarters of the U.S. public believes organic food is safer for the environment (79%) and

healthier (76%) than conventional foods ("Harris poll results," 2007). The price and

intense marketing of organic and other value-added animal products likely

communicates to the consumer that they are indeed better than their conventional

counterparts (Klonsky & Tourte, 1998). Higher prices and levels of advertising often









trigger a placebo effect in which consumers believe those products are of higher quality,

and subsequently, they have better experiences with the products than those less

advertised and/or with lower prices (Shiv, Carmon, & Ariely, 2005).

While some in the agriculture industry may still see organic agriculture as a

detriment to the conventional industry, today, many producers and companies have

embraced this niche market. This resulted in diversified production practices and

purchases of organic farms and brands to capture a piece of the premiums consumers

are willing to pay for these products and the positive corporate reputation that comes

from being attached to an initiative that is supposedly better for animal welfare, the

environment, and human health (Guthman, 2004).

Although the industry often points to animal agriculture opponent groups for the

shift in people's thinking about what is acceptable in livestock production practices in

the United States, marketing organic and more naturally produced products as better

than unlabeled ones may have unintended consequences. The messages consumers

receive in the grocery store week after week are likely far more memorable and

pervasive than what the HSUS puts in a video on YouTube or in an ad before a vote on

a ballot initiative. Consumers receive multiple exposures, which are more salient than a

single or few exposures to TV or Web ads/videos, to messages about meat production

through package labeling claims in the grocery store. A 2009 Nielsen poll found 61% of

consumers read food labels. Interestingly, shoppers at Whole Foods, Trader Joes,

Publix, Costco, and Safeway were mostly likely to read labels (Hale, 2009). Jauregi and

Ward (2006) surveyed a little over 14,500 households and found 57% check labels for

harmful ingredients and 60% base their food purchase on using the labels.









With less than 1% of the U.S. population involved in production agriculture (Hurt,

2002), most consumers may only learn of certain production inputs from reading food

labels. The question becomes, are production claims on meat labels affecting what

people believe about the unlabeled product? Limited research has been done to

examine the effects of production labeling claims on consumers' attitudes toward those

that do not carry such claims. Empirical research is needed to determine the effects of

production claims on consumer beliefs about the conventional meat product in the

United States. Such research may shed light on political actions that affect livestock

production, revealing why many consumers are unwilling to pay for product attributes

they perceive to be better, but are willing to support policy that would make such

attributes required of all animal products.

Loss Aversion

Food labels are intended to be persuasive communication to convince consumers

that the product is different and better than others to cause them to purchase the

product (Golan et al., 2001). Many communication and psychology scholars have found

numerous conditions under which persuasive communication is more effective.

Persuasion, or what convinces people to change beliefs, attitudes, and behaviors, is a

function of how our minds naturally want to think. Persuasive communication appeals to

those cognitive preferences. One of those cognitive preferences and "perhaps the most

successful and widely used explanatory construct in behavioral decision research"

(Brenner, Rottenstreich, Sood, & Bilgin, 2007, p. 369) is loss aversion. Loss aversion is

one of the main components of Kahneman and Tversky's (1979) prospect theory; it

shows that losses have a steeper value function than gains. In other words, losses loom

larger than equivalent gains. The concept of loss aversion does not necessarily imply









that people pay more attention to losses over gains but the hedonic reaction to a loss is

stronger than the reaction to a gain (Brenner et al., 2007).

Most studies examining nonloss- versus gain-message framing used quantitative

descriptors (Boettcher, 2004; Idson et al., 2004; Liberman et al., 2005; Kahneman &

Tversky, 1979; McDermott, 2004; Tversky & Kahneman, 1981), but limited research has

tested whether the predictions of loss aversion hold for qualitatively defined

frames/descriptors of equivalent gains and nonlosses. Also, additional research is

needed to test this cognitive preference when the implications of the decision are more

removed or not immediately known, such as consequences for personal health in the

long term, the environment, and animal welfare. The present study intends to test the

loss aversion theory in a context in which the consequences are more removed and not

immediately obvious. These are perceived consequences for animal welfare, personal

health, and the environment as marketed in production and processing claims on meat

labels. Most consumers will never experience the impact of their purchase on those

aspects first-hand or be able to directly attribute potential changes to it but may be more

persuaded by labeling claims that better match their goals in such purchase decisions. If

they are indeed loss averse, then conclusions will reflect that loss aversion is a powerful

cognitive preference that permeates consumer decision-making even when

consequences are more removed and qualitatively described. These findings will offer

suggestions for marketers of organic and natural foods, for environmental educators

and communicators seeking environmentally responsible behavior change in their

audiences, and for extension agents.









Regulatory Focus Framework

Regulatory focus theory (Higgins, 1998) offers another explanation of how

consumers' decision-making and behavior operate that adds to Kahneman and

Tversky's (1979, 1981) loss aversion concept. This theory states that whether negative

information is attended to more and weighted more heavily than positive information

depends on people's goals in the decision. The theory posits that individuals function

according to two different types of motivations based on mental depictions of an end-

state that will result from committing to a particular decision. Individuals using a

promotion focus view their goals as accomplishments, hopes, and aspirations (ideals or

maximal goals), and are sensitive to the presence or absence of positive outcomes, or

gains and non-gains. In contrast, individuals using a prevention focus are more

concerned with safety, responsibilities, and obligations (oughts or minimal goals), and

are sensitive to the absence or presence of negative outcomes, or non-losses and

losses.

One of the fundamental predictions of regulatory focus theory is that individuals

attend to information that is relevant to the activated regulatory focus and that they

weigh attributes compatible with this focus more carefully. This focus can be activated

by persuasive messages such as advertising and product labels. Previous research has

also shown some people are chronically self-regulating, using either the prevention

focus or promotion focus when making decisions (Higgins & Silberman, 1998).

Essentially, it can be a cognitive style; however, cues in the environment may cause a

shift from one regulatory focus to another. Situational variables can cause changes in

sensitivity, emotions, and strategic inclinations that can activate a promotion or

prevention focus (Higgins, 1998).









Idson, Liberman, and Higgins (2000) added to the theory when they found that the

pleasure of a gain (promotion success) is stronger than the pleasure of a nonloss

(prevention success), while the pain of a loss (prevention failure) is stronger than the

pain of a nongain (promotion failure). This is because success in the promotion focus is

achieving a maximal goal, whereas success in prevention is failure to achieve a minimal

goal. This is a different perspective than the predictions of loss aversion, which explains

losses loom larger than corresponding gains. Further research is needed to test how

consumers attend to nonlosses versus gains as previous contradictory evidence has

been found by others as well. Levin, Schneider, and Gaeth (1998) conducted a

comprehensive literature review and found conflicting evidence for loss aversion in

studies testing goal framing effects (gains vs. nonlossses and losses vs. nongains) and

called for more systematic research in this area.

In the context of the present study, the production labeling claims on meat

products often promote the avoidance of negative outcomes with a no labeling theme or

by saying how the raising of the livestock or poultry is absent of perceived risks/negative

outcomes (i.e., no antibiotics, no hormones), while others may have claims focusing on

the positive outcomes that will result from purchasing the product (i.e., eco-friendly,

great care). This experiment will test how participants react to gain versus nonloss

messages via attitudinal measures in a choice situation in which other product features

(price, appearance, cut, and weight) are controlled.

Purpose and Objectives

Food marketers and regulators use labels and claims to differentiate products and

inform consumers about their options. Government food regulators must consider the

effects of food labeling to ensure the policies, standards, and guidelines for such labels









are balancing the market for agricultural products and not misleading consumers

(Golan, Kuchler, & Mitchell, 2001).

While previous research shows there is a preference for products marketed as

improving animal care, personal health benefits, and environmental impacts (also called

credence attributes), no empirical research has examined the effects of marketing these

credence attributes on consumers' attitudes toward products without such attributes.

Messages about these credence attributes can be presented (framed) in different ways

potentially resulting in varying persuasive effects based on biases in people's cognitive

processing. The purpose of this study was to compare the persuasive effects of gain-

and nonloss-framed labeling claims. The research objectives were:

Objective 1: To compare the attitudinal effects of nonloss-framed claims to gain-
framed labeling claims using qualitative descriptors.

Objective 2: To determine whether and how consumers' attitudes toward products
with no claims are influenced by production labeling claims.

Objective 3: To assess whether consumers' voting intention on animal welfare
policy is affected by credence attribute labeling claims.








i-L Ir. .ir;;r]? X Aj-l.ih
TT IA T


Figure 1-1. Cover of Time magazine on August 21, 2009.


CERTIFIED
HUMANE


Figure 1-2. Humane Farm Animal Care's label.


The Real
Cost of
Cheap Food









CHAPTER 2
LITERATURE REVIEW

The previous chapter established the need for the current study and the research

objectives, which are to determine the effects of differently framed labeling claims on

consumers' attitudes toward the credence attribute product, the conventional product,

and voting intention. This chapter provides an overview of research related to consumer

decision-making research with specific emphasis on cognitive biases, prospect theory,

loss aversion, framing effects, and regulatory focus theory. Following this is an

explanation of the study's context in sustainable agriculture food products.

Cognitive Biases in the Construction of Choice

In explaining how people arrive at the choices they make, the rational choice

theory suggests a straightforward path in which people make logical calculations based

on an individually-defined preference ordering system with respect to 1) the benefits of

each alternative, 2) the costs of each alternative in terms of utilities foregone, and 3) the

best way to maximize utility (Simon, 1955). Therefore, actions or choices are a function

of knowledge and a cost/benefit analysis, in which costs are minimized and benefits

maximized. However, we now know that people do not always act or think rationally.

Numerous studies in decision-making have revealed that people can make unexpected

choices based on automatic heuristics and biases and how information and messages

about choice options are framed rather than careful reasoning (for a review see

Gilovich, 1991; Gilovich, Griffin, & Kahneman, 2002). While these simplifying heuristics

often lead to accurate judgments, they also yield systematic error.

Scholars often break down information processing into two parts or two paths: one

involving more systematic, careful processing and the other a more superficial or









heuristic processing of available information. One model in this area is the Heuristic-

Systematic Model (HSM) (Chaiken, 1987; Todorov, Chaiken, & Henderson, 2002). The

HSM describes two cognitive mechanisms called systematic processing and heuristic

processing. When engaging in systematic processing, people scrutinize the available

information to evaluate the validity of the message and form a judgment based on their

elaborations. The systematic mode requires sufficient cognitive resources-which is

affected by distraction, message repetition, time pressure, communication modality, and

knowledge and expertise-and motivation-which is affected by personal relevance,

need for cognition, task importance, accountability for one's attitudes, and exposure to

unexpected message content (Todorov, Chaiken, & Henderson, 2002). When people

are not sufficiently motivated or do not have the necessary cognitive resources, they

use heuristic processing. This mode is more nonanalytic in nature and is characterized

by the use of simple decision rules or heuristics to form a judgment. One common

example is the availability heuristic, which refers to the ease with which a circumstance

comes to mind. Vivid scenarios such as shark attacks tend to be more top-of-mind than

more objectively threatening ones like heart attacks. People may be completely

unaware of their heuristic processing and may even deny that they were influenced by

peripheral informational cues (Todorov, Chaiken, & Henderson, 2002).

Underlying heuristics are cognitive biases. Biases are defined as deviations from

some true or objective value or as violations of basic laws of probability (Gilovich &

Griffin, 2002). Tversky and Kahneman (1983) demonstrated how human judgments

depart from probability theory or simple logic in their famous "Linda problem." In this

experiment, subjects read the following personality description: "Linda is 31 years old,









single, outspoken, and very bright. She majored in philosophy. As a student, she was

deeply concerned with issues of discrimination and social justice and participated in

anti-nuclear demonstrations." They were then asked to determine which of two options

was more probable: (a) Linda is a bank teller or (b) Linda is a bank teller and active in

the feminist movement. Between 80% and 90% of participants tended to select (b) as

the more probable option, even though probability theory dictates a conjunction cannot

be more likely than either of its parts (Tversky & Kahneman, 1983). Although resulting in

systematic errors, cognitive biases allow people to make decisions efficiently and can

lead to correct decisions (Haselton, Nettle, & Andrews, 2005).

Prospect Theory

Prospect theory (Kahneman & Tversky, 1979) was one of the first influential

theories offering a descriptive model of how people make decisions that differed from

normative and rational choice theory. Kahneman and Tversky's (1979, 1981) studies

were able to show (1) that how the outcomes of a decision are framed can affect the

ultimate choice, and (2) that the decision maker's evaluation under uncertainty works on

a value function with three characteristics: diminishing sensitivity, reference

dependence, and loss aversion. The theory "sheds light on the interaction between the

person and the situation in decision-making environments" (McDermott, 2004, p. 293).

Before explaining the constructs of prospect theory, a description of the original work

and later tests of the theory will provide some context.

In the original work by Kahneman and Tversky (1979), they presented findings

indicating people treat losses differently than gains in the context of risky choice

decisions involving monetary outcomes. They found 80% of participants preferred a

certain outcome of $3000 to an 80% chance of $4000 and 20% chance of nothing. This









showed, with respect to gains, people tend to be risk averse. When they reversed the

prospects, 92% of participants preferred to gamble on an 80% chance of losing $4000

and 20% chance of losing nothing to a certain loss of $3000. With respect to losses,

people tend to be risk acceptant. In other words, if a person stands to gain something

for certain, they are less willing to take a risk to gain something more; however, if a

person stands to lose something for certain, they are more willing to take a risk to lose

more (Kahneman & Tversky, 1979).

In later studies advancing prospect theory, Tversky and Kahneman (1981) further

examined how the framing of choices/information affects the decision made in the often

cited "Asian disease problem." Subjects given the positively framed version of a sure

saving of one-third of the lives versus a one-third chance of saving all the lives and a

two-thirds chance of saving no lives chose the option with the certain outcome. Subjects

given the negatively framed version of a sure loss of two-thirds the lives versus a one-

third chance of losing no lives and a two-thirds chance of losing all of the lives chose the

risky option. The outcome of saving one-third of the lives is the same as losing two-

thirds of the lives, but in the positively framed version, that was an acceptable choice,

whereas in the negatively framed version, it was unacceptable (Tversky & Kahneman,

1981).

Boettcher (2004) found support for prospect theory when individuals evaluated a

political decision in the context of a terrorist-hostage situation at a U.S. embassy. In the

gain frame, hostages were rescued or not rescued, and in the loss frame, hostages died

or did not die. However, when subjects came together in a group to make a decision,

the preference reversal did not occur; therefore, support was not found for prospect









theory in group decision making. In a group context, people were more risk acceptant

regardless of the frame (Boettcher, 2004; McDermott, 2004). Numerous studies,

however, have indicated support for prospect theory in a wide variety of domains

including politics, business, finance, management, and medicine (see Maule &

Villejoubert, 2007, for a review).

The implications of prospect theory for the present study suggest individuals may

consider messages about credence attributes of meat products differently depending on

how the information is presented to them. A person's attitudes toward the product may

be stronger if the message about environmental impacts, for example, is framed as

avoiding a loss than if it is framed as achieving a gain for the environment.

Prospect theory (depicted in Figure 2-1) proposed an explanatory model of choice

that deviated from rational choice theory in four important ways, resulting in the main

constructs of the theory: diminishing sensitivity, reference dependence, loss aversion,

and framing effects. What follows is a brief explanation of the constructs relevant to this

study and a deeper explanation of the last two, which will be tested.

Kahneman and Tversky proposed that people make choices regarding gains and

losses in terms of deviations from a reference point, which is usually the status quo, but

can be deviations from an aspiration level or some other reference point (Heath, Larrick,

& Wu, 1999). Soman (2004) explained that values are coded as gains and losses

relative to a reference point, meaning the decision is reference dependant. Looking at

the Asian disease problem in Tversky and Kahneman's 1981 study, in one condition,

"the outcomes are framed in terms of saving lives; the potential disaster of losing all the

lives becomes the neutral reference point" (Soman, 2004, p. 383).









The frame, therefore, can change the perceived reference point of the question. A

frame refers to how information is described and interpreted (Bottecher, 2004).

Credence attribute messages on meat labeling can suggest a reference point of

personal health deterioration (when framed as nonloss), thereby causing people to want

to avoid the potential loss. These framing effects demonstrated the cognitive bias of

loss aversion, in which the notion of a certain loss is more aversive, causing people to

accept a risk (with poor odds to gain) to potentially avoid that loss. Following, is an in-

depth explanation of loss aversion and framing effects, which are the most relevant to

the present study.

Loss Aversion: A Bias and Construct in Prospect Theory

"Loss aversion is perhaps the most successful and widely used explanatory

construct in behavioral decision research" (Brenner et al., 2007, p. 369). As one of the

main components of Kahneman and Tversky's (1979, 1981) prospect theory, it shows

that losses have a steeper value function than gains. Examining Figure 2-1, losses and

non-losses are measured against the steep loss part of the value curve, whereas gains

and non-gains are measured against the shallow part of the value curve. In other words,

losses loom larger than equivalent gains. Loss aversion was originally proposed as an

explanation for the endowment effect, which is what explains people's tendency to place

a higher value on an item that they own than on an identical one that they do not own

(Kahneman, Knetsch, & Thaler, 1990).

The concept of loss aversion does not necessarily imply that people pay more

attention to losses over gains but the hedonic reaction to a loss is stronger than the

reaction to a gain (Brenner et al., 2007). Hedonics refer to the basic human emotions of

pleasurable and unpleasurable states of consciousness (Kahneman, Diener, &









Schwartz, 2003). Imagine the reaction of an elementary school student that loses a gold

star from their publicly displayed achievement tally. It is likely much more of a reaction

than when they received that gold star. Just as "happy marriages can be easily knocked

off-line, but it takes an enormous amount of time, effort, and commitment to repair a

marriage that is breaking apart, and in many instances even that cannot fix what has

already broken" (McDermott, 2004, p. 298). This cognitive bias is why gas stations

advertise a lower price for paying cash or using their credit card, rather than an

advertised surcharge for using a credit or debit card; people are more willing to forego

gains than to accept losses (Thaler, 1980).

Although seemingly irrational in the context of business and market transactions, it
has roots in lower-level psychological laws that seem adaptive to basic
environmental demands. Thus the asymmetry of people's reactions to pain versus
pleasure is eminently sensible in a world that punishes those who ignore danger
signs more than it rewards those who pursue signs of pleasure (Newell, Lagando,
& Shanks, 2007, p. 119).

Many studies have found support for loss aversion (see Levin et al., 1998 for a

review). One of the more recent studies testing loss aversion found a greater hedonic

reaction to losses than to non-gains (supportive of loss aversion) using three different

experimental scenarios, including the following one regarding labor union contract

negotiations:

Gain/non-gain condition: "one of the conditions listed in the proposal is an
increase in employee benefits of approximately $200."

Loss/non-loss condition: "one of the conditions listed in the proposal is a
decrease in employee benefits of approximately $200."

Participants in the gain and loss (non-gain and non-loss) conditions answered the
question "How would you feel if the condition is written (is removed and not
written) into the new contract?" on a scale ranging from -9 (very bad) to 9 (very
good). (Liberman, Idson, & Higgins, 2005, p. 531)









The authors found that gains were perceived as more intensely positive than non-

losses, a result that is opposite to the prediction derived from loss aversion. The figure

used to depict prospect theory and loss aversion in their study includes a depiction of

non-gains and non-losses (see Figure 2-2), which makes this alternative finding more

visible. Similar findings were presented in Idson et al. (2000). Both studies offered

regulatory focus theory as a possible explanation for the findings, although the authors

made no mention of measuring regulatory focus. Regulatory focus theory explains that

how people evaluate and react to gains and losses is controlled by how they envision

the goal or outcome of the decision (Higgins, 1998). The mechanism that moderates

this is called the regulatory focus. These findings suggest an area for continued study in

other contexts before conclusions can be made regarding the explanatory strength of

loss aversion with respect to gains versus non-losses.

In summary, loss aversion theory states that people have stronger reactions to

potential losses than potential gains. The theory predicts the same should hold true for

nonlosses (avoiding a loss) relative to gains (achieving a gain); however, some scholars

(Idson et al., 2000; Liberman et al., 2005) have found inconsistencies with this

prediction. More research is needed to compare the persuasive effects of gain- versus

nonloss-framed messages. Also, most studies examining nonloss- versus gain-

message framing have used quantitative descriptors (Boettcher, 2004; Idson et al.,

2004; Liberman et al., 2005; Kahneman & Tversky, 1979; McDermott, 2004; Tversky &

Kahneman, 1981). Additional research is needed to test whether the predictions of loss

aversion hold for qualitatively defined frames/descriptors of equivalent gains and

nonlosses.









Framing Effects

What Kahneman and Tversky (1979, 1981) and other previously mentioned

researchers have found is that people will arrive at different decisions depending on

how the choice information is framed. Framing, at a basic level, refers to the process

through which individuals or groups make sense of their environment; frames are

cultural structures that organize understanding of social phenomena. "Packets of

incoming information pass through various cognitive, affective, and/or social filters to

produce a 'perception' of the outside world. This construction of reality then drives

judgment and decision-making and ultimately behavior" (Bottecher, 2004, p. 332-333).

Although this may be an internal process, it is often constructed by some external actor

-either deliberately or unintentionally (Bottecher, 2004).

The focus of framing effects in psychology and marketing is slightly different from

media framing and political issue framing. Agricultural communication and general

communication researchers often use a different conception of framing, so it is

important to understand the differences in how framing effects can be operationalized

and studied. The psychology literature's definition of a framing effect is when two

"logically equivalent (but not transparently equivalent) statements of a problem lead

decision makers to choose different options" (Rabin, 1998, p. 36; Tversky & Kahneman

1981). This is called equivalency framing. In media and issue framing effects literature,

an alternative explanation is used. Druckman (2001) described media or issue framing

effects as when "a speaker's emphasis on a subset of potentially relevant

considerations causes individuals to focus on these considerations when constructing

their opinions" (p. 1042). In media and issue framing effects studies, the frames are

rarely logically equivalent. They are often "qualitatively different yet potentially relevant









considerations" (Druckman, 2004, p. 672). Issue and media framing involve the

selection of some aspects of a situation, making them more salient through

communicating text with the idea of advocating a particular solution or interpretation of

the topic (Entman, 1993). For example, the debate on universal health care is typically

framed by one side as "health care is a basic human right" and by the other side as "not

the government's or taxpayers' responsibility."

This study deals exclusively with equivalency framing and its effects on decision

makers, meaning two different, but logically equivalent frames are used. Levin and

Gaeth (1988) offer a good example. They found variation in quality preferences

regarding beef depending on whether a beef product was labeled as being 75% lean or

25% fat. The ground beef was evaluated by subjects as better tasting and less greasy

when it was labeled in the positive light (75%) lean. The common adage that pessimists

see the glass half empty, and optimists see it half full, demonstrates a complimentary

description of the same object that is viewed in two different ways. Objectively, a glass

half empty is a glass half full, but people will make different decisions about that object

depending on how it is presented to them.

"Throughout the literature, valence framing effects, wherein the frame casts the

same critical information in either a positive or a negative light, are often treated as a

relatively homogeneous set of phenomena" explained solely by prospect theory (Levin

et al., 1998, p. 150; emphasis in original). Levin et al. (1998) organized and interpreted

past literature on framing effects to explain contradictory and weak support of prospect

theory, thereby demonstrating the existence of different types of framing effects with

different underlying mechanisms and consequences.









Risky Choice Framing Effects

Framing, as defined by Tversky and Kahneman (1981), is a "the decision-maker's

conception of the acts, outcomes, and contingencies associated with a particular

choice" (p. 453). This means the choice involves their perceptions of the courses of

action, the outcomes associated with the alternative, and the likelihood associated with

the outcomes. To study framing effects in the vein of prospect theory, one would set up

an experiment as outlined in Figure 2-3; however, many recent studies of framing

effects have "deviated greatly from the operational definitions and theoretical concepts

used in the original studies" (Levin et al., 1998, p. 151).

The framing effects studied in prospect theory are what they call the risky choice

framing paradigm (Figure 2-3) in which the outcomes of a potential choice involving

options of differing risk levels are described/framed in different ways. In this type of

framing, risk preference is affected as seen in Kahneman and Tversky's (1979, 1981)

original studies. Overall, the evidence from multiple studies on framing effects in the

risky choice paradigm show a relatively consistent tendency for people to be more risk

acceptant when the options are framed to focus attention on the chance to avoid losses

than when options focus on the chance to realize gains (Levin et al., 1998).

Goal Framing Effects

Goals themselves can "govern or 'frame' what people attend to, what knowledge

and attitudes become cognitively most accessible, how people evaluate various aspects

of the situation, and what alternatives are being considered" (Lindenberg & Steg, 2007,

p. 119). Goal framing effects refer to the impact of persuasion depending on how a

consequence or implied goal of a behavior is framed (see Figure 2-4). What is different

about goal framing is that both frames should enhance the evaluation of the issue. It is a









matter of determining which type of goal-to avoid a loss or achieve a gain/benefit

(Levin et al., 1998).

Levin, Gaeth, Evangelista, Albaum, and Schreiber (2001) directly tested goal

framing effects in the context of reducing red meat consumption with American and

Australian subjects. The manipulation was:

Positive frame condition: If you discontinue eating red meat you will be able to
reduce the level of cholesterol in your blood. Thus, you will significantly decrease
the likelihood of the early onset of heart disease.

Negative frame condition: If you continue eating red meat you will not be able to
reduce the level of cholesterol in your blood. Thus, you will fail to significantly
decrease the likelihood of the early onset of heart disease.

Participants in each condition were then asked to write a number between 0 and
100 to indicate how likely they are to eliminate red meat from their diet, and to
write a number between 0 and 100 to indicate how likely they are to reduce by at
least 1/3 the amount of red meat in their diet. (Levin et al., 2001, p. 66)

American subjects rated the complete elimination of red meat and the reduction of red

meat significantly higher in the negative frame condition; however, this effect was not

significant for Australian subjects. Several other studies have generally found a similar

loss aversion bias in which avoiding a loss is greater than the desire to obtain a gain of

So for example, a meat product with credence attribute claims framed as avoiding loss

or damage to personal health, animal welfare, and the environment may create a

different attitudinal response in comparison to claims framed as achieving gains or

repairing personal health, animal welfare, and the environment.

Later research found this loss aversion bias in goal framing effects can be

mitigated by age (Shamaskin, 2009), involvement (Maheswaran & Meyers-Levy, 1990;

Miller & Miller, 2000), and culture (Levin et al., 2001) such that that those higher in age

and those higher in involvement are more influenced by positive frames.









Levin et al. (1998) found the evidence for goal framing is less homogenous than

for risky choice and attribute framing and called for more research in this area of

framing effects. Furthermore, the findings explained earlier from Idson et al. (2000) and

Liberman et al. (2005) revealed inconsistencies with loss aversion with respect to gains

versus non-losses. These scholars and others have developed theory to explain a

cognitive style underlying how people process information and their goals, called

regulatory focus.

Regulatory Focus Theory

Regulatory focus theory (Higgins, 1998) offers another explanation of how

consumers' decision-making and behavior operate that adds to Kahneman and

Tversky's (1979, 1981) loss aversion concept. This theory states that whether negative

information is attended to more and weighted more heavily than positive information

depends on people's goals in the decision, which is controlled by the individual's

regulatory focus. The theory posits that individuals function according to two different

types of motivations based on mental depictions of an end-state that will result from

committing to a particular decision. Those with a prevention focus will regulate their

behaviors away from negative outcomes, while those with a promotion focus will

regulate their behaviors toward positive outcomes (Higgins, 1998).

Individuals using a promotion focus view their goals as accomplishments, hopes,

and aspirations (ideals or maximal goals), and are sensitive to the presence or absence

of positive outcomes, or gains and non-gains. When the end-state is desired/positive,

individuals are said to have an approach goal. Approach goals are achieved by

maximizing the presence or minimizing the absence of positive outcomes (Higgins,

1998; Aaker & Lee, 2001). For example, an environmentally conscious consumer may









wish to improve the environment (desired end-state) and purchase meat with credence

attributes (strategy that maximizes the presence of a positive outcome). In contrast,

individuals using a prevention focus are more concerned with safety, responsibilities,

and obligations (oughts or minimal goals), and are sensitive to the absence or presence

of negative outcomes, or non-losses and losses. When the end-state is

undesired/negative, individuals are said to have an avoidance goal. Avoidance goals

are achieved by minimizing the presence or maximizing the absence of negative

outcomes. For example, an environmentally conscious consumer may wish to avoid

damaging the environment (undesired end-state) by purchasing meat with credence

attributes (strategy that minimizes the presence of a negative outcome). The examples

used with meat purchasing demonstrate that people can envision their goals for the

environment slightly differently (improve or repair vs. avoid damage) but still use the

same strategy (purchase meat with credence attributes) to attain the goal. Products, like

meat with credence attributes, can be "regarded as a means to approaching a positive

outcome or avoiding a negative one" (Florack, Scarabis, & Gosejohann, 2005, p. 240).

Sources of Regulatory Focus

Regulatory focus is affected by three sources: chronic regulatory focus of the

decision maker, contextual priming during or before the decision task, and the decision

task itself (Florack et al., 2005). A chronic regulatory focus is determined by caretaker-

child interactions. A child's behavior regulated by positive reinforcement increases their

sensitivity to promotion goals, whereas negative reinforcement increases their

sensitivity to prevention goals (Higgins & Silberman, 1998; Higgins, 1998). "Like other

motivational orientations, regulatory focus may vary between individuals not only

dispositionally, but also momentarily" and independently of the chronic focus (Florack et









al., 2005, p. 237). Situational variables can cause changes in sensitivity, emotions, and

strategic inclinations that can activate a promotion or prevention focus (Higgins, 1998).

This is shown in studies through which subjects are primed to adopt a promotion or

prevention focus. The priming is typically done through having individuals complete a

thought-listing activity. Freitas and Higgins (2002) offer the following induction script:

Promotion: Please think about something you ideally would like to do. In other
words, think about a hope or aspiration that you currently have. Please list the
hope or aspiration below.

Prevention: Please think about something you think you ought to do. In other
words, think about a duty or obligation that you currently have. Please list the duty
or obligation below (Freitas & Higgins, 2002, p. 3).

A decision task or consumer good can also be associated with a certain regulatory

focus. Zhou and Pham (2004) found participants who made prevention-related

investment decisions were more likely to have adopted a prevention focus as indirectly

measured through participants choosing a product with prevention claims. In examining

regulatory fit effects, Florack and Scarabis (2006) discovered that sunscreen is a

product category that prompts a prevention focus. Limited research has been done to

determine what other decision tasks and consumer goods can be associated with a

particular regulatory focus.

Regulatory Focus in Consumer Decisions

Regulatory focus theory has been tested in a number of consumer decision

making contexts to determine the robustness of the theory and continue its expansion.

What follows is a review of the key studies in this area.

Regulatory fit effect

One of the fundamental predictions of regulatory focus theory is that individuals

attend to and have stronger hedonic reactions to information that is relevant to the









activated regulatory focus and that they weigh attributes compatible with this focus more

carefully (Higgins, 2002). This is called the regulatory fit effect. Florack and Scarabis

(2006) found a relationship between an advertising claim and consumers' regulatory

focus had an impact on product preferences. Their study compared participants that

were primed to adopt a promotion focus or prevention focus, and their preferences for

sunscreen, based on the packaging claims on the bottle. One set of claims emphasized

a promotion focus ("Enjoy the sun") while the other emphasized a prevention focus

("Give sunburn no chance"). Participants who were asked to think about negative

(positive) things they try to avoid (pursue) while on vacation showed a stronger

preference for a brand with a prevention-focused claim (promotion-focused claim). The

product was a means to achieve the goals of approaching a positive outcome (getting a

tan) or avoiding a negative outcome (sunburn) (Florack & Scarabis, 2006). As

mentioned in the discussion on sources of regulatory focus, they found sunscreen is a

product category that prompts the prevention focus. Their manipulation checks,

however, did show successful priming.

One explanation for the regulatory fit effect "is that the fit of the message with a

person's regulatory focus leads to enhanced persuasion because individuals evaluate

messages more positively when they are in line with their attitudes, motivations, and

needs" (Florack et al., 2005, p. 244). The perception of fit may be used as a heuristic

and lead to biased message processing. Regulatory fit evokes a feeling of importance

or "feeling right," which gets interpreted as a positive evaluation (Higgins, 2002).

Furthermore, it increases the recipient's engagement with the message and is

perceived as more persuasive (Cesario, Higgins, & Scholer, 2008). If the consumer's









attention to meat with credence attributes prompts a particular regulatory focus, then the

messages that fit that focus would be the most appealing.

Moderators of regulatory focus effects

Wang and Lee (2006) looked at how regulatory focus theory affects consumers'

evaluations of products to determine if advertising promotion and prevention messages

simultaneously would enhance or diminish persuasion. In addition, they examined the

effects of involvement on the regulatory fit effect. They found subjects in the low-

involvement condition place more weight on features that fit their regulatory focus when

reviewing both fit and non-fit product feature claims. Timing subjects' evaluations of the

product feature claims showed those primed with a prevention focus spent more time

looking at the prevention claims, while those with a promotion focus spent more time on

promotion claims. Again, this only occurred in the low-involvement condition. In the

high-involvement condition, subjects spent about the same amount of time on both

types of claims. They did show that their evaluation of the products was driven more by

their perceived attractiveness of the features than by the extent of processing. That is

important because one could have argued that their preference for the product with

feature claims that fit their regulatory focus was a function of (mediated by) time spent

processing that information. In sum, "people rely on their regulatory focus as a filter to

process information selectively to construct their preferences when cognitive resources

are limited" (Wang & Lee, 2006, p. 36).

The research on regulatory fit effect demonstrates that it is more reflective of

heuristic versus systematic processing. Evan and Petty's (2003) finding that the

regulatory fit effect is moderated by need for cognition is consistent with Wang and

Lee's (2006) findings regarding involvement.









In a study aptly titled "I seek pleasures and 'we' avoid pains: The role of self-

regulatory goals in information processing and persuasion," Aaker and Lee (2001) found

individuals who viewed themselves as independent were more persuaded by

promotion-focused product information, whereas, those who viewed themselves as

interdependent were more persuaded by prevention-focused product information.

Subjects evaluated messages on the Welch's Grape Juice website. The independent

variable of self-view was manipulated in two ways: (1) picture focusing on an individual

or family, and (2) text that emphasized the individual ("you," "your") or the

interdependent-self ("family"). The independent variable of regulatory focus was

manipulated through product claims about Welch's Grape Juice. The promotion focus

emphasized messages its ability to increase energy. The prevention focus messages

emphasized its ability to prevent cancer and heart disease (Aaker & Lee, 2001). Figure

2-5 illustrates their findings that promotion information appeals more to the independent

self-view. Aaker and Lee's (2001) findings were important because they demonstrated

that accessible self-view moderates the persuasiveness of promotion-/prevention-

focused messages that could easily be manipulated through advertising and a strategy

to approach different audience segments (families vs. individuals).

Implications for Loss Aversion

Other researchers (Idson et al., 2000; Idson, Liberman, & Higgins, 2004; Liberman

et al., 2005) added to the theory when they found that the pleasure of a gain (promotion

success) is stronger than the pleasure of a non-loss (prevention success), while the

pain of a loss (prevention failure) is stronger than the pain of a non-gain (promotion

failure) (see p. 44 of this document for a complete description of the experiment).









Regulatory focus theory predicts that because promotion success (gain) is
success in achieving a maximal goal (a standard one hopes to achieve), it should
be experienced more intensely than prevention success (nonloss), which is
success in achieving a minimal goal (a standard one must achieve) (Liberman et
al., 2005, p. 269).

The pleasure of a gain being stronger than the pleasure of a non-loss is a different

perspective than the predictions of loss aversion, which explains losses loom larger

than corresponding gains. Idson et al. (2000), however, did not directly examine the

predictions derived from loss aversion, where as Idson et al. (2004) and Liberman et al.

(2005) did. They suggested "that more caution is needed in using [loss aversion] to

explain decision making phenomena in economics, political science, and social

psychology," particularly when examining gains versus non-losses (Liberman et al.,

2005, p. 534). Furthermore, they called for more research to determine whether

regulatory focus effects overwhelm the predicted loss aversion effect for gains versus

non-losses, which is what the present study intends to address.

Summary of Regulatory Focus Theory

The studies presented on regulatory focus theory demonstrate that decision

makers evaluate information that fits their focus more favorably than information that

does not. People tend to elaborate, better-understand, and "feel right" when presented

with information in tune with their regulatory focus (Cessario et al., 2008). Regulatory

focus theory offers implications for the design of messages as evidenced in Aaker and

Lee's (2001) self-view study and Liberman et al.'s (2005) study of gains versus non-

losses. Figure 2-6 provides a conceptual model of regulatory focus theory.

Measuring Framing Effects through Attitudes

Studies in agricultural economics often examine different ways of labeling food to

determine people's preferences (Hu, Woods, & Bastin, 2009). They determine









preference by measuring people's willingness to pay (WTP) for a product with certain

attributes. The concept of a preference is, in some ways, the counterpart in economics

to the concept of an attitude in psychology, "but the logic of attitudes and the logic of

preferences are quite different" (Kahneman & Sugden, 2005, p. 164). Preferences are

subjective, but their logical structure is objective. If a consumer prefers a ground beef

product that is 25% fat, they should prefer a product that is 75% lean. Attitudes are not

objective in structure; therefore, a consumer might have a negative attitude toward a

ground beef product that is 25% fat but a positive attitude toward one that is 75% lean.

The occurrence of framing effects does not violate the logic of attitudes as it does the

logic of preference (Kahneman & Sugden, 2005). Preferences are best measured by

making people choose between two options, while attitudes are best measured by

affective responses to a single object. Attitudes have a reasonable amount of stability.

"This stability of attitudes lends some stability to the choices that people make, but

attitudes are also susceptible to a lot of manipulations that are not allowed to have any

effect in a rational theory of preferences" (Kahneman & Sugden, 2005, p. 165).

Therefore, a framing effect should yield a change in attitude.

An attitude is defined as an association between an object of thought and a

valence evaluation with three components: cognitive, emotional, and behavioral

(Ostrom, Bond, Krosnick, & Sedikides, 1994). Cognitive responses are based on

beliefs, inferences, knowledge, and assumptions about the attitude object. Emotions are

the feelings connected to thinking or experiencing an attitude object. Behavior is the

action or actions taken in response to the attitude object (Ostrom et al., 1994). Similarly

to Ostrom et al. (1994), Batra and Ahtola (1991) state that "consumer attitudes have









distinct hedonic and utilitarian components" (p. 168). The hedonic component refers to

affective/emotional gratification from consumption behavior. The utilitarian component

refers to the instrumental, practical reasons. Attitude, therefore, can be measured

through utilitarian and hedonic descriptors.

Context of the Theoretical Research

The term "sustainable agriculture" is often used to incorporate the dimensions of

personal health (food safety), the environment, and animal welfare. It is difficult to define

because both conventional and organic agriculture attempt to frame their practices as

sustainable. Definitions of sustainable agriculture vary widely. A basic, conservative

definition is:

The primary goals of sustainable agriculture include: (1) providing a more
profitable farm income; (2) promoting environmental stewardship, including
protecting and improving soil quality, reducing dependence on non-renewable
resources, such as fuel and synthetic fertilizers and pesticides, and minimizing
adverse impacts on safety, wildlife, water quality and other environmental
resources; (3) promoting stable, prosperous farm families and communities.
(Sustainable Agriculture Research & Education [SARE], n.d., T3)

It is also defined as "a way of raising food that is healthy for consumers and animals,

does not harm the environment, is humane for workers, respects animals, provides a

fair wage to the farmer, and supports and enhances rural communities" (Sustainable

Table, n.d., T1). Even those using conventional agricultural practices could argue that

they are sustainable whether they ascribe to either definition. These two definitions may

lead one to conclude that 'the devil is in the details' and sustainability is in the eye of the

beholder. Regardless, most people have strong, pleasurable associations with the idea

of sustainable agriculture (Williams & Wise, 1997); therefore, products marketed on

dimensions of sustainability benefit from those associations.









The problem with the marketing of these food products is that it could suggest the

unlabeled or conventionally-produced foods are inferior and from unsustainable

agricultural systems. The United States government frames the organic label as a

"marketing label," and rejects the idea that organic food production would have relative

advantages to the environment, health or food quality (Bostr6m & Klintman, 2003). The

organic label and production claims are not meant to differentiate the food as safer, but

unintentionally, they may have. Government regulations have typically been used to

distinguish between safe and unsafe foods; therefore, organic standards could give

consumers the impression that conventionally produced foods are unsafe (Klonsky &

Tourte, 1998). In addition, the price and intense marketing of organic and other value-

added animal products likely communicates to the consumer that they are indeed better

than their conventional counterparts (Klonsky & Tourte, 1998). Higher prices and levels

of advertising often trigger a placebo effect in which consumers believe those products

are of higher quality, and subsequently, they have better experiences with the products

than those less advertised and/or with lower prices (Shiv et al., 2005). Food regulators

need to have an understanding of how production claims labeling affects consumers'

beliefs about meat in order to balance the market for such products and avoid

misleading consumers. Research investigating whether consumers' attitudes toward

conventionally produced products are affected by production claims and how these

attitudes might translate into behavioral intent (e.g., intent to support an animal welfare

ballot initiative) has yet to be done. Such research may also shed light on political

actions that affect livestock production, revealing why many consumers are unwilling to









pay for product attributes they perceive to be better, but are willing to support policy that

would make such attributes required of all animal products.

The attitudes toward products from sustainable agricultural systems are typically

positive. Consumers' reasons for preferring meat products from such agricultural

systems are: (1) health and nutritional benefits, (2) improved animal welfare, and (3)

decreased environmental impact (AMI & FMI, 2008, Yiridoe et al., 2005). What is

unknown is how consumers envision these three goals: are they trying to approach

positive outcomes, or avoid negative outcomes? Testing the theories of loss aversion,

and regulatory focus in the context of food labeling offers implications for marketing

sustainable agricultural products and the market for all agricultural products.

Furthermore, this application context can be used to test whether regulatory focus

effects can overwhelm the predictions of loss aversion, and how the predictions of loss

aversion hold when communicating gains and nonlosses qualitatively.

Summary

The review of the literature outlined in this chapter provided an overview of biased

information processing with an emphasis on loss aversion, framing effects, and

regulatory focus theory. Current gaps in the literature illustrate the need to further

explore message framing effects of gains versus nonlosses communicated qualitatively

and the effects of credence attribute labeling on consumers' attitudes toward products

without such claims and voting intention on an animal welfare ballot initiative.











value
AL


losses


convex
sector


concave
sector


-p


gains/
losses


\ point of reference


gains


Figure 2-1. Value function as proposed by prospect theory. Obtained from Jacob and
Ehret (2006).


subjective value


+ objective
value


Figure 2-2. Value function under prospect theory with reference to gains/non-gains and
losses/non-losses. Obtained from Liberman et al. (2005).


//














SURE THING
OPTION


POSITIVE
FRAME







NEGATIVE
FRAME

L.


RISKY
OPTION


COMPARE TO
SURE THING RISKY DETERMINE
OPTION OPTION FRAMING EFFECT


SOME LOST
FOR SURE


CHANCE NONE LOST PREFERENCE
wrrH PREFERENCE
CHANCE ALL LOST


Figure 2-3. Risky choice framing paradigm (Levin et al., 1998)


BEHAVIOR
X


POSITIVE :. k .:.: "i..
FRAME
FRAME OBTAIN GAIN .. ""' i

[APPROACH] (APPROACH BEH X)> :':.: "
-77.777 = ; ... .


BEHAVIOR
NOT-X


RATE OF BEH X



COMPARE TO
DETERMINE
FRAMING EFFECT


NEGATIVE :.': :' ..
FRAME ':..'.: ,.:'. i,: SUFFER LOSS -
[AVOID] "' "':i'", .i' (AVOID BEH NOT-X)[



Figure 2-4. Goal framing paradigm (Levin et al., 1998).










* Promotion focus
-- Prevention focus


Web site
evaluations

4.8

4.4

4.0 K

3.6

32


Independent Interdependent

Type of situational prime

Figure 2-5. Website evaluations as a function of situational prime and regulatory focus
(Aaker & Lee, 2001).













Promotion Focus


Chronic Priming Product or
Focus Thinking Decision
* Positive of hopes Promotion-
reinforce- and ideals oriented
ment from
caretakers


Approach Desired End-State
* Presence orabsence of positive outcomes
* Gainsand non-gains
* Eagerness


Maximal Goal
* Hopes
* Aspirations
* Ideals


Framing and persuasion effects
* Promotion-framed messages are more
persuasive
* Activated independent self-viewsare more
influenced by pronioionr-rrainied messages
* Low-involved people are more influenced by
promotion-framed messages


Prevention Focus


Chronic Priming Product or
Focus Thinking of Decision
* ',eillive dutiesand Prevention
reinforce- obligations -oriented
ment from
caretakers


Avoid Undesired End-State
* Presence orabsence of negative outcomes
* Lossesand non-losses
* Vigilance


Minimal Goal
* Safety
* Responsibility
* Obligations


Framing and persuasion effects
* Prevention-framed messages are more
persuasive
* Activated interdependent self-views are more
influenced b, pre- eriIcn-r r inied messages
* Low-involved people are more influenced by
prevention-framed messages


Framing effects independent of regulatory focus
* Highly-involved .jepie equ.Iiiv con'ila-r promotion and
prevention messages
* Galnsare experienced more intensely than non-losses
* Losses are experienced more intensely than non-gains


Figure 2-6. Regulatory focus theory conceptual model.









CHAPTER 3
METHODOLOGY

The purpose of this study was to compare the persuasive effects of gain- and

nonloss-framed labeling claims. The objectives of this study were to determine the

effects of differently framed labeling claims on consumers' attitudes toward the

credence attribute product, the conventional product, and voting intention. The theories

of loss aversion (Tversky & Kahneman, 1981) and goal framing effects (Levin et al.,

2001) predict that losses and potential losses garner a stronger hedonic reaction than

gains; therefore, avoiding a loss should yield a stronger response than achieving a gain.

Although two studies have suggested gains are reacted to more strongly than

nonlosses (Idson et al., 2005; Liberman et al., 2005) and offer the regulatory focus

theory as an explanation, the literature testing and supporting the predictions of loss

aversion is far more extensive. However, to ensure regulatory focus is not affecting the

attitudinal response, subjects' chronic regulatory focus will be measured and controlled

statistically. Subsequently, the following hypothesis is offered:

H1: When controlling for regulatory focus, subjects exposed to nonloss-framed
claims will have more positive attitudes toward the product with production claims
than those exposed to gain-framed labeling claims or control group claims.

The literature has suggested the intense marketing of sustainable agriculture

products or food products with production claims could communicate that the unlabeled

or conventionally produced foods are inferior and from unsustainable agricultural

systems (Klonsky & Tourte, 1998Therefore, in examining the effects of production

claims on attitudes toward conventional products that do not have production claim

labeling and subsequent voting behavior on an animal welfare ballot initiative, the

following hypotheses are:









H2: When controlling for regulatory focus, subjects exposed to nonloss-framed
claims will have less positive attitudes toward the product without production
claims than those exposed to gain-framed labeling claims or control group claims.

H3: Subjects exposed to a food product with production claims and a product
without such claims will have less positive attitudes toward the product without the
claims than those who do not see a food product with production claims.

H4: Subjects exposed to a food product with production claims will be more likely
to have intentions to vote "yes" for an animal welfare ballot initiative than those
who do not see a food product with production claims.

The experiment focused on determining how an individual's attitudes toward meat

products are influenced by differently framed credence claims and how voting behavior

on animal welfare initiatives is influenced. A subject's regulatory focus (prevention,

promotion) and the two independent variables (presence or absence of production

labeling claims, frame of production labeling claims) should lead to different effects on

attitudes toward the product with the claims and toward the product without the claims

and the subject's intent to support an animal welfare ballot initiative. See Figure 3-1.

Research Design

This study used a 2 (production claims: present and not present) x 3 (claim frame:

nonloss, gain, and neutral) between-subjects incomplete factorial design. This design

was chosen to determine (1) the effects of gain-framed claims and nonloss-framed

labeling claims regarding animal welfare and environmental impact on attitude toward

the product, and (2) the effects of production claims on attitudes toward products

without production labeling claims and voting intention. Factorial designs allow the

determination of the effect of two manipulated independent variables on the dependent

variables and the interaction among the variables.

The design of the study is depicted in Table 3-1 and was implemented as follows:

R= random assignment, X= treatment (independent variable), 0= dependent variable









XA1= Exposure to production claims and product without production claims

XA2= No exposure to production claims (control)

XB1= Environmental and animal welfare nonloss-framed claims

XB2= Environmental and animal welfare gain-framed claims

XB3= Neutral-framed general product claims (control)

01= posttest measure of attitude toward product with claims

02= posttest measure of attitude toward product without claims

03= posttest measure of voting on animal welfare ballot initiative

An incomplete factorial design is used when some combinations of values of factors are

non-sensical or not of theoretical interest (Shadish, Cook, & Campbell, 2002). The cells

of XA2B1 and XA2B2 are considered non-sensical because when the production claims are

not present, they clearly cannot also have a frame. XA2 and XB3 serve as the control

levels for each factor. The XA1B3 cell is of no theoretical interest.

Controlling Threats to Internal and External Validity

The research design accounted for a number of threats to internal and external

validity. The threat of selection to internal validity was controlled by using random

assignment to conditions using a random number generator, in addition to measuring

some antecedent and intervening variables that could be controlled for statistically if

necessary. Attrition/mortality was not a major concern in this design, however, extensive

protesting helped determine the ease and length of time it takes to complete the

experiment aided in preventing attrition and fatigue. The threat of instrumentation (the

instrument changing from person-to-person) was a concern in this study given the

reliance on technology to administer the treatment and collect data. The online survey

tool was extensively pretested on various computers, Internet browsers, and operating









systems to protect against this threat. Furthermore, subjects were asked if they could

view the images depicting the treatment and automatically skipped the dependent

variable measurements if they were not able to view the photos. Finally, other

extraneous variables related to the treatment that would potentially interfere with the

internal validity were controlled across conditions, including the chicken product, label

design, brand, and price. The labels were designed by the researcher, printed on label

paper, and placed directly on a package of boneless, skinless chicken breasts. The

labels were swapped for the different treatment groups on the same package of

chicken, which was photographed in a controlled studio environment by a professional

photographer to ensure reliability between the treatment groups.

Construct validity threats were controlled through protesting, pilot testing, and

manipulation checks in the experiment and also by ensuring the constructs were well-

defined and measured using multiple questions (which controls for non-measure bias).

Mono-operation bias was controlled by using two different treatments (gain and nonloss

frame) and a control group that did not receive the treatment. Suspicion/hypothesis

guessing was controlled for by not telling participants that they are participating in an

experiment. Instead, they were told upfront that it is a survey. This also controlled for

compensatory rivalry, since they will not know there may be different surveys (treatment

conditions). Administering the experiment online and by not using leading language or

leading questions also controlled for the experimental expectancies threat. The

interactions of other treatments on the outcomes were controlled for by measuring some

of those potential interactions (previous purchasing behavior, label attention) and using

random assignment to conditions.









Subjects

The convenience sample for this study included students from four courses at a

large southeastern university: a research and business writing class (N= 192), a public

speaking class (N= 176), an introduction to journalism class (N= 138), and an

introduction to mass media class (N= 234). Subjects were offered course extra credit to

incentivize participation. When the online experiment was sent out, 740 students were

enrolled in these classes. The courses contained students from a variety of colleges

and majors and at varying phases in their program (freshman, sophomore, juniors, and

seniors). Students enrolled in more than one of these courses were accounted for and

only allowed to participate in the study once; however, students enrolled in more than

one of the classes used in the sample were given extra credit in all of them by taking the

questionnaire once.

Convenience sampling is often used in psychology research, usually with easily

accessible college students (Peterson, 2001). This method involves choosing a sample

based on what is convenient to access. Cognitive psychologists argue that when

examining cognitive mechanisms, like memory, attention, or biases, college students

are an acceptable sample because they will maintain the same neural networks. Making

generalizations about consumer behavior from college students may be more difficult.

Peterson (2001) reviewed experimental studies using different subject samples and

found differences in the direction and magnitude of effect sizes between student and

non-student samples. He advised that making generalizations about consumer behavior

from college student samples to non-student samples should be done with caution.

However, when examining a theoretically interesting causal relationship (strictly theory

testing), the focus may need to be more on internal validity than external, and,









therefore, using a college student convenience sample is appropriate (Kam, Wilking, &

Zechmeister, 2007).

The nature of the study is to examine cognitive mechanisms (framing effects, loss

aversion, and regulatory focus) that have shown prevalence in multiple nonstudent

samples (Druckman, 2001) as well as student samples (Liberman et al., 2005; Tversky

& Kahneman, 1981). The theoretical contribution being whether exposure to production

claims affects attitudes toward conventional products without such claims and how the

frame affects attitudes toward the production attribute product. While generalizations

cannot be made to all consumers from this convenience sample, providing data about

the sample characteristics can aid in external validity conclusions because readers can

determine how similar other populations of consideration may be to the sample used in

this study. It is unlikely this group will have much previous exposure to or knowledge

about these kinds of labels; therefore, perhaps this sample is ideal to make a theoretical

contribution.

In the case of marketing value-added meat products, another key issue is

identifying users or potential users for the product category. Young adults, and

specifically college students, are one segment of consumers for food products.

Currently, there are over 15.9 million college students in the United States, representing

a $9.2 billion market that is viewed by packaged goods marketers as "a meaningful

segment" on its own, with distinct characteristics, brand loyalties, and preferences for

consumable goods, including food (Ness, Gorton, & Kuznesof, 2002, p. 506). As a

segment, traditional 18- to 24-year-old college students have been shown to differ from

their similar aged nonstudent peers, in that they are much more likely to live away from









home, and thus are able to establish an independent lifestyle, including the need to

develop life skills such as food shopping and meal preparation (Mintel, 1999). Students

may even spend more on food as a percentage of their total living expenses compared

with other consumers (Ness, Gorton, & Kuznesof, 2002). They are also more likely to be

aware of diet and health issues as compared with the population as a whole (Ness,

Gorton, & Kuznesof, 2002), which makes them a relevant target for marketing new food

products and technologies. Also, research shows that young consumers (18-32), and

those with a college education are more likely to purchase organic food products

(Onyango, Hallman, & Bellows, 2007).

Most researchers who use college student samples do so because of cost and

convenience factors, and the practice must therefore be viewed as a limitation of the

study. In the present study, however, college students were also used because they

represent a group of consumer prospects whose attitudes have long been tracked by

industry for their ability to influence and predict mainstream consumer trends, and this

predictive value is particularly significant for attitudes toward meat products.

Independent Variables

Regulatory Focus

Regulatory focus theory suggests that a cognitive mechanism regulates how

individuals attend to loss/non-loss and gains/non-gains (Higgins, 1998); therefore,

subjects' regulatory focus was measured to control for these effects. The regulatory

focus questionnaire (RFQ) (Higgins et al., 2001) contains 11 items (see Table 3-2) with

two subscales. Subjects were given the RFQ before the treatment was administered to

prevent regulatory focus priming from the framed production claims.









Pretesting of Message Stimuli

To determine the labeling claims that would be used as the treatment, a three-step

process was used. First, the researcher and an assistant visited six grocery stores

(three regional chain supermarkets, one national superstore, one local grocer, and one

natural and organic foods retailer chain) and recorded all unique meat labeling claims

addressing health, animal welfare, and environmental impacts. When the lists were

collapsed, 33 unique claims resulted. This first step was taken to improve the study's

external validity by using real labeling claims. The second step involved protesting these

33 claims with three focus groups (two face-to-face and one online) with a total of 20

college student participants (7 in the first, 8 in the second, 5 in the third). The claims

were assessed for basic understanding, claim type (health, animal welfare, or

environmental), and perceived frame (gain or nonloss). Participants used nominal group

assessment to categorize the claims first as individuals and then discussed

discrepancies and claim clarity as a group to resolve differences.

Because no clear animal welfare claim with a nonloss frame and no clear

equivalently framed health claims emerged from the nominal group assessments, a

third step was needed. The third step consisted of protesting the labeling claims using

an online survey with different samples of college students. Ultimately, two online

surveys were conducted: the first tested the original 33 claims with the addition of the

claim "No cages," and the second tested the original 33 claims, "No cages," and "No

fat." The first online survey was done with 23 college students, who were not part of the

original sample for the nominal groups. The second one also had 23 college students,

who were not a part of either of the samples already used. Between the nominal groups









and two online surveys, 66 college students participated in the protesting of the labeling

claims to determine type and frame. The results are presented in Table 3-3.

The claims were chosen based on 1) a Chi-square analysis of the combined data,

and 2) whether they were equivalent frames. Unfortunately, no clear equivalently

framed (nonloss and gain) health claims were found, and subsequently, the health claim

was eliminated from the study. Most studies examining nonloss- versus gain-message

framing used quantitative descriptors (Boettcher, 2004; Idson et al., 2004; Liberman et

al., 2005; Kahneman & Tversky, 1979; McDermott, 2004; Tversky & Kahneman, 1981),

but this study used qualitative descriptors to improve external validity, meet the applied

research objectives, and further test the limits of loss aversion. The environmental gain-

framed claim chosen was "Good for the environment," and the nonloss-framed claim

chosen was "No negative environmental impacts." These two claims are qualitatively

equivalent in that a product produced in a way that does not have negative

environmental impacts is good for the environment. In the same line of logic, a product

produced in a way that is good for the environment does not have negative

environmental impacts. The animal welfare gain-framed claim chosen was "No cages,"

and the nonloss-framed claim chosen was "Free to roam." These two claims are

qualitatively equivalent in that animals raised in a production system with no cages

would be free to roam, and animals free to roam are not in cages.

The claims were printed on a label, placed on a package of boneless, skinless

chicken breasts, and photographed. Chicken was chosen to ensure reliability of the

study because it is a uniform product with little to no differences of product

characteristics that are able to be visually detected. In addition, chicken is a product









consumers choose primarily based on color with no consideration for marbling or other

visual quality cues (Becker, Benner, & Glitsch, 2000). Chicken ranks number one in

total meat consumption in the United States (USDA Economic Research Service, 2007).

Demerritt (2004) found that organic poultry is a gateway organic food and an important

frontline product for the organic industry (as cited in Oberholtzer, Greene, & Lopez,

2005). Because of the standardization of this product, chicken is ideal for experimental

purposes to ensure participants are making their decision based on the claim and not

on physical quality characteristics. Results from this study are transferable to other meat

products with more distinguishing characteristics, like beef or lamb. The same package

of chicken was used for both treatment conditions to control for any quality differences.

Price, cut, weight, and brand were also held consistent between the conditions.

The ballot initiative regarding animal welfare used the same language from

California's Proposition 2 that passed in November 2008. The same language has been

consistently used by HSUS in states like Illinois, Ohio, Michigan, Arizona, and Florida as

it either attempts or successfully proposes to ban certain animal confinement practices

(HSUS, 2009). It read:

Calves raised for veal, egg-laying hens, and pregnant pigs can be confined only in
ways that allow these animals to lie down, stand up, fully extend their limbs, and
turn around freely. Under the measure, any person who violates this law would be
guilty of a misdemeanor, punishable by a fine of up to $1,000 and/or imprisonment
in county jail for up to six months. (Prop 2: Standards for confining farm animals,
2008)

Dependent Variables

Attitudinal Measures

After viewing the product with claims and product without claims simultaneously,

subjects' attitudes toward each product were measured. The scale developed by Batra









and Ahtola (1991) measures the hedonic and utilitarian sources of consumer attitudes

using eight semantic-differential questions. The scale reliabilities exceeded a

Chonbach's alpha of .89 (Crowley, Spangenberg, & Hughes, 1992). The utilitarian

components were measured using Batra and Ahtola's (1991) scale by the five-point

semantic differential items of useful/useless, valuable/worthless, beneficial/harmful, and

wise/foolish. The hedonic component was measured by the items pleasant/unpleasant,

nice/awful, agreeable/disagreeable, and happy/sad. Overall attitudes were measured by

using items of good/bad, positive/negative, like/dislike, and favorable/unfavorable.

Again, all of these items were measured on a five-point semantic differential scale (see

Table 3-4).

Voting Behavior

After completing the attitudinal measures, subjects' voting intention was assessed

to test H4. To measure voting intention, a ballot was presented with the following

proposition:

On the next ballot in your state, the following initiative regarding the
confinement of livestock is being proposed:

Calves raised for veal, egg-laying hens, and pregnant pigs can be confined only in
ways that allow these animals to lie down, stand up, fully extend their limbs, and
turn around freely. Under the measure, any person who violates this law would be
guilty of a misdemeanor, punishable by a fine of up to $1,000 and/or imprisonment
in county jail for up to six months.

How do you plan to vote?
O Yes
o No

Attribute Variables

To improve the generalizability of the findings, several attribute variables were

included in the measures. Providing data about the sample characteristics can aid in









external validity conclusions because readers can determine how similar other

populations of consideration may be to the sample used in this study (Ary, Jacobs, &

Razavieh, 2002). These variables included chronic regulatory focus, age, gender,

organic food purchasing behavior, attention to meat labeling claims, and personal/family

ties to agriculture.

Instrumentation

The instrumentation for this study was implemented using an online questionnaire

tool. Experiments administered online offer several advantages including higher

statistical power from larger sample sizes, savings in time, resources, space, and

manpower, reduction of experimenter bias, and ease of access for the subjects (Reips,

2000). This study followed recommendations from Reips (2000) for conducting

experiments online, which included extensive protesting, using the subjects' first name

in contacts with them, and sending several reminders. The online questionnaire tool

provided the ability to collect the data completely electronically using photos to

represent the treatment and e-mail to administer it. Selection bias was not a concern in

this study because most college students are comfortable using the Web and tend to be

early adopters of new Internet technologies (Jones, Johnson-Yale, Millermaier, &

Seoane-Perez, 2009).

Instrument Content

Three different online questionnaires were created, one for each treatment group.

The only differences between all three questionnaires were the images of chicken

packages with different labeling claims. One of the manipulation check questions was

different between the two treatment groups and the control. The treatment groups were

asked which product was better for animal welfare and the environment, while the









control group was asked which product offered more information. Subjects responded to

65 questions in total. The instrument (Appendix C) was developed as a series of pages

to minimize scrolling and was comprised of the following elements:

Page 1 included the informed consent.

Page 2 included the prompt to type in their unique participant identification number,
which was used to avoid collecting names in the data set but still be able to
provide the students the extra credit incentive.

Page 3 included a checkbox question for the subject to indicate what courses they were
enrolled in. This was to ensure that students enrolled in more than one of the
courses included in the sample received the extra credit and only took the survey
once.

Page 4 was a transition page briefly explaining the first set of questions that would be
asked.

Page 5 included seven items from the RFQ that had the same response items.

Page 6 included the last two items from the RFQ that had the same response items.

Page 7 was a transition page briefly explaining the second set of questions that would
be asked.

Page 8 displayed the two photos of the packages of chicken and asked whether they
could view the images. This was done to ensure the subject's Internet browser
was displaying the images. If they answered "no," the survey tool automatically
skipped the treatment pages and questions and took them to the demographics
questions to prevent subjects from answering the questions without the ability to
view the treatment.

Page 9 displayed the two photos again, this time with the 16 semantic-differential
questions that make up the attitudinal index to measure subjects' attitudes toward
the product with the claims. The question was "I feel that Product A is..."

Page 10 displayed the two photos again, with the attitudinal index to measure subjects'
attitudes toward the product without the claims. The question was "I feel that
Product B is..."

Page 11 was a transition page briefly explaining the next screen would contain a
potential state law regarding the confinement of livestock to vote on.

Page 12 stated "On the next ballot in your state, the following initiative regarding the
confinement of livestock is being proposed." Following that text, was the language









from the California Proposition 2 that appeared on the state ballot in 2008. They
were then asked, "How do you plan to vote?"

Page 13 was a transition page explaining the next set of questions would be about
demographics and grocery shopping decisions.

Page 14 included four demographic questions to assess age, gender, community of
origin, and connections with livestock production.

Page 15 asked "Approximately how often do you eat meat (including all meals and
snacks) in a typical week?"

Page 16 included two questions to assess political party affiliation and views.

Page 17 asked whether they currently shopped for groceries for themselves or
household. If the subject indicated "no," it sent them to page 18. If they indicated
"yes," it sent them to page 19.

Page 18 was only seen if the subject indicated they did not do the grocery shopping.
This page asked if they helped make the decisions about the food that would be
bought. If they indicated "yes," it sent them to page 19. If they indicated "no," it
skipped them to page 21.

Page 19 included five yes/no questions regarding their attention to organic and
production labeling on poultry or meat products.

Page 20 included five questions to assess how often they purchased organic and
credence-labeled products.

Page 21 was a transition page explaining the final set of questions would be related to
their experience taking the survey.

Page 22 included one manipulation check question to see if they noticed the difference
in the labeling between the two products.

Page 23 included one manipulation check question to see if they noticed one was
intended to be labeled as better for the environment and animal welfare.

Page 24 was the thank you and debriefing page explaining the purpose of the study and
that the voting situation was only hypothetical.

Pilot Test

The procedure was pilot tested with 30 undergraduate students. After the pilot

subjects completed the experimental procedure, they were asked a series of qualitative

questions to determine fatigue, technical, and understanding issues. Using SPSS 16.0









for WindowsTM, item analysis statistics were run to determine the construct validity of

each of the scales measuring concepts of interest. In the social sciences, a reliability

coefficient of .70 or larger indicates an index is adequate (Traub, 1994). This data

analysis indicated one necessary change to the instrument before final distribution.

The 16-item attitudinal scale had an alpha reliability of .97 overall, with a reliability

of .97 on the 12-item index developed by Batra and Ahtola (1991) subscale and .88 on

the 4-item researcher developed subscale. DeVellis (2003) states that, ideally, the

Chronbach alpha coefficient of a scale should be above .7; therefore, all items were

kept in this scale. The 11-item regulatory focus questionnaire contains two subscales: a

5-item prevention scale and a 6-item promotion scale. The prevention scale had an

alpha reliability of .77, while the promotion scale had an alpha reliability of .67.

Removing the item "I have found very few hobbies or activities in my life that capture my

interest or motivate me to put effort in them" was deleted, resulting in a final reliability of

.72 for the promotion index.

Procedure

Subjects (N= 740) were randomly assigned (with the use of a random number

generator) to either the nonloss-framed claims condition, the gain-framed claims

condition, or the control claims condition to test the hypotheses. The claims and

treatment conditions are shown in Table 3-5. In the gain frame and nonloss frame

conditions, subjects simultaneously viewed a package of chicken with two production

claims (animal welfare and environmental impact), cut, weight, and price on the label

and a package of chicken with only cut, weight, and price on the label (referred to

hereafter as the product without production claims). In the control condition, subjects









simultaneously viewed a product without production claims and a product with general

product claims (boneless and skinless, and chicken breasts).

Subjects were verbally given a pre-notice from their course instructor two to three

days before the initial contact e-mail with directions and the link were sent out (see

Appendix A). The first contact was sent to students as a personalized e-mail (Dear [First

name]) to better establish a connection with the subjects (Reips, 2000) with a brief

explanation of the survey, their unique participant identification number, and a unique

link matched with their randomly assigned treatment group (see Appendix B). Subjects

had 10 days to complete the questionnaire. Three reminder e-mails were sent to

subjects who had not yet responded: the first was sent four days after the first contact,

the second three days after that, and the third was sent the morning of the final day (see

Appendices B-1 and B-2).

Data Analysis

Data analysis for this study was completed using SPSS 16.0 for WindowsTM PC.

Cronbach's coefficient alpha was used as an internal-consistency measure of reliability.

This measure is used with Likert type questions when a score can take on a range of

values (Ary et al., 2002). One-way ANOVAs and t-tests were used to address the first

three hypotheses and a Chi square test for independence was used to address the

fourth hypothesis.











Table 3-1. Incomplete factorial research design
R XA1B1 01, 02, 03

R XA1B2 01, 02, 03

R XA2B3 01, 02, 03


Table 3-2. Regulatory focus questionnaire (Higgins et al., 2001)
Never Seldom Sometimes Often Very often

Compared to most people, are 1 2 3 4 5
you typically unable to get what
you want out of life?

Growing up, would you ever 1 2 3 4 5
"cross the line" by doing things
your parents would not
tolerate?
How often have you 1 2 3 4 5
accomplished things that got
you psychedd" to work even
harder?
Did you get on your parents' 1 2 3 4 5
nerves often when you were
growing up?
How often did you obey rules 1 2 3 4 5
and regulations that were
established by your parents?
Growing up, did you ever act in 1 2 3 4 5
ways that your parents thought
were objectionable?
Do you often do well at different 1 2 3 4 5
things that you try?
Not being careful enough has 1 2 3 4 5
gotten me into trouble at times.
Certainly Somewhat Neither true Somewhat Certainly
false false nor false true true
When it comes to achieving 1 2 3 4 5
things that are important to me,
I find that I don't perform as well
as I ideally would like to.
I feel like I have made progress 1 2 3 4 5
toward being successful in my
life.
I have found very few hobbies 1 2 3 4 5
or activities that capture my
interest or motivate me to put
effort into them.










Table 3-3. Results of nominal group assessment
Claim Claim Type Frame

Health Environment Animal Unclear Nonloss Gain Unclear


Earth friendly

Raised without antibiotics

Great care

Family farmers in
harmony with nature
Good for you

Minimally processed

Lean

Eco-friendly

Fed vegetarian diet

No hormones
administered
No artificial ingredients

Good for the environment

No animal or poultry
products in feed
No cows injected w/ rgbh

No environmental
contamination
No negative
environmental impacts
Raised to reduce
environmental impacts
No antibiotics

No antibiotics
administered
Produced without
antibiotics, synthetic
hormones, or pesticides
No hormones

Raised without hormones

Cage-free

No growth stimulants


46* 19

44 19


0 48* 19


3 9


15 51*

52* 14

52* 13

32 22










Table 3-3. Continued
Claim Claim Type Frame

Health Environment Animal Unclear Nonloss Gain Unclear
No preservatives 62* 1 1 2 44 22 0

No fillers 53* 0 9 4 44 22 0

No steroids 35 0 19 12 50* 14 2

Meets humane society 5 1 60* 0 26 39 0
standards
No chemical medicines 27 1 23 15 39 12 0

Humanely raised 0 0 66* 0 16 50* 0

No nitrates 41 17 2 6 49* 17 0

No sodium 65* 0 1 0 40 25 1

Free to roam 0 0 66* 0 16 50* 0

No cages (n= 46) 1 1 44* 0 47* 22 0

No fat (n= 23) 23* 0 0 0 6 17* 0

Note: Labels in italics were added later and only tested with online survey. Chi-square
analysis p < .05.

Table 3-4. Hedonic/utilitarian sources of attitude scale (Batra & Ahtola, 1991) and
researcher-developed measures
I feel that the product is...


Useless

Worthless

Harmful

Foolish

Unpleasant

Awful

Disagreeable

Sad

Bad


Useful

Valuable

Beneficial

Wise

Pleasant

Nice

Agreeable

Happy

Good

Positive

Like


Negative

Dislike









Table 3-4. Continued


I feel that the product is...
Unfavorable 1 2 3 4 5 Favorable

Unhealthy 1 2 3 4 5 Healthy

Unsafe to eat 1 2 3 4 5 Safe to eat when
when cooked cooked
From an animal 1 2 3 4 5 From an animal
treated treated humanely
inhumanely
Bad for the 1 2 3 4 5 Good for the
environment environment
Note: Italicized text indicates researcher-developed item.

Table 3-5. Experiment treatment groups
Claim Type Nonloss Frame Gain Frame Condition Control Condition
Condition
Environmental No negative Good for the X
Impact environmental impacts environment
Animal Welfare No cages Free to roam

Control X X Boneless and skinless

Chicken breasts











Covariate
* Chronic Regulatory Focus
(prevention or promotion)




Message Frame/Treatment
* Nonloss/No claims
* Gain/No claims
* Neutral/No claims
(control)


\
J \
\I


Framing Effect
* Attitude toward
product with
credence claims

(More Positive)


- -

Framing Effect
Attitude toward
product with
claims


(Less Positive)


Exposure to
production claim
effect
* Attitude toward
product w/o
production claims

(More Positive)



Exposure to
production labeling
claim effect
* Attitude toward
product without
production claims

(Less Positive)


Figure 3-1. Operational framework for the current study


Exposure to
production claim
effect
* Voting behavior
on animal welfare
ballot initiative

(Support)



Exposure to
production claim
effect
* Voting behavior
on animal welfare
ballot initiative

(Does not
support)









CHAPTER 4
RESULTS

With loss aversion and regulatory focus theory as the theoretical framework, the

objectives of this study were to determine the effects of differently framed labeling

claims on consumers' attitudes toward the product with production claims, the

conventional product, and voting intention. The treatment and assessments were

delivered online through the use of an online questionnaire tool to college students in

four different large university courses. The three independent variables were the

presence/absence of production labeling claims, claim frame (nonloss or gain), and

regulatory focus (promotion, prevention). The two dependent variables were attitudes

toward the product and voting intention on an animal welfare ballot initiative.

This chapter provides an analysis of the data beginning with sample

demographics, followed by analysis of the variables of interest. Next is a discussion of

the scale reliabilities used to develop the indexes that measure the independent and

dependent variables, followed by an overview of manipulation checks. The chapter

concludes with a discussion of the tests of hypotheses used in the study.

Descriptive Analysis

Using an online survey tool for development and administration, the 65-item

questionnaire was administered to 740 college students from four different large

university courses. The overall response rate was 89.2% (n= 660).

Demographics

The demographic characteristics included in the instrument were: age, gender,

political party affiliation, political views, rural-urban background, connection with the

livestock industry, meat consumption frequency, and attention to and purchase









frequency of meat and/or poultry with five types of production labeling claims. Subjects

who did not grocery shop for themselves or their household, nor help make the food

purchasing decisions skipped the questions about attention to and purchase frequency

of meat and/or poultry products with production labeling claims.

Descriptive analysis indicated 459 of subjects were female (69.5%) and 201 were

male (30.5%). The undergraduate student population from which the sample was

chosen contains more females (55%) than males (45%) (University of Florida Office of

Institutional Planning and Research, 2009). The age range of the respondents was 18

to 33 years old, with a mean of 21 years old (SD= 1.69). The majority of subjects

described the community in which they grew up in as a subdivision in a city or town (n=

491, 74.4%), followed by rural, not a farm (n= 98, 14.8%), downtown in a city or town

(n= 47, 7.1%), and farm (n= 23, 3.5%). Most subjects indicated that neither they nor

their immediate family work in livestock production (n= 563, 85.3%).

Subjects' political party affiliation was fairly evenly distributed among the response

items with most identifying themselves as Independent-leaning Democrat (n= 105,

15.9%), followed by Democrat (n= 97, 14.7 %), and Republican (n= 85, 12.9%). When

the response items were collapsed, 265 were Democrat (40.2%), 228 were Republican

(34.5%), and 78 were Independent (11.8%). Subjects' political views leaned slightly

more toward Liberal than Conservative. Most indicated they were somewhat Liberal (n=

116, 17.6%), followed closely by Liberal (n= 115, 17.4%), while 90 considered

themselves somewhat Conservative (13.6%) and 98 Conservative (14.8%). When the

response items were collapsed, 277 were Liberal (42.0%), 222 were Conservative

(33.6%), and 107 were neither (16.2%).









The majority of subjects consumed meat on a regular basis with most eating it 4-7

times per week (n= 258, 39.1%) and 8-14 times per week (n= 216, 32.7%). Only 27

(4.1%) indicated that they never eat meat, and 14 (2.1%) indicated they eat it less than

once per week (see Figure 4-1).

Attention to and Purchase Frequency of Meat/Poultry with Production Claims

Before assessing attention to and purchase frequency of meat and poultry

products with production labeling claims, subjects were asked if they do the grocery

shopping or help make the decisions for food purchases. The majority of subjects (n=

601, 91.1%) do the grocery shopping for themselves or their household and 41 (6.2%)

help make the decisions as to what food to purchase. Only 17 (2.6%) indicated they do

not purchase nor help make the decisions; therefore, they automatically skipped over

the production label attention and purchase frequency questions.

Subjects were asked whether they pay attention to five different types of

production labeling claims: 1) organic labels, 2) labels that address the way the animal

was raised, 3) labels that say "no hormones," 4) labels that say "no antibiotics," and 5)

labels that suggest the product is better for the environment ("green"). The sample was

fairly evenly split (with the exception of claims addressing the way the animal was

raised) between "yes" and "no," with slightly more indicating "no" on all five of the

labeling claim types. Table 4-1 displays the results in entirety.

When asked how often they purchase meat or poultry products with these five

production labeling claims, most indicated they purchase them never or less than once

a month. The means were all less than 2, with the way the animal was raised having the

lowest purchase frequency (M= 1.13, SD= 1.44) and no hormones having the highest

purchase frequency (M= 1.55, SD= 1.71). See Table 4-2 for the complete results.









Scale Reliabilities

Within the loss aversion and regulatory focus framework, three scales were used

in this study to measure independent and dependent variables. The regulatory focus

questionnaire contained two scales: one to assess promotion focus and the other to

assess prevention focus. This was an independent variable in the study. Attitude, a

dependent variable, was measured using a scale containing 16 items, which consisted

of two subscales.

Regulatory Focus Scales

Regulatory focus was measured using a questionnaire developed by Higgins et al.

(2001). The total questionnaire contained 11-items, five prevention items and six

promotion items. They are considered two separate scales. In the pilot study, one of the

promotion items was eliminated to improve the reliability alpha, thereby making the

scale consist of five items. Subjects indicated their responses using a 5-point Likert

scale. Each scales' total score could be 25 at the strongest regulatory focus down to

five at the weakest regulatory focus.

The promotion scale had a range of standard deviations from .82 to .93,

demonstrating a minimal amount of variance in the data. The corrected item-total

correlations on the prevention scale ranged from .43 to .67 (Table 4-3). The alpha

reliability coefficient for the entire index was a= .79 and would not be improved by

removing any item. DeVellis (2003) states that, ideally, the Chronbach alpha coefficient

of a scale should be above .7. The grand mean for the promotion scale was 17.88 (SD=

3.22).

The prevention scale had a range of standard deviations from .70 to 1.12

demonstrating some variance in the data. The correction item-total correlations on the









prevention scale ranged from .43 to .67 (Table 4-4). The alpha reliability coefficient for

the entire index was a= .68 and could not be improved by removing any item. This is

slightly lower than what DeVellis (2003) recommends, and is therefore, a limitation of

the study. The grand mean for the prevention scale was 19.48 (SD= 2.73).

Attitude Scales

Attitude was treated as one dependent variable that included product-specific

attitude and general attitude because the overall reliability of the scale was strong. The

total attitude scale with all 16 items had standard deviations ranging from .94 to 1.15 on

attitude toward the product without claims and .85 to 1.09 on attitude toward the product

with the claims. The alpha reliability coefficient for the entire index was a= .96 on

attitude toward product without the claims and a= .96 on attitude toward product with

the claims. Neither would be significantly improved by removing any item (Table 4-5).



Descriptive Analysis of Variables of Interest

Regulatory Focus

Subjects' regulatory focus was measured using Higgins et al. (2001) Regulatory

Focus Questionnaire. One item (question 11) from this index was deleted based on the

scale reliabilities in the pilot testing data. This questionnaire contains two scales: one to

assess promotion focus and the other to assess prevention focus. The lowest score on

each scale was seven and the highest was 25. The mean promotion score was 19.48

(SD= 2.73) and the mean prevention score was 17.88 (SD= 3.22), indicating that this

sample tended to be more promotion focused. Table 4-6 shows the results for each

item.









Attitude Toward Product

The attitude score ranged from 1 (most negative) to 3 (neutral) to 5 (most

positive). The grand mean on attitude toward the products without the claims was 3.53

(SD= .84). The grand mean attitude toward the products with the claims was higher (M=

4.04, SD= .74). Overall, attitude toward the product with the claims was more positive

than attitude toward the product without the claims. Table 4-7 displays the results for

each item, and Table 4-8 displays the results for attitude between treatment groups.

Voting Intention

Voting intention was measured using a one-item measure. Subjects were asked

how they would vote on an animal confinement law in their state. The language of the

proposed initiative was the same that appeared on California's 2008 ballot for

Proposition 2. The question was posed as follows:

On the next ballot in your state, the following initiative regarding the
confinement of livestock is being proposed:

Calves raised for veal, egg-laying hens, and pregnant pigs can be confined only in
ways that allow these animals to lie down, stand up, fully extend their limbs, and
turn around freely. Under the measure, any person who violates this law would be
guilty of a misdemeanor, punishable by a fine of up to $1,000 and/or imprisonment
in county jail for up to six months.

How do you plan to vote?
O Yes
o No

Subjects were not provided any additional information. Most subjects indicated they

plan to vote "yes" for this law (n= 510, 77.3%), while only 150 (22.7%) indicated "no."

Manipulation Checks

To evaluate the labeling claim stimuli used in the treatment, two manipulation

checks were conducted. The labels for the two products in each condition were









designed identically, with the exception of the presence of the claims. The first

manipulation check was designed to determine if subjects noticed the difference

between the labels on the packages of chicken. In the gain-frame condition, 97.5% (n=

196) noticed the differences between the two labels. In the nonloss-frame condition,

98.3% (n= 227) noticed the differences. In the control condition, 93.4% (n= 197) noticed

the differences. A one-way between-groups analysis of variance with the Welch

correction showed that the differences between the groups was not significant F (2,

660) = 2.17, p = .12.

The second manipulation check was designed to determine if subjects recognized

that the product with the claims was better for animal welfare and the environment. In

the control condition, subjects were instead asked which product contained more

information. In the nonloss-frame condition, 98.3% (n= 232) identified the product with

claims as better for animal welfare and the environment. In the gain-condition, 96.6%

(n= 201) identified the product with claims as better for animal welfare and the

environment. In the control condition, a different question was asked since general

product claims were used. In this condition, 96.3% (n= 208) identified the product with

the claims as the one containing more information. A one-way between-groups analysis

of variance with the Welch correction showed that the differences between the groups

was not significant F (2, 640) = 1.13, p = .33.

Tests of Hypotheses

To determine if the theoretical covariate was influencing the dependent variables,

the relationships between the regulatory focus scores (promotion, prevention) and

attitude toward the products were investigated using Pearson product-moment

correlation. Preliminary analyses were performed to ensure no violation of the









assumptions of normality, linearity, and homoscedasticity. There was a small positive

correlation between promotion score and attitude, r= .08, n = 660, p = .04, with a

greater promotion focus associated with a more positive attitude. There were no

significant correlations between prevention score and attitude. Because the promotion

score only helps explain .64% of the variance in subjects' scores on the attitude scale,

the promotion focus effect was considered negligible. A correlation coefficient less than

.09 means there is no relationship between the variables (Cohen, 1988). Subsequently,

both covariates (promotion score, prevention score) were removed from the data

analysis of the hypotheses.

H1: When controlling for regulatory focus, subjects exposed to nonloss-framed
claims will have more positive attitudes toward the product with production claims
than those exposed to gain-framed labeling claims or neutral general product
claims.

The covariates for regulatory focus were not included in the analysis because

Pearson product-moment correlations revealed no relationship between the covariates

and the dependent variables. A one-way between-groups analysis of variance was

conducted to compare the different labeling claim framing effects on attitudes toward

the product with the claims. The independent variable was the frame of the claim

(nonloss, gain, neutral), and the dependent variable was attitude toward the product

with the claims. Preliminary assumption testing showed no serious violations noted.

There was a significant effect of labeling claim frame on attitudes toward the

product with production claims, F (2, 657) = 16.87, p < .001 (see Table 4-9).

Planned contrasts revealed that subjects exposed to gain-framed claims had more

positive attitudes toward the product with the claims than those exposed to neutral

product claims t(657) = -5.26, p < .001, and those exposed to nonloss-framed claims









had more positive attitudes in comparison to the control group as well t(657) = -4.79, p <

.001. The difference between gain and nonloss labeling claim frames, however, was not

significant t(657) = -.64, p = .52 (2-tailed) (see Table 4-10 and Figure 4-2).

H2: When controlling for regulatory focus, subjects exposed to nonloss-framed
claims will have less positive attitudes toward the product without production
claims than those exposed to gain-framed labeling claims or neutral general
product claims.

The theoretical covariate of regulatory focus was not included in this analysis

either. A one-way between-groups analysis of variance was conducted to compare the

different labeling claim framing effects on attitudes toward the product without the

claims. The independent variable was the frame of the claim (nonloss, gain, neutral),

and the dependent variable was attitude toward the product without the claims.

Preliminary assumption testing was conducted with no serious violations noted.

There was a significant effect of labeling claim frame on attitudes toward the

product without production claims, F (2, 657) = 6.41, p = .002 (see Table 4-11).

Planned contrasts revealed that subjects exposed to gain-framed claims had less

positive attitudes toward the product without the claims than those exposed to neutral

product claims t(657) = 2.12, p = .035, and those exposed to nonloss-framed claims had

less positive attitudes in comparison to the control group as well t(657) = 3.56, p < .001.

The difference between gain and nonloss labeling claim frames, however, was not

significant t(657) = -.1.37, p = .17 (2-tailed) (see Table 4-12 and Figure 4-3).

H3: Subjects exposed simultaneously to a food product with production claims and a
product without such claims will have less positive attitudes toward the product without
the claims than those who do not see a food product with production claims.
While the data analyses for H1 and H2 offered insight into this hypothesis, a

specific analysis was conducted to offer a complete picture. The treatment groups

(nonloss, gain, control) were recorded into a new independent variable that grouped









together subjects in the nonloss and gain conditions because they were the groups that

saw the production labeling claims, whereas the control group saw general product

claims (boneless and skinless, chicken breasts). An independent samples t-test was

conducted to compare the presence of the production claims effects on attitudes toward

the product without the claims. The independent variable was the presence of the

production claims (present, absent), and the dependent variable was attitude toward the

product without claims. Preliminary assumption testing was conducted with no serious

violations noted.

The independent samples t-test showed a significant difference between the two

groups t(658) = -.3.31, p = .001 (2-tailed). An inspection of the mean scores indicated

that subjects' exposed to the production labeling claims had less positive attitudes

toward the product without claims than those who were not (see Table 4-13 and Figure

4-4).

H4: Subjects exposed to a food product with production claims will be more likely
to have intentions to vote "yes" for an animal welfare ballot initiative than those
who do not see a food product with production claims.

A Chi-square test for independence indicated no significant association between

subjects' exposure to production labeling claims and voting decision on the animal

welfare ballot initiative. The majority of subjects voted yes (n = 510, 77.3%).

Post Hoc Analyses

Several post hoc analyses were conducted to explore other relationships among

the variables collected in this study that were not included in the hypotheses. These

analyses provide a more complete picture of other variables that influence the

dependent variables in this study.









Attention To and Purchase of Products With Production Claims

Subjects were asked 1) whether they pay attention to, and 2) how often they

purchase products with five different types of production labeling claims on meat and/or

poultry when grocery shopping: 1) organic labels, 2) labels that address the way the

animal was raised, 3) labels that say "no hormones," 4) labels that say "no antibiotics,"

and 5) labels that suggest the product is better for the environment ("green"). These

items were re-coded into a single variable for attention (Chronbach alpha coefficient of

.86) and a single variable for purchase frequency (Chronbach alpha coefficient of .89).

The relationship between these grocery shopping behaviors (attention and purchase

frequency) and attitudes toward the products was investigated using Pearson product-

moment correlation coefficient. The analyses revealed three key findings (see Tables 4-

14 and 4-15 for all results). The first is that attention to production claims had a strong,

positive correlation with purchase frequency of these products, r = .58, p < .001. The

second is that attention (r= -.28, p < .001) and purchase frequency (r= -.21, p < .001)

both had small, negative correlations with attitude toward the product without production

claims (see Table 4-15). The more production claims subjects' indicated they pay

attention to and the more frequently they purchase products with these claims, the less

positive their attitudes were toward the product without production claims.

Community Upbringing and Livestock Background

Subjects' identified themselves as growing up in a: subdivision in a city or town (n=

491, 74.4%), rural area, not a farm (n= 98, 14.8%), downtown area in a city or town (n=

47, 7.1%), and farm (n= 23, 3.5%). Table 4-16 displays the descriptive statistics for

community upbringing and attitude toward the products. Examining the attitudinal

means between the two products, it appears that those who identified themselves as









growing up on a farm or in a rural area had more positive attitudes toward both products

than those from a subdivision or urban area.

Means for attitudes toward the products were compared using one-way ANOVAs

to determine if there were any differences based on their self-identified community

upbringing. There was no significant difference between the groups on attitudes toward

the product with the claims, F(3, 654) = 1.10, p = .35. There was a significant difference

between the groups on attitudes toward the product without the claims (Table 4-17).

Despite reaching statistical significance, the actual difference in mean scores

between groups was quite small given an effect size of .02. Post hoc comparisons using

the Bonferroni correction indicated that the mean score for those who grew up in a rural

area, not a farm (M = 3.76, SD = .83) had significantly more positive attitudes than

those who grew up in a subdivision (M = 3.49, SD = .83), or an urban ("downtown") area

(M = 3.34, SD = .82). Table 4-18 displays the results of the post hoc comparisons.

An analysis of variance was conducted to determine if having a personal or family

background in livestock production affected attitudes toward the product. There was no

significant effect of planning to or having a background in livestock production on

attitudes toward the product with the claims F(3, 653) = 1.39, p = .25 or attitudes toward

the product without the claims F(3, 653) = 2.29, p = .08.

Political Affiliation and Voting Intention

As mentioned earlier in the chapter, subjects indicated their political party affiliation

and political viewpoint. With respect to political party identification, 265 indicated

themselves as Democrat (40.2%), 228 Republican (34.5%), and 78 Independent

(11.8%). For political viewpoint, 277 considered themselves Liberal (42.0%), 222

Conservative (33.6%), and 107 were neither (16.2%). A Chi-square test for









independence did not show any association between political party or viewpoint and

voting intention on the animal welfare ballot.

Other Non-Significant Relationships

Several other relationships were investigated using either Pearson product-

moment correlations, Chi-square tests for independence, and ANOVAs depending on

the type of variables. No statistically significant relationships were found between age

and regulatory focus, political affiliations and regulatory focus, gender and regulatory

focus, age and attitudes toward the products, gender and attitudes, or course and

attitudes.











Table 4-1. Attention to selected production labeling claims on meat/poultry
Yes No
Type of Labeling Claim n % n %

Organic 306 47.6 337 52.4
Way animal was raised 226 35.1 417 64.9
No hormones 319 49.6 324 50.4
No antibiotics 285 44.4 357 55.6
Better for environment 283 44.1 359 55.9

Table 4-2. Purchase frequency of meat/poultry with select production labeling claims

Type of Labeling Claim n M SD
Organic 640 1.21 1.45
Way animal was raised 641 1.13 1.44
No hormones 642 1.55 1.71
No antibiotics 638 1.42 1.66
Better for environment 642 1.21 1.42
Note: Scores based on Likert scale with 0= never, 1= less than once a month, 2= once
a month, 3= twice a month, 4= weekly, 5= every time.

Table 4-3. Promotion focus scale inter-item consistency statistics
M SD Corrected Item-Total Alpha if Item Deleted
Correlation

RF2 3.51 .92 .67 .72

RF4 3.34 .93 .58 .75
RF5 4.13 .82 .55 .76

RF6 3.53 .85 .63 .73
RF8 3.37 .84 .43 .79

Table 4-4. Prevention focus scale inter-item consistency statistics
M SD Corrected Item-Total Alpha if Item Deleted
Correlation

RF1 3.73 .72 .44 .64
RF3 3.83 .76 .40 .65

RF7 3.91 .75 .40 .65
RF9 3.58 1.12 .49 .62

RF10 4.43 .70 .53 .60


100











Table 4-5. Total attitude scale inter-item consistency statistics

Product Without Claims Product With Claims

M SD Corrected Alpha if M SD Corrected Alpha if Item
Item-Total Item Item-Total Deleted
Correlation Deleted Correlation


Useless: Useful 3.98 .98 .70 .96 4.18 .87 .71 .96

Worthless: Valuable 3.86 .97 .75 .96 4.14 .85 .75 .96

Harmful: Beneficial 3.62 1.08 .84 .96 4.17 .85 .82 .96

Foolish: Wise 3.42 .94 .79 .96 3.82 .96 .75 .96

Unpleasant: Pleasant 3.49 1.04 .84 .96 3.93 .93 .82 .96

Awful: Nice 3.49 .96 .87 .96 3.91 .91 .84 .96

Disagreeable: Agreeable 3.54 .99 .85 .96 3.90 .93 .82 .96

Sad: Happy 3.23 .99 .79 .96 3.71 .92 .78 .96

Bad: Good 3.53 1.08 .88 .96 4.01 .92 .84 .96

Negative: Positive 3.41 1.05 .87 .96 4.04 .91 .85 .96

Dislike: Like 3.57 1.13 .86 .96 4.05 .93 .84 .96

Unfavorable: Favorable 3.42 1.15 .85 .96 4.08 .95 .85 .96

Unhealthy: Healthy* 3.70 1.09 .78 .96 4.27 .84 .75 .96

Unsafe to eat when 4.25 .97 .51 .97 4.46 .85 .63 .96
cooked: Safe to eat when
cooked*

From an animal treated 2.92 1.15 .60 .97 3.95 1.09 .63 .96
inhumanely: From an
animal treated humanely*

Bad for the environment: 3.08 1.04 .66 .96 3.96 1.02 .68 .96
Good for the
environment*
Note: Scores based on semantic differential scale from 1= useless to 5= useful.
*Researcher-developed item.

Table 4-6. Regulatory focus questionnaire descriptive statistics

n M SD
Compared to most people, are you typically unable to get 660 3.73 .72
what you want out of life? (promotion item)*
Growing up, would you ever "cross the line" by doing 660 3.51 .92
things your parents would not tolerate? (prevention
item)*


101











Table 4-6. Continued


n M SD
How often have you accomplished things that got you 660 3.83 .76
psychedd" to work even harder? (promotion item)
Did you get on your parents' nerves often when you were 660 3.34 .93
growing up? (prevention item)*
How often did you obey rules and regulations that were 660 4.13 .82
established by your parents? (prevention item)
Growing up, did you ever act in ways that your parents 660 3.53 .85
thought were objectionable? (prevention item)
Do you often do well at different things that you try? 660 3.91 .75
(promotion item)
Not being careful enough has gotten me into trouble at 660 3.37 .84
times. (prevention item)*
When it comes to achieving things that are important to 660 3.58 1.12
me, I find that I don't perform as well as I ideally would
like to. (promotion item)*
I feel like I have made progress toward being successful 660 4.43 .70
in my life. (promotion item)
Note: Scores based on Likert scale of 1= never to 5= very often or 1= certainly false to
5= certainly true. *Item was reverse coded

Table 4-7. Attitude toward product (product specific* + general attitude)


Useless:Useful

Worthless:Valuable

Harmful:Beneficial

Foolish:Wise

Unpleasant:Pleasant

Awful:Nice

Disagreeable:Agreeable

Sad:Happy

Bad:Good

Negative: Positive

Dislike:Like

Unfavorable: Favorable

Unhealthy:Healthy

Unsafe to eat when cooked:
Safe to eat when cooked*


Attitude Toward Product
Without Claims

n M

660 3.98

660 3.86

660 3.62

660 3.42

660 3.49

660 3.49

660 3.54

660 3.23

660 3.53

660 3.41

660 3.57

660 3.42

660 3.70

660 4.25


Attitude Toward Product
With Claims


SD

.98

.97

1.08

.94

1.04

.96

.99

.99

1.08

1.05

1.13

1.15

1.09

.97


M

4.18

4.14

4.17

3.82

3.93

3.91

3.90

3.71

4.01

4.04

4.05

4.08

4.27

4.46


102










Table 4-7. Continued
Attitude Toward Product Attitude Toward Product
Without Claims With Claims
n M SD n M SD

From an animal treated 660 2.92 1.15 660 3.95 1.10
inhumanely: From an animal
treated humanely*
Bad for the environment: Good 660 3.08 1.04 660 3.96 1.02
for the environment*
Note: Scores based on semantic differential scale from 1= useless to 5= useful.
*Researcher-developed item to measure product-specific attitude.

Table 4-8. Attitude toward product grand means among treatment groups
Attitude Toward Product Without Claims Attitude Toward Product With Claims
Treatment Group n M SD n M SD
Gain Frame 208 3.51 0.81 208 4.17 0.67
Nonloss Frame 236 3.41 0.87 236 4.13 0.68
Control 216 3.68 0.81 216 3.80 0.81
Total 660 3.53 0.84 660 4.04 0.74
Note: Scores ranged from 1 (most negative) to 3 (neutral) to 5 (most positive).

Table 4-9. Effects of labeling claim frame on attitudes toward product with claims
Source SS df MS F p

Claim Frame 17.49 2 8.75 16.87 .000
Error 340.53 657 .518
Total 358.02 659

Table 4-10. Planned comparisons t-test for differences between treatment groups on
attitude toward product with claims

n M SD t df p

Gain Framed Production Claims 236 4.17 0.67 -5.26 657 .000

Neutral Framed Product Claims (Control) 216 3.80 0.81

Nonloss Framed Production Claims 236 4.13 0.68 -4.79 657 .000

Neutral Framed Product Claims (Control) 216 3.80 0.81

Gain Framed Production Claims 236 4.17 0.67 -.64 657 .52

Nonloss Framed Production Claims 236 4.13 0.68


103










Table 4-11. Effects of labeling claim frame on attitudes toward product with claims
Source SS df MS F p

Claim Frame 9.86 2 4.43 6.41 .002

Error 453.90 657 .691

Total 462.76 659


Table 4-12. Planned comparisons t-test for differences between treatment groups on
attitude toward product without claims

n M SD t df p

Gain Framed Production Claims 236 3.51 0.81 2.12 657 .035

Neutral Framed Product Claims (Control) 216 3.68 0.81

Nonloss Framed Production Claims 236 3.41 0.87 3.56 657 .000

Neutral Framed Product Claims (Control) 216 3.68 0.81

Gain Framed Production Claims 236 3.51 0.81 -1.37 657 .17

Nonloss Framed Production Claims 236 3.41 0.87

Scores ranged from 1 (most negative) to 3 (neutral) to 5 (most positive).

Table 4-13. Independent samples t-test for differences between subjects exposed to
production claims and subjects exposed to general product claims
n M SD t df p

Production Claims Absent 216 3.68 0.81 -3.31 658 .001

Production Claims Present 444 3.46 0.84

Note: Scores ranged from 1 (most negative) to 3 (neutral) to 5 (most positive).

Table 4-14. Pearson product moment correlations between grocery shopping behavior
and attitude toward product with production claims
Attitude toward Attention to Purchase frequency
product with production claims in of products with
production claims grocery store production claims
Attitude toward product with 1 -.08 .002
production claims
Attention to production claims in N= 629 1 .59**
grocery store
Purchase frequency of products N= 629 N= 629 1
with production claims


104










Table 4-15. Pearson product moment correlations between grocery shopping behavior
and attitude toward product without production claims
Attitude toward Attention to Purchase frequency
product without production claims in of products with
production claims grocery store production claims
Attitude toward product without 1 -.22** -.17**
production claims
Attention to production claims in 629 1 .59**
grocery store
Purchase frequency of products 629 629 1
with production claims
** p < .001 (2-tailed).

Table 4-16. Mean attitude toward products between community upbringing
Attitude Toward Product With Attitude Toward Product Without
Claims Claims
n M SD n M SD
Farm 23 4.11 .95 23 3.86 .92
Rural, Not a Farm 98 4.14 .77 98 3.76 .83
Subdivision in Town or City 491 4.02 .73 491 3.49 .83
Downtown in Town or City 47 3.95 .65 47 3.36 .82


Table 4-17. Effects of community upbringing on attitude toward product without claims
Source SS df MS F p rf

Label Presence 7.63 1 7.63 11.22 .001 .017

Community Upbringing 9.94 3 3.31 4.87 .002 .022

Error 445.04 654 .68

Total 8682.90 659


Table 4-18. Post hoc comparisons between community upbringing
95% Cl
Comparisons Mean Attitude SE p Lower Upper Bound
Difference Bound
Farm vs. Rural .095 .191 1.0 -.411 .600
Rural vs. Subdivision .274* .091 .017 .032 .516
Subdivision vs. Urban .128 .126 1.0 -.206 .461
Urban vs. Farm -.496 .210 .11 -1.052 .059
Rural vs. Urban .402* .146 .037 .014 .789
Farm vs. Subdivision .369 .176 .220 -.097 .834
Note: p-values reflect Bonferroni adjustment for multiple comparisons


105









* Never
* Less Than Once
E 1-3
S4-7
E18-14
* Overl4


Figure 4-1. Subjects' weekly meat consumption


106












4.2-


S4.1-


0
5

1i
a-





3.9-












Figure 4-2. Means between the attitudes toward product with claims in each treatment
group. Total attitude includes product-specific and general attitude.
group. Total attitude includes product-specific and general attitude.


107

















3.65-

0

S3.6-



IL 3.55-
0g


I-
i 3.5-



C 3.45-



3.4-

gain nonloss control
treatment


Figure 4-3. Means between the attitudes toward product without claims in each
treatment group.


108


















,
3.65-
0



I-
S3.6-
0







m




3.45

No Credence Claims Credence Clains Present
label present

Figure 4-4. Means between the attitudes toward product without claims in each
treatment group.


109









CHAPTER 5
CONCLUSION

Overview


This study addressed two gaps found in the literature. The first one being whether

consumers' beliefs about conventionally produced food products are affected by

production labeling claims and whether this on-package marketing can also affect intent

to support an animal welfare ballot initiative. The second gap was to test the predictions

of loss aversion and regulatory focus theories using a qualitative means of presenting

equivalent gains and nonlosses in the context of labeling claims on meat. The purpose

of this study was to compare the persuasive effects of gain- and nonloss-framed

labeling claims. With loss aversion and regulatory focus theory as the theoretical

framework, the objectives of this study were to determine the effects of differently

framed labeling claims on consumers' attitudes toward the product with production

claims, the conventional product, and voting intention.

Two types production labeling claims were constructed (gain and nonloss) to

market a package of boneless, skinless chicken breasts as being good for animal

welfare and the environment. Originally, health claims were also intended to be used,

but extensive protesting warranted eliminating them from the study. A convenience

sample of college students were randomly assigned to receive one of three sets of

labeling claims: nonloss-framed claims on animal welfare and environmental impact,

gain-framed claims on animal welfare and environmental impact, or general/neutral

claims related to the cut of meat. To determine the effects of the treatments, attitudes

toward the product with the claims, attitudes toward the product without the claims, and

the decision on a hypothetical voting scenario were measured.


110









Chapter 4 discussed data analyses and results from 660 subjects. The mean age

was 21 years old and the sample was primarily female. Most subjects' considered

themselves from suburban areas and did not have a livestock production background

nor planned to in the near future. The subjects tended to be more Democrat and Liberal

in their political viewpoint than Republican and Conservative or Independent and neither

Liberal nor Conservative. Most subjects' indicated that they consume meat on a daily to

twice daily basis and are the primary grocery shopper for themselves or household.

Less than half of the subjects pay attention to five types of production labeling claims

and most do not purchase products with these claims or only every so often (less than

once per month). They pay the most attention to hormone labeling and the least to

claims about how the animal was raised. This chapter presents the key findings,

discussion/implications, limitations, recommendations, and conclusions.

Key Findings

Descriptive analysis of the data found that subjects were generally more promotion

focused than prevention focused; however, having a prevention focus did not influence

attitudes toward the product with or the one without the claims, and a promotion focus

accounted for a negligible amount (<1%) of the variance in those attitudes.

A total of four hypotheses were tested in the present study. Based on the theories

of loss aversion, framing effects, and regulatory focus, the first hypothesis predicted that

when controlling for regulatory focus, subjects exposed to nonloss-framed claims will

have more positive attitudes toward the product with production claims than those

exposed to gain-framed labeling claims or control group claims. This hypothesis was

partially supported. Regulatory focus (prevention, promotion) did not influence attitudes

toward the product with the claims so it was not included as a covariate in the analysis.


111









The gain- and nonloss-framed production labeling claims did not lead to significantly

different attitudes toward the product with the claims. Subjects in both treatment

conditions had positive attitudes toward the product with the claims that were, in fact,

nearly the same whether they were exposed to nonloss or gain claims. Subjects

exposed to gain or nonloss claims had more positive attitudes towards the product with

the claims than those exposed to neutral, general product claims, but this was more

likely an effect of the treatment conditions' use of production claims than the frames

themselves.

The second hypothesis predicted that when controlling for regulatory focus,

subjects exposed to nonloss-framed claims will have less positive attitudes toward the

product without production claims than those exposed to gain-framed labeling claims or

control group claims. This hypothesis was partially supported. Subjects exposed to gain

claims did not differ from those exposed to nonloss claims in their attitudes toward the

product without the claims. Subjects exposed to gain or nonloss claims had less positive

attitudes towards the product without the claims than those exposed to neutral, general

product claims. Again, this was likely because of the production claims subjects were

exposed to in the treatment conditions rather than the framing.

The third hypothesis stated that subjects exposed to a food product with

production claims and a product without such claims will have less positive attitudes

toward the product without the claims than those who do not see a food product with

production claims. This hypothesis was supported. The analyses for H1 and H2

suggested this would be the case, but a direct analysis was conducted to confirm it. An

independent samples t-test revealed a significant effect of exposure to the production


112









labeling claims on attitudes toward the product without the claims. Subjects exposed to

production claims had less positive attitudes toward the product without the claims,

whereas those who were not exposed to production claims had a more positive attitude

toward the product without the claims.

The fourth and final hypothesis predicted that subjects exposed to a food product

with production claims will be more likely to vote "yes" for an animal welfare ballot

initiative than those who do not see a food product with production claims. This

hypothesis was not supported. The majority of the subjects voted yes for the animal

welfare ballot initiative.

Implications

The results of this study offer several theoretical implications for loss aversion,

framing effects, and regulatory focus theories. The theoretical implications are followed

by an explanation of the practical implications based on the effects production claim

labeling had on attitudes toward the product without those claims and voting intentions

on an animal welfare ballot initiative.

Theoretical

Researchers agree that the way information is framed can influence consumers'

judgment and decision about products (see Levin et al., 1998; Rabin, 1998; Boettecher,

2004). Previous loss aversion research consistently showed that people have stronger

reactions to information presented as potential losses/nonlosses in comparison to

equivalent potential gains/nongains (Kahneman & Tversky, 1979; Tversky & Kahneman,

1981; Boettcher, 2004; McDermott, 2004). Conversely, a few other studies suggested

that gains garner a stronger reaction than nonlosses (Idson et al., 2000; Liberman et al.,

2005).


113









The present study did not find loss/gain asymmetry in support of either prediction.

Whether subjects were exposed to gain-framed production labeling claims or nonloss

claims did not matter; their attitudes toward the products were affected similarly. This

could be because the application of the message/information was directly connected

with an ordinary market good: food. Horowitz and McConnell (2002) found that the more

a good is like an "ordinary market good" then the lower is the degree of gain/loss

asymmetry. The production claims themselves, however, were less about the product

itself and more about the implications for environmental impact and animal welfare as a

result of producing that product. The environment and animal welfare are non-market

goods and cannot be directly experienced by the consumer; that is the nature of

credence attributes (Darbi & Karni, 1973). Perhaps the predictions of loss aversion

would hold when testing the production labeling claims in the absence of the food

product. While that would be a clearer test of the prediction, it is less representative of

the reality of how these production claims are frequently encountered by consumers.

Also, since attitudes toward the product (rather than the claims) were measured, the

utilitarian value inherit in the chicken may be confounding the framing effect. This is

commonly the case in advertising and marketing research; the basis of the field of

advertising is that the information provided with and/or about the product affects

consumer judgment of the product (Young, 2008).

Another reason could be that the information (the production labeling claims) in

this study was presented in a qualitative manner rather than the typical quantitative

manner used in many previous studies supporting loss aversion (Kahneman & Tversky,

1979; Tversky & Kahneman, 1981; Levin et al., 1998; Boettcher, 2004; McDermott,


114









2004) and in those supporting regulatory focus theory (Idson et al., 2000; Liberman et

al., 2005). These studies did not always use numbers, but some used examples that

could be quantified (i.e., reducing cholesterol in Levin et al., 2001). Holistic

environmental impact and animal welfare are difficult to quantify objectively (Broom,

1991; Stolze, Piorr, Haring, & Dabbert, 2000), or, at best, would be difficult for the

average consumer to fully interpret (Bateman, Dent, Peters, Slovic, & Starmer, 2007).

Consumers rely on food production certification agencies (government and third-party)

to make the interpretations and provide them a trustworthy generalization of the

meanings of good animal welfare and environmental impact (Caswell & Mojduszka

1996; Golan et al., 2001; Auriol & Schilizzi 2003). Also, framing information as gains

and nonlosses primarily affects the reference point people use to make judgments and

decisions (Heath et al., 1999). Soman (2004) explained that values are coded as gains

and losses relative to a reference point, meaning a decision is reference dependant.

Presenting information about a product in a qualitative manner might cause consumers

to automatically adjust their reference point because numerical values are not available

to encode the message as a gain or a nonloss. Therefore, qualitatively created frames

(i.e., no negative environmental impacts vs. good for the environment) may not

communicate the intended reference point strongly enough, but are, therefore, equally

persuasive on attitudes. Bateman, Day, Jones, and Jude (2009) suggested that an

individual is able to interpret that one numeric value is larger than another without

necessarily understanding its meaning, thereby leading to the reliance on heuristics and

biases to form judgment. This study attempted to frame nonlosses and gains

equivalently, but qualitatively. The results suggest that in the absence of numbers or


115









quantifiable information, the biases of loss aversion and framing effects are minimized.

The message may need to include terms that more strongly suggest a reference point,

such as "reduce environmental impact" or "improve environmental impact," to induce the

biases.

Regulatory focus (prevention, promotion) also did not explain much of the variance

in attitudinal response. This could be for the same reasons as explained earlier with

respect to loss aversion and framing effects; however, regulatory focus studies are

more likely to use what researchers call "prevention and promotion messages" that do

not involve quantifiable information and are usually not equivalent (see Aaker & Lee,

2001; Wang & Lee, 2006; Florack & Scarabis, 2006; Zhao & Pechmann, 2007).

According to regulatory focus theory, subjects' with a high promotion (prevention) focus

score should have been more sensitive to the presence of gains (nonlosses), meaning

they should have had a stronger attitudinal response to the labeling claim frame that fit

their regulatory focus. Previous research has found that negative (loss) frames are more

persuasive with prevention-focused individuals than positive (nonloss) frames, and that

positive (gain) frames are more persuasive than negative (nongain) frames with

promotion-focused individuals (Zhao & Pechmann, 2007). This may explain why a very

small correlation existed between promotion-focus and attitudes but not between

prevention focus and attitudes. Many regulatory focus researchers advise marketers to

create and frame messages that are in line with audiences' regulatory focus for effective

persuasion (Aaker & Lee, 2001; Wang & Lee, 2006; Florack & Scarabis, 2006; Zhao &

Pechmann, 2007). This study implies that a simplistic message, such as a few words in

a labeling claim, may not be strong enough to elicit the regulatory fit effect. It may take


116









additional priming of a regulatory focus to fully induce it (Freitas & Higgins, 2002;

Florack & Scarabis, 2006) or a more precise manipulation of message wording (e.g.,

avoid environmental damage, achieve positive environmental impacts).

Practical

The theoretical hypotheses testing showed that production claims framed as gains

or nonlosses produced similar positive attitudinal effects on the credence attribute

product with the production claims. Gain-framed claims produced slightly (but not

statistically significant) more positive attitudes toward the product with the claims, but

slightly less negative attitudes toward the product without the claims. Marketers of

credence attribute food products could potentially encourage purchase by placing

products with gain-framed claims in their own section of the grocery store (away from

the conventional products without the claims) and those with nonloss-framed claims

next to the conventional items. However, additional research adding price variation as

an additional independent variable would need to be considered because this study

held price consistent across both product types.

When examining the attitudinal effects of production labeling claims, subjects'

exposed to those claims had less positive attitudes toward the product without the

claims than those exposed to general product claims. The product without the claims

was meant to represent the conventional commodity product to determine how the

marketing of credence attribute products affects people's attitudes toward the

conventional product. The results did not show that exposure to production claims

produces negative attitudes toward conventional products but it did produce markedly

less positive attitudes. The results show that consumers view the conventional product

inferior to the credence attribute product on the aspects of safety, healthiness, humane


117









animal treatment, and environmental-friendliness, as well as on more general aspects.

Thus, the production claims are a source of information reducing consumers' positive

attitudes toward those aspects of conventional agriculture production and its food

products. Previous work suggested this may be the case (Klonsky & Tourte, 1998), but

this study shows that it is. The claims could serve as a prompt, causing consumers to

recall negative information from the news media and mass media (Craven & Johnson,

1999). It is unclear how much of consumers' beliefs and subsequent attitudes can be

accounted for by various communication channels (i.e., advertisements, labels, media,

websites, social media, etc.). For example, would the majority of consumers know (or

be concerned about) antibiotic and hormone use or confinement in livestock production

if they weren't inundated with more expensive products claiming to be absent of those

inputs? This study shows that the production claims are a source of information that

produces inferior attitudes toward conventional products without such claims.

The exposure to production labeling claims and subsequent attitudes produced,

however, did not translate into voting intention on an animal welfare ballot initiative.

Subjects overwhelmingly supported this law. The reason no treatment effects were seen

could be due to several reasons. First, political decision making information that affects

decisions typically comes in forms of communication (i.e., TV ads, websites, news and

editorials, etc.). This study intended to determine if food labels could be a source of

communication affecting political decision making, but did not find that to be the case.

Another reason could be that the measure was a one-item, dichotomous measure of

behavioral intent. Multiple item measurement with a wider scale would better capture

the variance that naturally exists in complex decision making.


118









Interestingly, the subjects indicated they pay the least amount of attention to

animal welfare claims and purchase food with such claims the least in comparison to

four other types of claims (no hormones, no antibiotics, organic, and environmentally

friendly). This food shopping characteristic is similar to studies surveying general adult

consumer populations (Verbeke & Viaene, 1999; Yiridoe et al., 2005; Hughner,

McDonagh, Prothero, Shultz, & Stanton, 2007). The data shows, on the other hand, that

they are willing to support legislation that would make it required of all livestock

producers to provide their livestock more space in confinement, which is an animal

welfare consideration. While subjects were willing to support a government policy, they

are not willing to "put their money where their mouth is." When evaluating a potential

risk, decision-making research has shown people must be paid more to accept a risk

than they are willing to pay to avoid that risk (Thaler, 1980; Horowitz & McConnell,

2002). While this concept is not entirely related to the findings, it suggests that people

are generally willing to pay less to avoid a risk, and supporting this legislation costs no

money (at least immediately) to avoid the risk completely. These subjects' were not

provided any additional information about the consequences of passing such a law.

When this particular law was passed in California, 63.5% of the voters supported it. This

study shows that, without education and persuasive communication efforts, that number

could be much higher with young adults.

From the post hoc analyses that explored the relationships between some of the

demographic variables and dependent variables, it is interesting to note the relationship

between community upbringing and effects on attitude toward the product without the

claims. Those who indicated they grew up in a rural area (not a farm) had more positive


119









attitudes toward the conventional product than those from subdivisions or urban areas.,

While no other statistically significant relationships were found, examination of the

means across the four groups shows that those from farm and rural communities

generally have more positive attitudes towards both products. This relationship and

pattern could be because those from a farm or rural community feel a stronger

connection with agriculture and its products as a result of knowing farmers in their

community, exposure to high school agriculture programs, and/or involvement in 4-H

and/or FFA. Rural communities are more likely to have farmers and youth agricultural

programs in comparison to suburban and urban communities (Frick, Birckenholz,

Gardener, & Matchtmes, 1995).

The other relationship of interest found in the post hoc analyses was the small,

negative correlation between attention to production claims and attitude toward the

product without claims, and purchase frequency and attitude toward the product without

claims. While paying attention to more claims and purchasing these products more

frequently does not correlate with having more positive attitudes toward products that

carry those claims, these behaviors do reduce their attitudes toward products without

such claims. It seems that what potentially encourages attention to and purchase of

products with production claims is the devaluation or fear of conventional products.

Limitations

While the present study offers several useful theoretical and practical insights,

there were some limitations that should be considered. The convenience sample of

college students is one of the key limitations, primarily for the practical implications and

recommendations. College students are still developing their consumer habits and civic

engagement. These may change with further maturity, experience, and when starting a


120









family. For example, consumers with children are more likely to learn about and

purchase organic foods (Hughner et al., 2007). Readers should carefully consider the

demographic information before applying conclusions to other populations.

Research consistently shows that consumers purchase and prefer organic and

natural meat products for health reasons (AMI & FMI, 2008, Yiridoe et al., 2005;

Hughner et al., 2007). While consumers often make connections between

environmental impacts and animal welfare and health concerns (Hughner et al., 2007),

those elements themselves are not as concerning as chemicals, hormones, and

antibiotics, which are primarily health-related (Yiridoe et al., 2005). This study was

unable to find a plausible health claim in equivalent gain and nonloss frames related to

those elements to test the effects.

The present study used a one-time only exposure to the production labeling

claims. Strong attitudes resistant to change require repeated exposure to persuasive

messages (Perloff, 2008). Repeated exposure over longer periods of time could reveal

a greater influence of the production labeling claims on attitudes. Though, the claims

used in this study are those that do exist in reality. Given about half of the subjects in

this study at least pay attention to such claims, this was likely not their first exposure to

them.

Another limitation of the study was the less-than-ideal reliability of the prevention

focus scale. Perhaps the use of another regulatory focus scale (e.g., Lockwood, Jordan,

& Kunda, 2002) would prove more reliable.


121









Recommendations


For Future Research and Theory

From a theoretical perspective, more research needs to be done examining the

effects of gains versus nonlosses. This study attempted to further some of the previous

research in that area (Idson et al., 2004; Liberman et al., 2005), but perhaps due to the

qualitative nature of the frames and the nature of the application (food product), did not

find asymmetry in the attitudinal reactions to gains versus nonlosses. Researchers in

these theoretical areas should consider future studies that attempt to manipulate gains

and nonlosses qualitatively to determine if biases are minimized as a result. The finding

that regulatory focus did not influence attitudinal response to gain- or nonloss-framed

claims should be supplemented with using regulatory focus priming to determine

potential effects of this cognitive style in combination with the use of a more reliable

measure for regulatory focus (e.g., Lockwood et al., 2002).

One of the goals of this research was to determine if and how production labeling

claims affects attitudes toward the product, and therefore, only assumptions can be

made regarding how it affects beliefs and attitudes about agricultural production. A

branch of this study would be to determine how this type of labeling affects consumers'

beliefs and attitudes about agricultural production and food safety directly. Furthermore,

a national survey of where consumers obtain their information about agriculture

production, food, and farming life would be beneficial to agricultural communications

researchers, especially as the U.S. population employed in agriculture continues to

dwindle. This would help determine the role marketing and advertising, as well as other

forms of communication, play in forming beliefs and attitudes.


122









The present study held several variables consistent to determine the effect of the

differently framed production labeling claims on attitudes. Additional manipulations of

variables such as product type, price, brand, and other packaging characteristics would

be beneficial to marketers and may produce different attitudinal effects. Also, as

mentioned in the limitations, further research into production claim labeling should test

health claims since that is the primary reason driving consumer perceptions and

purchase of such products (AMI & FMI, 2008, Yiridoe et al., 2005; Hughner et al., 2007).

A follow-up study should also include other dependent measures that may be

affected by food labeling claims. Behavior, such as willingness to pay and purchase

likelihood, would offer additional insight into the effects of food labels. In addition, while

attitudes can be a useful measure of food label communication effects, it would be

worthwhile to examine other effects such as risk perceptions. In people's subjective

evaluation of risk, nine general properties of activities or technologies emerge: (1)

voluntariness of risk, (2) immediacy of effect, (3) knowledge about the risk by the person

who are exposed to the potentially-hazardous risk source, (4) knowledge about the risk

in science, (5) control over the risk, (6) newness, i.e. are the risks new and novel or old

and familiar ones, (7) chronic/ catastrophic, (8) common/dread, i.e. whether people

have learned to live with and can think about the risk reasonably and calmly, or is it a

risk that people have great dread for, and (9) severity of consequences (Fischoff, Slovic,

Lichtenstein, Read, & Combs, 2000). Using those dimensions of risk, a measure of

livestock production risk perceptions could be measured. As previously mentioned, a

more complex measure of voting intention would also be useful in capturing greater

variability in the potential effects of food labeling as communication affecting political


123









decisions. The manipulations of nonloss and gain messages in these studies should

include terms like "reduce" and "improve" to more strongly suggest a reference point

that is moved toward or away from to determine if the biases of loss aversion and

regulatory focus fit effect are subsequently induced.

For Practitioners

In this study, exposure to production labeling claims about animal welfare and

environmental impact reduced positive attitudes toward the product without such claims.

Specifically, the conventional product was viewed as inferior to the credence attribute

product on the aspects of safety, healthiness, humane animal treatment, and

environmental-friendliness, as well as on more general aspects. While this is likely

viewed as a positive finding for those with a vested interest in alternative agriculture

production and products, it is probably concerning to those who believe in the merits of

conventional agriculture. The marketing of credence attribute products contributes to the

devaluation of products that do not have such claims; however, many products, even

those from conventional systems could qualify for many different types of production

claims. It is recommended that those within the agricultural industry develop a system to

explore the facets of farming operations that may qualify food products for production

and/or processing claims, especially those related to health and food safety, animal

welfare, and environmental impact. The results of this study also imply that agricultural

communicators working on behalf of conventional agriculture need to help rebuild

attitudes toward that type of production system and help consumers understand the

meanings and implications of various food labels. They also need to assist in

communication efforts with opinion leaders, policy makers, and voters on agricultural

policy issues. Beyond that, agricultural communicators should help their organizations


124









and businesses understand and value these attitudes because the controversy over

alternative agriculture and conventional agriculture is far from over (see Paarlberg, 2010

and Lappe, 2010).

Government food regulators must consider the effects of food labeling to ensure

the policies, standards, and guidelines for such labels are balancing the market for

agricultural products and not misleading consumers (Golan et al., 2001). If organic and

other credence attribute labeled foods continue to be perceived as the safer and better

food choice, the market for conventional foods could potentially suffer. Government

regulators must meticulously consider these types labeling claims before approving

them and be responsible for communicating their meaning to consumers.

Marketers and advertisers intending to apply the principles of loss aversion,

framing effects, and regulatory focus need to carefully consider the limitations of these

theories, especially when making general, qualitative claims. Careful development and

testing of the framed messages needs to be conducted to ensure the message-frame-

audience combinations produce the intended effects.

Conclusions

Using loss aversion and regulatory focus as a theoretical framework, the

objectives of this study were to determine the effects of differently framed labeling

claims on consumers' attitudes toward the product with production claims, the

conventional product, and voting intention. This study attempted to frame nonlosses and

gains equivalently, but qualitatively. The results suggest that in the absence of numbers

or quantifiable information, the biases of loss aversion, framing effects, and regulatory

focus fit effect are minimized. Advertisers and marketers should carefully consider the


125









potential limitations of these theories and thoroughly test differently framed messages or

claims before intending to leverage the power cognitive heuristics and biases.

The mere exposure to the production labeling claims, no matter how framed,

produced equally positive attitudes toward the product; however, their presence also

decreased the positive attitudes held toward the product without such claims. These

types of food labels are a source of information affecting consumers' attitudes towards

conventional agriculture products and perhaps even the production system. Agricultural

communicators should not underestimate the effects that food marketing and

advertising can have on consumers' attitudes toward conventional agriculture and its

products, and consider these effects in addition to messages put forth by activist groups

and mass media.


126







APPENDIX A
VERBAL PRE-NOTIFICATION

Whoa, extra credit?!

15-minute online survey to gather
students' opinions about food products
and labeling

E-mail from Katie Abrams with link to
survey and your unique participant ID
will be sent to your GATORLINK e-mail
account this Thursday

Survey expires:
Sunday, March 28 at 11:59 p.m.

Contact Katie with questions/problems:
kchodil@ufl.edu

PowerPoint slide used by instructors of courses which contained the sample for the
study to announce the upcoming study.


127









APPENDIX B
FIRST CONTACT E-MAIL SENT TO SUBJECTS

Dear ${m://FirstName},

I am conducting a study about food products and would like to gather your opinions
about them through an online survey. Your instructor has agreed to offer you extra
credit in ${e://Field/course} for participating in the study. The survey will take
approximately 15 minutes of your time.

To take the survey, you will need to know your unique participant ID number, which is
${e://Field/ID}. Please take extra care to type in the correct ID number.

Do NOT delete this e-mail if you think you might want to take the survey later because it
contains your unique participant ID number. The subsequent reminder e-mails may not
contain this required information.

The survey is online. Click or copy-paste the following link into your Internet browser:
${l://SurveyURL}. The link will only be active until 11:59pm, March 28.

If you have questions or problems accessing the survey, please e-mail me at
kchodil@ufl.edu or call 352-392-0502 ext. 238.

Follow this link to the Survey:
${l://SurveyLink?d=Take the Survey}


Or copy and paste the URL below into your internet browser:
${l://SurveyURL}


Thank you,

Katie Abrams
Graduate Assistant
University of Florida


128









APPENDIX C
FIRST AND SECOND REMINDER E-MAIL SENT TO SUBJECTS

Dear ${m://FirstName},

You are receiving this e-mail because I do not have record of you completing the survey
for extra credit in ${e://Field/course} for completing it. You certainly are not obligated to
complete the survey; this is simply a reminder message. It will take approximately 15
minutes of your time.

To take the survey, you will need to know your unique participant ID number, which is
${e://Field/ID}. Please take extra care to ensure you type in the correct participant ID
number.

Your instructor will have the list of names for extra credit by March 31, so please check
with them at that time to ensure you got the points.

The survey is online. Click or copy-paste the following link into your Internet browser:
${l://SurveyURL}. The link will only be active until 11:59pm, Sunday, March 28.

If you have questions or problems accessing the survey, please e-mail me at
kchodil@ufl.edu or call 352-392-0502 ext. 238.

Follow this link to the Survey:
${l://SurveyLink?d=Take the Survey}


Or copy and paste the URL below into your internet browser:
${l://SurveyURL}


Thank you,

Katie Abrams
Graduate Assistant
University of Florida


129









APPENDIX D
THIRD REMINDER E-MAIL SENT TO SUBJECTS

Dear ${m://FirstName},

Today is the final day to take the online survey for extra credit in ${e://Field/course}. The
survey will take approximately 15 minutes of your time.

To take the survey, you will need to know your unique participant ID number, which is
${e://Field/ID}. Please take extra care to ensure you type in the correct participant ID
number.

Your instructor will have the list of names for extra credit by March 31, so please check
with them at that time to ensure you got the points.

The survey is online. Click or copy-paste the following link into your Internet browser:
${l://SurveyURL}. The link will only be active until 11:59pm tonight.

If you have questions or problems accessing the survey, please e-mail me at
kchodil@ufl.edu or call 352-392-0502 ext. 238.

Follow this link to the Survey:
${l://SurveyLink?d=Take the Survey}


Or copy and paste the URL below into your internet browser:
${l://SurveyURL}


Thank you,

Katie Abrams
Graduate Assistant
University of Florida


130













APPENDIX E

INSTRUMENT



UNIVERSITY of

UF FLORIDA


IRB.
Informed Consent
Protocol Tite: Food Labeling Claims
Protocol Number 2010-U-132

Please read this consent statement carefully before you decide to participate in this study.

Purpose of the research study: The purpose of this research is to examine college students' attitudes toward food labeling.
You wil be asked questions about your attitudes and personal experiences about purchasing food.

What you wl be asked to do in this study: First, you wil answer some questions about your personality. Then, you will be
shown food products and provide your opinions about the products you see. You wi also be asked to hypothetically vote for
a change in state laws. Finally, you wil answer some demographic and food preferences questions. You can skip any
question you do not wish to answer.

Time required: About 15 minutes

Risks and Benefits We do not anticipate there wll be any risks or direct benefits to you as a consequence of your decision
to complete the study.

Compensation: You wil receive extra course credit from your instructor.

Confidential Your name wil not appear on the questionnaire itself. Your identity wi be kept confidential to the extent
provided by law. Your instructor wil need your name to give you extra credit, but wll not receive your name before the final
exam. Only the researcher wi have access to a ie to match your participant ID number and name. Your responses are
completely confidential and no reference wil be made in any oral or written report that would link you individually to the
study. Furthermore, the principal investigator works independently of your instructor.

Voluntary participation: Your participation in this study is completely voluntary.

Right to withdraw from the stud: You have the full right to withdraw from the study at anytime without consequence.

Whom to contact if you have questions about the study: Katie Abrams, graduate assistant, Department of Agricultural
Education and Communication, University of Florida, address: PO Box 110540, University of Forida, Gainesvile, FL,
32611-0540. phone: 352.392.0502 ext 238 emai kchoditufledu or supervisor Dr. Tracy Irani, associate professor,
Department of Agricultural Education and Communication, University of Florida, address: PO Box 110540, University of
Florida, Gainesvile, FL, 32611-0540. phone: 352.392.0502 ext 225 email: rani@uffedu.

Whom to contact about your right as a research participant in this study: UFIRB, Box 112250, University of Florida,
Gainesvile, FL 32611-2250; ph 352.392.0433

c I have read the .'...-i.:.. .. .;.1 ..=:.1 above. I .i.ii rnilli agree to participate in


0 I have read the .r..iii: i- i.v.1- : ii.. l above. I do NOT agree to participate in the
procedure.


____ Sif^CCTnpLag^ ___
Q%[~l "jlOP


131











UNIVERSITY of
UF FLORIDA

Q1. Please enter your participant ID number that can be found in the e-mail that contains the link to this
survey.



oI ra


Survey Powered Bv Qual ric


m0m


U F UNIVERSITY of
UF FLORIDA

Q2 What classes) are you taking this survey for? (Select ALL that you are enrolled in. You
will get extra credit in each class you are enrolled by taking the survey ONCE.)

0 AEE 3030: Effective Oral Communication
E AEE ,::.::' Research and Business Writing
E JOU 1100: Intro to Journalism
] MMC 2604: Mass Media and You

JSurvBe GcoY m
ffjtj "|lOa~i


Survey Powered By Qualtiics


m m


132













UNIU ERSITY of
UF FLORIDA

T1. At any time during this survey, you can use the back-button feature at the bottom of each
page to go back to previous questions to review or change your response.

The first set of questions are about your personality. Please read each question and
respond thoughtfully.


SuE;, Corpdalr
*:riJ 10OW


Suvey Powered By Qualtrir


m Gm


133












Q3. Compared to most people, are you typically UNABLE to get what you want out of life?

Never Seldom Sometimes Often Very Often
0 0 0 : 0



04. Growing up, how often would you "cross the line" by doing things your parents would
not tolerate?

Never Seldom Sometimes Often Very Often




Q6. How often have you accomplished things that got you psychedd" to work even
harder?

Never Seldom Sometimes Often Very Often




Q6. Did you get on your parents' nerves often when you were growing up?

Never Seldom Sometimes Often Very Often
( D C) :7) C,



07. How often did you obey rules and regulations that were established by your parents?

Never Seldom Sometimes Often Very Often
) DC C7 C: ",



Q8. Growing up, did you ever act in ways that your parents thought were objectionable?

Never Seldom Sometimes Often Very Often




Q9. Do you often do well at different things that you try?

Never Seldom Sometimes Often Very Often




010. Not being careful enough has gotten me into trouble at times.

Never Seldom Sometimes Often Very Often


veC, Crp r c
--- ----- I .--


134











U F LI UNIVERSITY of
UF FLORIDA

011. When it comes to achieving things that are important to me, I find that I don't perform
as well as I ideally would like to.


Certainly False
0


Somewhat Neither True Nor
False False


Somewhat True Certainly True


012. I feel like I have made progress toward being successful in my life.


Certainly False


Somewhat Neither True Nor
False False


Somewhat True Certainly True


sunruv carpim
0%r I OY%)4


Survey Prered By C~uAtrir


mC mI


I UNIVERSITY of
U (FLORIDA

T2. On the next screen you will see two photos of food products. You will be asked to
examine these products and asked for your opinions about them.


0% 10Y%


Survey Powered By Qualtrcs


m8 m


135












jJIf UNIVERSITY of
UF FLORIDA

Directions. Carefully examine these two packages and their labels.

Product A


Product A


Product B


Product B


Q13. Are you able to view the images above?
SYes
SNo


Survey Pcere5d By Qualtes


m m


136


______ Sjr -- -













Gain condition


Directions. Carefully examine these two packages and their labels and answer the questions beklw the photos.

Product A Product B


Product A


Product B


Q14A. I feel that Product A is...

Useless
Worthless
Harmful
Foolish
Unpleasant
Awful
Disagreeable
Sad
Bad
Negative
Dislike
Unfavorable
Unhealthy
Unsafe to eat when cooked
From an animal treated inhumanely
Bad for the environment


I0 I: I ? i




(D'













C' ci i































137


Useful
Valuable
Beneficial
Wise
Pleasant
Nice
Agreeable
Happy
Good
Positive
Like
Favorable
Healthy
Safe to eat when cooked
From an animal treated humanely
Good for the environment


chicken breadi





M AP I Im h nAT
II v roam ,













Gain condition




Directions. Carefully examine these two packages and their labels and answer the questions below the photos.


Product A


Product A


Product B


Product B


Q14B. I feel that Product B is...

Useless
Worthless
Harmful
Foolish
Unpleasant
Awful
Disagreeable
Sad
Bad
Negative
Dislike
Unfavorable
Unhealthy
Unsafe to eat when cooked
From an animal treated inhumanely
Bad for the enAronment


') 0 y O Useful
,, ,0 0 ,:) 0 Valuable
S (0 Beneficial
Wise
-, Pleasant
0 0 Nice
( Agreeable
0 K O 0 0 Happy
0 0 0 0 Good
( D :, 0 Positive
Like
' ) 0 0 Favorable
S O 0 0 Healthy
: Safe to eat when cooked
, 0 From an animal treated humanely
'? i) 'i i 5 uGood for the enAronment


138


bL. eats & -uman
chicken breast

good *r trhe nnrcKnm W
Ire e r*or


VA--IWA|jD


'A II~E~












Other Treatment Groups


Nonloss Condition Photos


Product B


Product B


Control Condition Photos


Product B


Product B


139


Product A


Product A


Product A


Product A


bjnilrss hlnlan
cr-icken breasts

bhoiele. & skknlei
chiderm breastr


EEP fEFRml6RATED











|UNI U 'VERSIT) of
UFLORIDA

T3. Thank you for providing your opinions about the two chicken products.

Next, we would like you to read a potential state law regarding the confinement of livestock and
indicate whether you would or would not support such a law.


Sue;, Cormpn
0r l10rh


Survey PoWered By Qualtis




IJ|f UNIVERSITY of
UF FLORIDA

Q15.
On the next ballot in your state, the following initiative regarding the confinement of
livestock is being proposed.

Calves raised for veal, egg-laying hens, and pregnant pigs can be confined only in ways
that allow these animals to lie down, stand up, fully extend their limbs, and turn around
freely. Exceptions made for transportation, rodeos, fairs, 4-H programs, lawful slaughter,
research and veterinary purposes. Under the measure, any person who violates this law
would be guilty of a misdemeanor, punishable by a fine of up to $1,000 andlor
imprisonment in county jail for up to six months.

How do you plan to vote?

SYes
N No


0% 100%paa


Survey Powered ByQualics


m m


140












W f UNIVERSITY of
UF IFLORIDA

T4. Now, we'll ask you some questions about your demographics and grocery shopping
choices.











jJ f |UNIVERSITY of
UF IFLORIDA

Q16. What year were you born?




Q17. What is your gender?

0 Male
0 Female



Q18. Which of the following categories represents where you lived growing up?

A farm in a rural area
C Rural area, not a farm
SSubdivision in town or city
0 Downtown area in a city or town



Q19. Do you or anyone in your immediate family work in the livestock production industry?

C Yes, n the past
C0 Yes, currently
C I or my family plan to within the next 4 years
0 No


Survey Pweered By -ualivs


m m


141


S-YO.PMWm
njj jj jjjj 1











SU'NIV tRSITY of
U IFLORIDA

Q20. Approximately how often do you eat meat (including all meals and snacks) in a typical
week?

C) Never
0 Less than once per week
O 1 to 3 times per week
0 4 to 7 times per week
0 8 to 14 times per week
C More than 14 times per week






Survey Po ered B'y "Quai


142












j If UNIVEERSIT) of
U IFLORIDA

Q21. Generally speaking, would you call yourself a(n):

C Strong Democrat
0 Not very strong Democrat
0 Independent leaning Democrat
C Independent
O Independent leaning Republican
O Not very strong Republican
D i::I iiong Republican
) Other
0 Prefer not to answer



Q22- When it comes to politics, would you describe yourself as:

0 Very Liberal
C Liberal
O Somewhat Liberal
O Neither Liberal nor Conservative
0 Somewhat Conservative
0 Conservative
0 Very Conservative
0 Prefer not to answer

su cmpirr c %




Survey PoRwred By Qualtns m m


143










yW f UNIVERSITY of
UF FLORIDA

Q23 Do you currently shop for groceries for yourself or your household?
0 Yes
C' No


Survey Powered By Qualtiks


m mr


F L|UNI VERSITY of
UF FLORIDA

Q23A. Do you help decide what food should be purchased at the grocery store for yourself
or your household?
0' Yes
0 No


Jrm,- DXTPLXacr
ltF 10D,%~


Survey Powered By Qualtrs


m m


144











f I |LINIV' ERSI T Y of
UFFLORIDA

Q24. When grocery shopping, do you pay attention to...

Yes No
organic labels on
meat and chicken?
labels that address
the w_%, r-ie animal
was raised (i.e.,
humanely raised,
free range,
cage-free)?
labels on meat or
chicken that say no 0
hormones?
labels on meat or
chicken that say no 0
antibiotics?
labels on meat or
chicken that
suggest it is better
for the environment
'.i reCenp")?




Survey e:i Byc ui~r sp


Survey Pcwerei By Qualtikismm


145












yU F UNIVERSITY of

U FLORIDA


025 Approximately how often do you purchase meat or chicken with an organic label?


Less than
Never once a month


Once a Twice a
month month
IC


Q26. Approximately how often do you purchase meat or chicken with a label that
addresses the way the animal was raised (i.e., humanely raised, free range, naturally
raised)?


Less than
Never once a month


Once a Twice a
month month


Every time I
''--e 1 purchase


Q27. Approximately how often do you purchase meat or chicken with a label that says no
hormones?


Less than
Never once a month


Once a Twice a
month month


Every time I
' e.' e I' purchase


Q28. Approximately how often do you purchase meat or chicken with a label that says no
antibiotics?


Less than
Never once a month


Once a Twice a Every time I
month month ''e purchase
_, K, C


Q29. Approximately how often do you purchase meat or chicken with a label that suggest
it is better for the environment rareen)?


Less than
Never once a month
K) QI


m mbi


Every time I
"'ee 1 purchase


Once a
month


Twice a
month
0


Every time I
'ee purchase


Survey Poered y CaEmlr


146


S-r Cop er o
CAI lo'












Finally, w'd like to aRSTk you about your experience taking this survey.f
UK FLORIDA

TS. Finally, we'd like to ask you about your experience taking this survey.


Su rvey Powered By Quates


m[ m


Gain condition


jW |UNIVERSITY of
UF FLORIDA

Pics.
ProductA


ProductA


Product B


Product B


Q30 When you were providing your opinions about these two products earlier in the survey, did you notice the difference
between the information provided on the labels?
Yes
SNo
SI could not see the images


Survey Pewerea By QuaMrf


mB m


147


Inellfl & irlkrl.
chicken breasts
gcd fr i nvron1It -

n n m 1- I -4
^*ft"












Gain condition


Product B


chicken brcaris

fre- 10 rolan v
<,n,,,M'..
mramlu


Product B


Q31. Which of these two products was better for the environment and the animal's welfare?

Product A
Product B


Survey Powered By Quaftn


m m


148


Product A


Product A


ri-.












Control condition


Product A


Product A


Q31. Which of these two products contained more information on the label?

SProduct A
C Product B


Suvey Pn vred By Qualnst


SUNIV ERSITY of

U FLORIDA

Thank you for completing the survey. The purpose of this study was to see how different
food labels affect people's attitudes toward the product and their voting decision on an
animal confinement law. The state statute you voted on was only hypothetical and will not
affect any current state laws regarding the confinement of livestock. This was, however, the
language from a proposition passed in the state of California in 2008. If you have questions
or concerns, please e-mail the principal investigator, Katie Abrams at kchodll@ufl.edu.

Sujw;y xrefl on


Survey Pomered By Qualtric


149


Product B


Product B


>-hicl;en hre ,ti

6--aIcG -. & dM =.


iMfPIrEiPATED


m mr


i7a,anrian
nz n~m~









LIST OF REFERENCES


Abrams, K. M., & Meyers, C. A. (2009). A comparison of persuasive message factors
and frames in animal agriculture communication campaigns on the Web. Paper
presented at the Association for Communication Excellence Conference in
Agriculture, Natural Resources, and Life Sciences. Des Moines, IA.

Abrams, K. M., Meyers, C. A., & Irani, T. A. (2009). Naturally confused: Consumers'
perceptions of organic and all-natural pork products. Journal of Agriculture and
Human Values [in press]. doi: 10.1007/sl 0460-009-9234-5

American Meat Institute & Food Marketing Institute. (2008). The power of meat: An in-
depth look at meat through the shoppers' eyes. Paper presented at the Annual
Meat Conference, Nashville, TN. Retrieved August 17, 2009, from
http://www.meatconference.com/ht/a/GetDocumentAction/id/38142

Ary, D., Jacobs, L. C., & Razavieh, A. (2002). Introduction to research in education (6th
ed). Belmont, CA: Wadsworth/Thomson Learning.

Auriol, E. & Schilizzi, S.G.M. (2003). Quality signaling through certification: Theory and
an application to agricultural seed markets. Working paper no 165, IDEI University
of Toulouse.

Bateman, I. J., Day, B. H., Jones, A. P., & Jude, S. (2009). Reducing gain-loss
asymmetry: A virtual reality choice experiment valuing land use change. Journal of
Environmental Economics and Management, 58(1), 106-118.

Bateman, I. J., Dent, S., Peters, E., Slovic, P., & Starmer, C. (2007). The affect heuristic
and the attractiveness of simple gambles. Journal of Behavioral Decision Making
20(4), 365-380.

Batra, R., & Ahtola, O. T. (1991). Measuring the hedonic and utilitarian sources of
consumer attitudes. Marketing Letters, 2(2), 159-170.

Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D. (2001). Bad is
stronger than good. Review of General Psychology, 5(4), 323-370.

Bech-Larsen, T., Grunert, K. G., & Poulsen, J. B. (2001). The acceptance of functional
foods in Denmark, Finland and the United States: A study of consumers' conjoint
evaluations of the qualities of functional foods and perceptions of general health
factors and cultural values. MAPP working paper no 73. Aarhus, Germany: MAPP
Center.

Becker, T., Benner, E., & Glitsch, K. (2000). Consumer perception of fresh meat quality
in Germany. British Food Journal, 102(3), 246-266.


150









Blandford, D., Bureau, J.-C., Fulponi, L., & Henson, S. (2002). Potential implications of
animal welfare concerns and public policies in industrialized countries for
international trade. In B. Krissof, M. Bohman, & J. Caswell, Global Food Trade and
Demand for Quality (pp. 77-100). New York: Kluwar Academic/Plenum Publishers.

Bostr6m, M., & Klintman, M. (2003). Framing, debating, and standardising "natural food"
in two different political contexts: Sweden and the U.S. Score Working Paper no.
2003:3. Stockholm, Sweden: Stockholm Center for Organizational Research,
Stockholm School of Economics.

Brenner, L., Rottenstreich, Y., Sood, S., & Bilgin, B. (2007). On the psychology of loss
aversion: Possession, valence, and reversals of the endowment effect. Journal of
Consumer Research, 34, 369-376.

Bruns0, K., Fjord, T., & Grunert, K. (2002). Consumers food choice and quality
perception. MAPP working paper no 77. Aarhus, Germany: MAPP Center.

California Institute for Rural Studies. (2005). Regulating organic: Impacts of the national
organic standards on consumer awareness and organic consumption patterns.
Retrieved August 20, 2009, from
http://www.cirsinc.org/Documents/Pub1205.2. PDF

Caswell, J. A. (1998). How labeling of safety and process attributes affects markets for
food. Agricultural and Resource Economics Review, 27(2), 151-158.

Caswell, J. A., & Mojduszka, E. M. (1996). Using informational labeling to influence the
market for food quality. American Journal of Agricultural Economics, 78(5), 1248-
1253.

Cesario, J., Higgins, E. T., & Scholer, A. A. (2008). Regulatory fit and persuasion: Basic
principles and remaining questions. Social and Personality Psychology Compass,
2(1), 444-463.

Craven, B., & Johnson, C. (1999). Politics, policy, poisoning and food scares. In Morris,
J., & Bate, R. (Eds.). Fearing Food: Risk, Health and Environment (pp. 141-169).
Oxford: Butterworth Heinemann.

Crowell, S. (2009, April 2). The three R's of the HSUS agenda. Farm and Dairy.
Retrieved August 19, 2009, from http://www.farmanddairy.com/columns/the-three-
rs-of-the-hsus-agenda/l 1606.html

Crowley, A. E., Spangenberg, E. R., & Hughes, K. R. (1992). Measuring the hedonic
and utilitarian dimensions of attitudes toward product categories. Marketing
Letters, 3(3), 239-249.

Dangour, A. D., Dodhia, S. K., Hayter, A., Allen, E., Lock, K., & Uauy, R. (2009).
Nutritional quality of organic foods: A systematic review. American Journal of
Clinical Nutrition [in press]. doi: 10.3945/ajcn.2009.28041


151









Darbi, M. R., & Karni, E. (1973). Free competition and the optimal amount of fraud.
Journal of Law and Economics, 16(1), 67-88.

DeGregori, T. R. (2003). Origins of the organic agriculture debate. Ames, IA: Iowa State
Press.

DeVellis, R. F. (2003). Scale development: Theory and applications (2nd ed). Thousand
Oaks, CA: Sage.

Downing, B. (2009, July 19). Humane Society, farmers prepare for war: Battle lines are
forming over proposal to change Ohio rules on methods of confining livestock.
Akron Beacon Journal. Retrieved August 19, 2009, from
http://www.ohio.com/news/51120387.html

Dubisch, J. (2004). You are what you eat: Religious aspects of the health food
movement. In C. L. Delaney (Ed.). Investigating culture: An experiential
introduction to anthropology (pp. 311-319). Malden, MA: Blackwell Publishing.

Dunlap, R., & Mertig, A. (Eds.). (1992). American environmentalism: The US
environmental movement, 1970-1990. London: Taylor and Francis.

Druckman, J. N. (2004). Political preference formation: Competition, deliberation, and
the (ir)relevance of framing effects. American Political Science Review, 98, 671-
86.

Druckman, J. N. (2001). On the limits of framing effects: Who can frame? The Journal of
Politics, 63(4), 1041-1066.

Eicher, A. (2003). Organic agriculture: A glossary of terms for farmers and gardeners.
Eurka, CA: University of California Cooperative Extension. Retrieved May 7, 2010
from http://ucce.ucdavis.edu/files/filelibrary/1068/8286.pdf

Entman, R. (1993). Framing: Toward clarification in a fractured paradigm. Journal of
Communication, 43(4), 51-58.

Evans, L. M., & Petty, R. E. (2003). Self-guide framing and persuasion: Responsibility
increasing message processing to ideal levels. Personality and Social Psychology
Bulletin, 29(3), 313-324.

Factory. (1989). Oxford English Dictionary. Retrieved May 14, 2010, from
http://dictionary.oed.com/cgi/entry/50081547?

Fischoff, B., Slovic, P., Lichtenstein, S., Read S., & Combs, B. (2000). How safe is safe
enough? A psychometric study of attitudes toward technological risks and benefits.
In: P. Slovic (Ed.), The perception of risk (pp. 80-104). London: Earthscan.

Fitzgerald, D. (2003). Every farm a factory: The industrial ideal in American agriculture.
New Haven, CT: Yale University Press.


152









Florack, A. & Scarabis, M. (2006). How advertising claims affect brand preferences and
category-brand associations: The role of regulatory fit. Psychology and
Marketing, 23(9), 741-755.

Florack, A., Scarabis, M., & Gosejohann, S. (2005). Regulatory focus and consumer
information processing. In Kardes, F., Herr, P., & Nantel, J. (Eds.), Applying social
cognition to consumer-focused strategy (pp. 235-263). Mahwah, NJ: Lawrence
Erlbaum Associates.

Freitas, A. L., & Higgins, E. T. (2002). Enjoying goal-directed action: The role of
regulatory fit. Psychological Science, 13, 1-6.

Frick, M., Birckenholz, R., Gardener, H., & Matchtmes, K. (1995). Rural and urban
inner-city high school student knowledge and perception of agriculture. Journal of
Agricultural Education, 36(4), 1-9.

Gabbett, R. J. (2008, Dec. 10). Meat industry faces emboldened animal rights lobby
next year. Meatingplace Industry News. Retrieved August 19, 2009, from
http://www.meatingplace.com/MembersOnly/webNews/details.aspx?item=10700

Garner, R. A. (1993). Animals, politics & morality. Manchester University Press:
Manchester.

Gilovich, T. (1991). How we know what isn't so: The fallibility of human reason in
everyday life. New York: Free Press.

Gilovich, T., & Griffin, D. W. (2002). Heuristics and biases: Then and now. In Gilovich,
T., Griffin, D. W., & Kahneman, D. (Eds.), Heuristics and biases: The psychology
of intuitive judgment (pp. 1-18). Cambridge: Cambridge University Press.

Gilovich, T., Griffin, D. W., & Kahneman, D. (2002). Heuristics and biases: The
psychology of intuitive judgment. Cambridge: Cambridge University Press.

Golan, E., Kuchler, F., & Mitchell, L. (2001). Economics of food labeling. Journal of
Consumer Policy, 24, 117-184.

Gold, M. V. (2010, March). Should I purchase organic foods? USDA Alternative Farming
Systems Information Center. Retrieved May 24, 2010, from
http://www.nal.usda.gov/afsic/pubs/faq/BuyOrganicFoodsD.shtml

Goodwin, J., & Rhoades, E. (2009). Agricultural legislation: The presence of California
Proposition 2 on YouTube. Paper presented at the National American Association
for Agricultural Education Conference, Louisville, KY. Retrieved August 19, 2009,
from http://www.aaaeonline.org/files/national_09/papers/2.pdf

Gottlieb, R. (2005). Forcing the spring: The transformation of the American
environmental movement. Washington DC: Island Press.


153









Grant, J. (2008). The green marketing manifesto. John Wiley & Sons. Available from
http://common.books24x7.com/book/id_23397/book.asp

Greene, C., Dimitri, C., Lin, B., McBride, W., Oberholtzer, L., & Smith, T. (2009).
Emerging issues in the U.S. organic industry (Economic Information Bulletin No.
55). Washington DC: United States Department of Agriculture Economic Research
Service. Retrieved August 19, 2009, from
http://www.ers.usda.gov/Publications/EIB55/EIB55.pdf

Guthman, J. (2004). The trouble with 'organic lite' in California: A rejoinder to the
conventionalismm' debate. Sociologia Ruralis, 44(3), 301-316.

Hale, T. (2009, June 12). Most households read food labels. Nielsen Wire. Retrieved
August 24, 2009, from http://blog.nielsen.com/nielsenwire/consumer/most-
households-read-food-labels

Harris poll results show who is buying organic foods, how frequently. (2007, October).
Nutrition Business Journal, 12(10), 21.

Heath, C., Larrick, R. P., & Wu, W. (1999). Goals as reference points. Cognitive
Psychology, 38, 79-109.

Higgins, E. T. (1998). Promotion and prevention: Regulatory focus as a motivational
principle. In Zanna, M. P. (Ed.), Advances in Experimental Social Psychology (pp.
1-46). New York: Academic Press.

Higgins, E. T. (2002). How self-regulation creates distinct values: The case of promotion
and prevention decision making. Journal of Consumer Psychology, 12(3), 177-
191.

Higgins, E. T., Friedman, R. S., Harlow, R. E., Idson, L. C., Ayduk, O. N., & Taylor, A.
(2001). Achievement orientations from subjective histories of success: Promotion
pride versus prevention pride. European Journal of Social Psychology, 31, 3-23.

Higgins, E. T, & Silberman, I. (1998). Development of regulatory fo- cus: Promotion and
prevention as ways of living. In Heckhausen, J., & Dweck, C. (Eds.), Motivation
and self-regulation across the life span (pp. 78-113). New York: Cambridge
University Press.

Hooker, R. J. Food and drink in America. Indianapolis: The Bobs-Merrill Company, Inc.

Horowitz, J. K., & McConnell, K. E. (2002). A review of WTA/WTP studies. Journal of
Environmental Economics and Management 44, 426-447.

Hu, W., Woods, T., & Bastin, S. (2009). Consumer acceptance and willingness to pay
for blueberry products with non conventional attributes. Journal of Agricultural and
Applied Economics, 41(1), 47-60.


154









Hughner, R., McDonagh, P., Prothero, A., Shultz II, C., & Stanton, J. (2007). Who are
organic food consumers? A compilation and review of why people purchase
organic food. Journal of Consumer Behaviour, 6(2/3), 94-110. doi: 10.1002/cb.210.

Humane Farm Animal Care. (n.d.). What is certified humane raised and handled?
Retrieved September 2, 2009, from
http://www.certifiedhumane.org/about/whatis.html

Hurt, R. D. (2002). American agriculture: A brief history. West Lafayette, IN: Purdue
University Press.

Hwang, Y., Roe, B., & Teisl, M. F. (2005). An empirical analysis of United States
consumers' concerns about eight food production and processing technologies.
AgBioForum, 8(1), 40-49.

Idson, L. C., Liberman, N., & Higgins, E. T. (2004). Imagining how good or bad you'd
feel: A motivational experience beyond outcomes. Personality and Social
Psychology Bulletin, 30, 926-937.

Jacob, F., & Ehret, M. (2006). Self-protection vs opportunity seeking in business buying
behavior: An experimental study. Journal of Business & Industrial Marketing,
21(2), 106-117.

Jauregui, C., & Ward, R. W. (2006, July). Do consumers really use food labels? Paper
presented at the American Agricultural Economics Association (New Name 2008:
Agricultural and Applied Economics Association) 2006 Annual meeting, July 2006,
Long Beach, CA. Retrieved August 24, 2009, from http://purl.umn.edu/21142

Jones, S., Johnson-Yale, C., Millermaier, S., & Seoane Perez, F. (25 September, 2009).
Everyday life, online: U.S. college students' use of the Internet. First Monday
[Online], 14(10). Retrieved May 14, 2010 from
http://www.uic.edu/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/2649/2301

Kahneman, D., Diener, E., & Schwartz, N. (Eds.) (2003). Well-being: The foundations of
hedonic psychology. New York: Russell Sage Foundation Publications.

Kahneman, D., Knetsch, J., & Thaler, R. (1990). Experimental test of the endowment
effect and the case theorem. Journal of Political Economy 98(6), 1325-1348.

Kahneman, D., Ritov, I., & Schkade, D. (1999). Economic preferences or attitude
expressions? An analysis of dollar responses to public issues. Journal of Risk and
Uncertainty, 19, 220-242.

Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under
risk. Econometrica 47, 263-291.


155









Kam, C. D., Wilking, J. R., & Zechmeister, E. J. (2007). Beyond the "narrow data base":
Another convenience sample for experimental research. Political Behavior, 29(4).
doi: 10.1007/s11109-007-9037-6

Kempton, W. Boster, J. & Hartley, J. (1995). Environmental values in American culture.
Cambridge: MIT Press.

Klonsky, K., & Tourte, L. (1998). Organic agricultural production in the United States:
Debates and directions. American Journal of Agricultural Economics, 80(5), 1119-
1124.

Knorr, D., & Watkins, T. R. (1984). Alterations in food production. New York: Van
Nostrand Reinhold.

Lappe, A. (2010, April 29). Don't panic, go organic. Foreign Policy. Retrieved May 7,
2010, from http://www.foreignpolicy.com/articles/2010/04/29/
dont_panic_go_organic

Levin, I. P., & Gaeth, G. J. (1988). Framing of attribute information before and after
consuming the product. Journal of Consumer Research, 15, 374-378.

Levin, I., Schneider, S., & Gaeth, G. (1998). All frames are not created equal: A
typology and critical analysis of framing effects. Organizational behavior and
human decision processes, 76(2), 149-188.

Levenstein, H. (2003). The paradox of plenty: A social history of eating in modern
America. Berkeley and Los Angeles, CA: University of California Press.

Liberman, N., Idson, L. C., & Higgins, E. T. (2005). Predicting the intensity of losses vs.
non-gains and non-losses vs. gains in judging fairness and value: A test of the loss
aversion explanation. Journal of Experimental Social Psychology, 41(5), 527-534.

Lindenberg, S., & Steg, L. (2007). Normative, gain and hedonic goal frames guiding
environmental behavior. Journal of Social Issues, 63(1), 117-137.

Lobao, L., & Meyer, K. (2001). The great agricultural transition: Crisis, change, and
social consequences of twentieth century US farming. Annual Review of
Sociology, 27(1), 103-124.

Lockwood, P., Jordan, C., & Kunda, Z. (2002). Motivation by positive or negative role
models: Regulatory focus determines who will best inspire us. Journal of
Personality and Social Psychology, 83(4), 854-864. doi: 10.1037/0022-
3514.83.4.854

Loureiro, M. L., McCluskey, J. J., & Mittelhammer, R. C. (2005). Assessing consumer
preferences for organic, eco-labeled, and regular apples. Journal of Agricultural
and Resource Economics, 26(2), 404-416.


156









Magkos, F., Arvaniti, F., & Zampelas, A. (2006). Organic food: Buying more safety or
just peace of mind? A critical review of the literature. Critical Reviews in Food
Science and Nutrition, 46(1), 23-56.

Maheswaran, D. & Meyers-Levy, J. (1990). The influence of message framing and issue
involvement. Journal of Marketing Research, 27, 361-67.

Maule, J., & Villejoubert, G. (2007). What lies beneath: Reframing framing effects. In
Lagnado, D. A., & Read, D. (Eds.), Judgment and choice: Perspectives on the
work of Daniel Kahneman. Thinking and Reasoning, 13, 25-44.

McDermott, R. (2004). Prospect theory in political science: Gaines and losses from the
first decade. Political Psychology, 25(2), 289-312.

Miller, M. G., & Miller, K. U. (2000). Promoting safe driving behaviors: The influence of
message framing and issue involvement. Journal of Applied Social Psychology,
30(4), 853-866.

Morrison, P. C., Nehring, R., Banker, D., & Somwaru, A. (2004). Scale economies and
efficiency in U.S. agriculture: Are traditional farms history? Journal of Productivity
Analysis, 22, 185-205.

Nestle, M. (2007). Food politics: How the food industry influences nutrition and health.
Berkeley, CA: University of California Press.

Ness, M., Matthew G., & Kuznesof, S. (2002). The student food shopper: Segmentation
on the basis of attitudes to store features and shopping behavior. British Food
Journal, 104(7), 506-526.

Nisbet, M. C., Scheufele, D. A., Shanahan, J., Moy, P., Brossard, D., & Lewenstein, B.
V. (2002). Knowledge, reservations, or promise? A media effects model for public
perceptions of science and technology. Communication Research, 29(5), 584-608.

Obach, B. K. (2007). Theoretical interpretations of the growth in organic agriculture:
Agriculture modernization or an organic treadmill? Society & Natural Resources
An International Journal, 23(3), 229-244.

Oberholtzer, L., Greene, C., & Lopez, E. (2006). Organic poultry and eggs capture high
price premiums and growing share of specialty market. (LDP-M-150-01). Retrieved
November 17, 2009, from the USDA Economic Research Service website:
http://www.ers.usda.gov/Publications/LDP/2006/12Dec/LDPM15001/ldpm 15001.p
df

Organic Trade Association (2008). Health of the planet and its inhabitants. Retrieved
August 19, 2009 from http://www.ota.com/organic/benefits/health.html


157









Onyango, B. M., Hallman, W. K., & Bellows, A. C. (2007). Purchasing organic food in
US food systems: A study of attitudes and practice. British Food Journal, 109(5),
399-411.

Ostrom, T. M., Bond, Jr., C. F., Krosnick, J. A., & Sedikides, C. (1994). Attitude scales:
How we measure the unmeasurable. In S. Shavitt & T.C. Brock (Eds.) Persuasion:
Psychological insights and perspectives (pp. 15-42). Needham Heights, MA: Allyn
and Bacon.

Padel, S., & Foster, C. (2005). Exploring the gap between attitudes and behavior:
Understanding why consumers do or do not buy organic food. British Food
Journal, 8, 606-625.

Paarlberg, R. (2010, May/June). Attention Whole Foods shoppers. Foreign Policy.
Retrieved May 7, 2010, from http://www.foreignpolicy.com/articles/2010/04/26/
attention_whole_foods_shoppers

Perloff, R. M. (2008). The dynamics of persuasion: Communication and attitudes in the
21st century (3rd ed.). New York: Lawrence Erlbaum Associates.

Peterson, R. A. (2001). On the use of college students in social science research:
Insights from a second-order meta-analysis. Journal of Consumer Research, 28.
doi: 10.1086/323732

Petty, R. E., & Cacioppo, J. T. (1986). Communication and persuasion: Central and
peripheral routes to attitude change. New York: Springer Verlag.

Pollan, M. (2004). The cheapest calories make you the fattest: A food-chain journalist
looks for stories in our meals. Retrieved August 18, 20009, from
http://www.michaelpollan.com/press.php?id=7

Prop 2: Standards for confining farm animals. (2008). Retrieved November 19, 2009,
from California Voter Information Guide:
http://voterguide.sos.ca.gov/past/2008/general/title-sum/prop2-title-sum.htm

Rabin, M. (1998). Psychology and economics. Journal of Economic Literature, 36(1),
11-46.

Reips, U-D. (2000). The web experiment method: Advantages, disadvantages, and
solutions. In M. H. Birnbaum (Ed.), Psychological experiments on the web (pp. 89-
117). London: Academic Press.

Rollin, B. E. (1990). Animal welfare, animal rights, and agriculture. Journal of Animal
Science, 68(10), 3456-3461.

Rollin, B. E. (2003). Farm animal welfare: Social, bioethical, and research issues. Ames,
IA: Iowa State Press.


158









Rozin, P., Spranca, M., Krieger, Z., Neuhaus, R., Surillo, D., Swerdlin, A. et al. (2004).
Preference for natural: Instrumental and ideational/moral motivations, and the
contrast between foods and medicines. Appetite, 43(2), 147-154.

Sassenrath et al. (2008). Technology, complexity and change in agricultural production
systems. Renewable Agriculture and Food Systems, 23, 285-295.

Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-
experimental designs for generalized causal inference. Boston: Houghton Mifflin
Company.

Shamaskin, A. (2009). Getting the message across: Examining information presentation
and healthcare decision making among older adults (Unpublished thesis, Cornell
University, 2009). Retrieved September 13, 2009, from Cornell University Library
website http://hdl.handle.net/1813/12656

Shiv, B., Carmon, Z., & Ariely, D. (2005). Placebo effects of marketing actions:
Consumers may get what they pay for. Journal of Marketing Research, 42(4), 383-
393.

Simon, H. A. (1955). A behavioral model of rational choice. The Quarterly Journal of
Economics, 69(1), 99-118.

Smith, R. (2009, March 5). Prop 2 opening door to promote veganism. Feedstuffs
Foodlink. Retrieved August 19, 2009, from http://www.feedstuffsfoodlink.com/ME2/
dirmod.asp?sid=F4A490F89845425D8362C0250A1 FE984&nm=&type=news&mo
d=News&mid=9A02E3B96F2A415ABC72CB5F516B4C10&tier=3&nid=BOD5904C
A7FE4934ABEF9024D30444A8

Soman, D. (2004). Framing, loss aversion, and mental accounting. In Koehler, D. J., &
Harvey, N. (Eds.), Blackwell handbook ofjudgment and decision making, (pp. 379-
398). Maiden, MA: Blackwell Publishing.

Stolze, M., Piorr, A., Haring, A.M., & Dabbert, S. (2000) Environmental impacts of
organic farming in Europe. Organic Farming in Europe: Economics and Policy, 6.
Universitat Hohenheim, Stuttgart-Hohenheim. Retrieved from
http://orgprints.org/8400.

Storck, A. (2008). Organic meat and poultry will keep growing in 2009: trade group.
Meat Place Industry News. Retrieved August 17, 2009, from
http://www. meatingplace.com/MembersOnly/webNews/details.aspx?item=1 0801

Sustainable Agriculture Research & Education. (n.d.). Exploring sustainability in
agriculture. Retrieved September 20, 2009, from
http://www.sare.org/publications/explore/explore.pdf

Sustainable Table. (n.d.). What is sustainable agriculture? Retrieved September 20,
2009, from http://www.sustainabletable.org/intro/whatis/


159









Thaler, R. (1980). Toward a positive theory of consumer choice. Journal of Economic
Behavior and Organization, 1, 39-60.

The Humane Society of the United States. (2009, Oct. 12). Mich. Gov. Granholm signs
historic farm animal welfare measure. Retrieved November 18, 2009, from
http://www.humanesociety.org/news/press_releases/2009/10/mich_gov_granholm
_signs.html

The Humane Society of the United States. (2008). Research. Retrieved August 19,
2009, from http://www.hsus.org/farm/resources/research/

The NPD Group, Inc. (2009). "Better for you foods" to grow significantly over the next
decade. The NPD Group, Inc. Press Release. Retrieved August 17, 2009, from
http://www.npd.com/press/releases/press_090707a.html

Todorov, A., Chaiken, S., & Henderson, M. D. (2002). The Heuristic-Systematic Model
of social information processing. In Dillard, J., & Pfau, M. (Eds.), The persuasion
handbook: Developments in theory and practice (pp. 195-211). Thousand Oaks,
CA: Sage Publications, Inc.

Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology of
choice. Science, 211, 453-458.

Traub, R. E. (1994). Reliability for the social sciences: Theory and applications.
Thousand Oaks, CA: Sage.

United States Department of Agriculture Agricultural Marketing Service. (2008). LS /SO
Guide 65 Program. Retrieved September 2, 2009, from http://www.ams.usda.gov/
AMSv1.0/ams.fetchTemplateData.do?template=TemplateD&navlD=GradingCertifi
cationandVerfication&leftNav=GradingCertificationandVerfication&page=LSIS065
Program

USDA Food and Safety Inspection Service. (2003). Food standards and labeling policy
book. Retrieved November 17, 2009, from
http://www.fsis.usda.gov/oppde/larc/Policies/PolicyBook.pdf

United States Department of Agriculture Food Safety and Inspection Service. (n.d.).
Animal production claims. Retrieved September 2, 2009, from
http://www.fsis.usda.gov/OPPDE/larc/Claims/RaisingClaims.pdf.

University of Florida Office of Institutional Planning and Research. (2009). Enrollment:
Final headcount enrollment by class level, gender and ethnicity (1997-2009).
Retrieved from http://www.ir.ufl.edu/factbook/enroll.htm.

Verbeke, W., & Viaene, J. (1999). Beliefs, attitude and behavior towards fresh meat
consumption in Belgium: Empirical evidence from a consumer survey. Food
Quality and Preference, 10, 437-445.


160









Vermeir, I., & Verbeke, W. (2006). Sustainable food consumption: Exploring the
consumer "attitude-behavioral intention" gap. Journal of Agricultural and
Environmental Ethics, 19, 169-194.

Wakker, P., Kobberling, V., & Schwieren, C. (2007). Prospect theory's diminishing
sensitivity versus economics' intrinsic utility of money: How can the introduction of
the Euro be used to disentangle the two empirically. Theory and Decision, 63, 205-
231.

Wang, J., & Lee, A. Y. (2006). The role of regulatory focus in preference construction.
American Marketing Association, 43, 28-38.

Williams, D. L., & Wise, K. L. (1997). Perceptions of Iowa secondary school agricultural
education teachers and students regarding sustainable agriculture. Journal of
Agricultural Education, 38(2), 15-20.

Yale Center of Environmental Law and Policy's Environmental Attitudes and Behavior
Project. (2007). Yale environmental poll. Retrieved September 20, 2009, from
http://envirocenter.research.yale.eduluploadslepoll/YaleEnvironmentalPoll2007Ke
yfindings.pdf

Yiridoe, E. K., Bonti-Ankomah, S., & Martin, R. C. (2005). Comparison of consumer
perceptions and preference toward organic versus conventionally produced foods:
A review and update of the literature. Renewable Agriculture and Food Systems,
20(4), 193-205.

Young, C. (2008). The advertising research handbook (2nd ed). Seattle: Ideas in Flight.

Zeithaml, V. A. (1988). Consumers' perceptions of price, quality, and value: A means-
end model and synthesis of evidence. Journal of Marketing, 52(3), 2-22.

Zhao, G., & Pechmann, C. (2007). The impact of regulatory focus on adolescents'
response to antismoking advertising campaigns. Journal of Marketing Research,
44(4), 671-687.


161









BIOGRAPHICAL SKETCH

Katherine (Katie) Abrams was born and raised for 18 years in Shorewood, IL. She

received her B.S. from Purdue University and her M.S. from the University of Florida,

both in agricultural communications. Her overarching research goal is to gain an

understanding of how people make sense of and participate in debates about

agricultural and environmental issues. She often researches in the context of organic

and natural foods because it provides an intersection for examining food safety, animal

welfare, and environmental attitudes and thought processes. Her communication skills

and teaching expertise are in visual communications, Web design and usability, social

media, and public relations. In August 2010, Katie will begin her career as a visiting

assistant professor at the University of Illinois, Champaign-Urbana in the department of

advertising teaching agricultural communications.


162





PAGE 1

1 THE POWER OF FOOD LABELS: MARKETING ENVIRONMENTAL IMPACTS AND ANIMAL WELFARE ON MEAT LABELS AS GAINS VERSUS NONLOSSES AND THE INFLUENCE ON ATTITUDES AND VOTING INTENTIONS By KAT HERINE M. ABRAMS A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Kat herine M. Abrams

PAGE 3

3 To all of my boys, Brian, Wrigley, and Copper

PAGE 4

4 ACKNOWLEDGMENTS This doctoral dissertation may signal the completion of my 24 years of formal education, but the journey led by curiosity and learning will continue for the rest of my life. That journey is inspired by many people in my life who Id like th ank for their support, friendship, guidance, and love. My dissertation committee members were assets in my development as a researcher and scholarly writer. I thank Lyle Brenner for teaching me about some of the most fascinating consumer psychology theories. His class and dissertation guidance helped position me well for broader fields in communication and increased my confidence in experimental methodology. Paul Monaghan imparted on to me his impeccable insight into the complexities of conservation behavior. I hope we can work together in the future because his community based social marketing research fascinates me. Ricky Telg was a source of positive encouragement throughout my program at the Univers ity of Florida in roles as a teacher and researcher. He helped me maintain my sanity through some of the toughest dilemmas and inspired me to be a strong leader in any academic role I may take on. I have many faculty in the Department of Agricultural Education and Communication to thank. Glenn Israel was one o f the best teachers I have ever had, in his class and through his help on my research. Brian Myers was always willing to answer my methodological questions when I needed a quick answer. Ed Osborne was caring and supportive throughout my program and helped me secure my first job. Hannah Carter was a faculty member who I could be completely open and honest with and count on for great words of encouragement and advice.

PAGE 5

5 Courtney Meyers has and continues to be a positive role model, friend, and colleague in agri cultural communications. She inspired me to continue on for my PhD and to be a teacher and researcher. I could always count on her as a listener, adviser, research buddy, conference hotel roommate, and amazing friend. I envision us remaining close for our entire lives, helping each other grow professionally all the while remaining good friends. Tracy Irani has been my academic and career adviser and constant teacher throughout my graduate program. She is one of the main reasons I have been successful and wi ll continue to take on academic challenges. Her willingness to share research opportunities with me gave me an excellent learning experience that will continue throughout my career in academia. I can only hope to be as successful and well respected as she is in our field. Id like to thank Lauri Baker for her friendship and guidance as well. She helped me get my dissertation idea on paper and was always willing to answer my questions or verify my thinking. She has been a great workout buddy and friend. We n ot only work great together, but share a strong bond that will last a lifetime. Meredith Cochie, where has she been all my life? I have her to thank for an amazing friendship, laughter, adventures, gripe sessions, sarcasm, and confidence. She has a lot of faith in me, and I in her. I know we will continue to be best friends and colleagues, and maybe one day well share that backyard and academic department again. We can only hope, but god save the community! My family (Mom, Dad, and Scott) has always been supportive of my goals and life decisions. Even though I may seem very independent in a lot that I do, I will always

PAGE 6

6 need them for advice and support. I know how proud they are of my accomplishments, and they are the ones I have always been trying to impress my entire life. My newer family, Brians family, is a welcome addition in my life. I cannot thank them enough for adopting me with open arms, love, and support. Last, but certainly not l east, I tha nk my husband, Brian. Thank you for your confidence, encouragement, and pride in me throughout my graduate education. We are about to embark on a crazy journey together as a young couple in our first house, in a new place, and in new jobs. No matter what this first step leads to, I couldnt imagine doing it w ithout you. I am so lucky to have you in my life as my husband and best friend.

PAGE 7

7 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES .......................................................................................................... 10 LIST OF FIGURES ........................................................................................................ 12 LIST OF DEFINITIONS ................................................................................................. 13 ABSTRACT ................................................................................................................... 14 CHAPTER 1 INTRODUCTION .................................................................................................... 16 Types of F ood Labeling .......................................................................................... 19 Organic and Natural Foods Market ......................................................................... 21 Consumers Considerations in Purchasing Food with Credence Attributes ............ 23 Motivators of Green Food Consumerism .............................................................. 26 Personal Health ................................................................................................ 26 Environmentalism ............................................................................................. 27 Animal Welfare ................................................................................................. 28 Political Actions Affecting Meat Production ............................................................. 29 Loss Aversion ......................................................................................................... 33 Regulatory Focus Framework ................................................................................. 35 Purpose and Objectives .......................................................................................... 36 2 LITERATURE REVIEW .......................................................................................... 39 Cognitive Biases in the Construction of Choice ...................................................... 39 Prospect Theory ..................................................................................................... 41 Loss Aversion: A Bias and Construct in Prospect Theory ....................................... 44 Framing Effects ....................................................................................................... 47 Risky Choice Framing Effects .......................................................................... 49 Goal Framing Effects ........................................................................................ 49 Regulatory Focus Theory ........................................................................................ 51 Sources of Regulatory Focus ........................................................................... 52 Regulatory Focus in Consumer Decisions ........................................................ 53 Regulatory fit effect .................................................................................... 53 Moderators of regulatory focus effects ....................................................... 55 Implications for Loss Aversion .......................................................................... 56 Summary of Regulatory Focus Theory ............................................................. 57 Measuring Framing Effects through Attitudes ......................................................... 57 Context of the Theoretical Research ...................................................................... 59 Summary ................................................................................................................ 61

PAGE 8

8 3 METHODOLOG Y ................................................................................................... 66 Research Design .................................................................................................... 67 Controlling Threats to Internal and External Validity ......................................... 68 Subjects ............................................................................................................ 70 Independent Variables ............................................................................................ 72 Regulatory Focus ............................................................................................. 72 Pretesting of Message Stimuli .......................................................................... 73 De pendent Variables .............................................................................................. 75 Attitudinal Measures ......................................................................................... 75 Voting Behavior ................................................................................................ 76 Attribute Variables .................................................................................................. 76 Instrumentation ....................................................................................................... 77 Instrument Conte nt ........................................................................................... 77 Pilot Test .......................................................................................................... 79 Procedure ............................................................................................................... 80 Data Analysis .......................................................................................................... 81 4 RESULTS ............................................................................................................... 87 Descriptive Analysis ................................................................................................ 87 Demographics .................................................................................................. 87 Attention to and Purchase Frequency of Meat/Poultry with Production Claims ........................................................................................................... 89 Scale Reliabilities .................................................................................................... 90 Regulatory Focus Scales .................................................................................. 90 Attitude Scales ................................................................................................. 91 Descriptive Analysis of Variables of Interest ........................................................... 91 Regulatory Focus ............................................................................................. 91 Attitude Toward Product ................................................................................... 92 Voting Intention ................................................................................................ 92 Manipulation Checks ............................................................................................... 92 Tests of Hypotheses ............................................................................................... 93 Post Hoc Analyses .................................................................................................. 96 Attention To a nd Purchase of Products With Production Claims ...................... 97 Community Upbringing and Livestock Background .......................................... 97 Political Affiliation and Voting Intention ............................................................. 98 Other Non Significant Relationships ................................................................. 99 5 CONCLUSION ...................................................................................................... 110 Overview ............................................................................................................... 110 Key Findings ......................................................................................................... 111 Implications ........................................................................................................... 113 Theoretical ...................................................................................................... 113 Practical .......................................................................................................... 117 Limitations ............................................................................................................. 120

PAGE 9

9 Recommendations ................................................................................................ 122 For Future Research and Theory ................................................................... 122 For Practitioners ............................................................................................. 124 Conclusions .......................................................................................................... 125 APPENDIX A VERBAL PRENOTIFICATION ............................................................................. 127 B FIRST CONTACT E MAIL SENT TO SUBJECTS ................................................ 128 C FIRST AND SECOND REMINDER E MAIL SENT TO SUBJECTS ..................... 129 D THIRD REMINDER E MAIL SENT TO SUBJECTS .............................................. 130 E INSTRUMENT ...................................................................................................... 131 LIST OF REFERENCES ............................................................................................. 150 BIOGRAPHICAL SKETCH .......................................................................................... 162

PAGE 10

10 LIST OF TABLES Table page 3 1 Incomplete factorial research design .................................................................. 82 3 2 Regulatory focus questionnaire (Higgins et al., 2001) ........................................ 82 3 3 Results of nominal group assessment ................................................................ 83 3 4 Hedonic/utilitarian sources of attitude scale (Batra & Ahtola, 1991) and researcher developed measures ........................................................................ 84 3 5 Experiment treatment groups ............................................................................. 85 4 1 Attention to selected production labeling claims on meat/poultry ..................... 100 4 2 Purchase frequency of meat/poultry with select production labeling claims ..... 100 4 3 Promotion focus scale inter item consistency statistics .................................... 100 4 4 Prevention focus scale inter item consistency statistics ................................... 100 4 5 Total attitude scale inter item consistency statistics ......................................... 101 4 6 Regulatory focus questionnaire descriptive statistics ....................................... 101 4 7 Attitude toward product (product specific* + general attitude) .......................... 102 4 8 Attitude toward product grand means among treatment groups ....................... 103 4 9 Effects of labeling clai m frame on attitudes toward product with claims ........... 103 4 10 Planned comparisons t test for differences between treatment groups on attitude toward product with claims ................................................................... 103 4 11 Effects of labeling claim frame on attitudes toward product with claims ........... 104 4 12 Planned comparisons t test for differences between treatment groups on attitude toward product without claims .............................................................. 104 4 13 Independent samples t test for differences between subjects exposed to production claims and subjects exposed to general product claims ................. 104 4 14 Pearson product moment correlations between grocery shopping behavior and attitude toward product with production claims .......................................... 104 4 15 Pearson product moment correlations between grocery shopping behavior and attitude toward product without production claims ..................................... 105

PAGE 11

11 4 16 Mean attitude toward products between community upbringing ....................... 105 4 17 Effects of community upbringing on attitude toward product without claims ..... 105 4 18 Post hoc comparisons between community upbringing .................................... 105

PAGE 12

12 LIST OF FIGURES Figure page 1 1 Cover of T ime magazine on August 21, 2009 ..................................................... 38 1 2 Humane Farm Animal Cares label. .................................................................... 38 2 1 Value function as proposed by prospect theory.. ................................................ 62 2 2 Value function under prospect theory with reference to gains/nongains and losses/non losses.. ............................................................................................. 62 2 3 Risky choice framing paradigm ........................................................................... 63 2 4 Goal framing paradigm. ...................................................................................... 63 2 5 W ebsite evaluations as a function of situational prime and regulatory focus. ..... 64 2 6 Regulatory focus theory conceptual model. ........................................................ 65 3 1 Operational framework for the current study ...................................................... 86 4 1 Subjects weekly meat consumpt ion ................................................................. 106 4 2 Means between the attitudes toward product with claims in each treatment group.. .............................................................................................................. 107 4 3 Means between the attitudes toward product without claims in each treatment group. ............................................................................................... 108 4 4 Means between the attitudes toward product without claims in each treatment group. ............................................................................................... 109

PAGE 13

13 LIST OF DEFINITIONS Credence Attribute Quality a ttributes of a product that cannot be assessed by the consumer before or after use and affect products perceived quality only so much as consumers trust in the claims; for example, a label on a pork product indicat ing raised under environmentally friendly practices would be considered a credence attribute (Darbi & Karni, 1973). Factory Farm A term proliferated by animal and environmental activists to describe conventional livestock production and concentrated anim al feeding operations. Organic Agriculture A type of agriculture that promotes the use of renewable resources and management of biological cycles to enhance biological diversity, without the use of genetically modified organisms, or synthetic pesticides, herbicides, or fertilizers. Organic livestock production promotes concern for animal welfare, without the use of synthetic foodstuffs, growth hormones, or antibiotics. (Eicher, 2003) Production Claim A claim referring to how a food product was produced be fore slaughter or harvest. Promotion Focus Refers to a cognitive mechanism in which people view their goals as accomplishments, hopes, and aspirations (ideals or maximal goals), and are sensitive to the presence or absence of positive outcomes, or gains an d nongains. Prevention Focus Refers to a cognitive mechanism in which people are more concerned with safety, responsibilities, and obligations (oughts or minimal goals), and are sensitive to the absence or presence of negative outcomes, or nonlosses and losses. Regulatory Focus A cognitive style/mechanism that regulates how people attend to information.

PAGE 14

14 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE POWER OF FOOD LABELS: MARKETING ENVIRONMENTAL IMPACTS AND ANIMAL WELFARE ON MEAT LABELS AS GAINS VERSUS NONLOSSES AND THE INFLUENCE ON ATTITUDES AND VOTING INTENTIONS By Katherine M. Abrams August 2010 Chair: Tra cy Irani Major: Agricultural Education and Communication Consumers receive information about how their food is (or is not) produced on a regular basis through the labels they see in the grocery store. Production labeling claims like eco friendly, cage free and no hormones offer information about the product they are on and about the conventionally produced products that do not carry these claims. The theories of loss aversion and regulatory focus suggest that messages, such as food production clai ms, can be framed as gains or nonlosses and have different persuasive effects but the theories contradict each other This study used an experimental design with a convenience sample of 660 college students to examine how consumers attitudes toward food products are affected by gain and nonloss framed production labeling claims about animal welfare and environmental impact and whether this onpackage marketing can also affect intent to support an animal welfare ballot initiative. The results did not reveal different attitudinal effects between gainand nonloss framed production claims as predicted by loss aversion and regulatory focus theories ; however, the presence of the production claims did significantly reduce positive

PAGE 15

15 attitudes toward the product w ithout claims. Exposure to the production claims increased positive attitudes toward the product they were on, but these attitudes did not translate into intentions to support the animal welfare ballot initiative. Over 75% of the sample indicated they intended to support the policy regardless of the treatment. This study attempted to frame nonlosses and gains equivalently, but qualitatively. The results suggest that in the absence of numbers or quantifiable information, the biases of loss aversion, framing effects, and regulatory focus fit effect are minimized. Regardless of how production claims were framed, it is clear that they are a source of information affecting consumers attitudes towards conventional agriculture products and perhaps even the product ion system. Agricultural communicators should not underestimate the effects that food marketing and advertising can have on consumers attitudes toward conventional agriculture and its products, and consider these effects in addition to messages put forth by activist groups and mass media.

PAGE 16

16 CHAPTER 1 INTRODUCTION Now that I know how supermarket meat is made, I regard eating it as a somewhat risky proposition so I don't buy industrial meat (Michael Pollan, 2004, 6). What the well known author of The Omnivores Dilemma has to say about meat from conventional agricu lture that eating it is a risky proposition is a viewpoint growing in popularity (DeGregori, 2003) This stems from uncertainty about how animal agriculture production practices, such as administering subtherapeutic antibiotics, confining livestock in cr ates or cages, and concentrating animals in large numbers, might affect human health, animal welfare, and the environment (DeGregori, 2003; Hughner, McDonagh, Prothero, Shultz, & Stanton, 2007) What makes Michael Pollan different from most people is that he has visited many livestock production facilities while doing research for his books and Washington Post articles, whereas the majority of Americans have probably only seen a snapshot of animal agriculture from their car windows at 65 miles per hour. Sc ience literacy and communication research consistently shows that most people get their science informationwhich includes agricultural sciences from news and entertainment media (Nisbet et al., 2002). Many U.S. consumers are likely getting their perspecti ves on the United States agricultural system from journalists and authors like Michael Pollan, from talk shows like Oprah and Ellen and from the news media. These perspectives from mass media are not typically favorable toward conventional agriculture as evidenced in movies like Food, Inc., Fast Food Nation, and King Corn ; and in books like The Omnivores Dilemma and Chew on This A recent cover story in Time magazine entitled Getting Real About the High Price of Cheap Food (see Figure 1 1 ) represents what is typical in this type of coverage. The se massmediated channels

PAGE 17

17 portray large, conventional farms as having negative characteristics that are detrimental to human health, the environment, and animal welfare. Th e United States agricultural system has intensified over time as a result of technological and market forces, urban/suburban sprawl and a decrease in interest of farming as an occupation ( Fitzgerald, 2003; Sassenrath et al., 2008) Livestock production, in particular is highly associated with trends toward greater farm concentration and corporate industrialization, due in part to urban encroachment, government policies, and the geographic availability of feed (Lobao & Meyer, 2001; Morrison, N ehring, Banker, & Somwaru, 2004). These external pressures have led to greater human input and control of food animals lives from conception to slaughter. The intensification of production was noted by activist groups who began referring to large, conventional farms as factory farms. According to the Oxford English Dictionary, the first documented use of the term is attributed to a journal of economics in 1890, but started to proliferate in publications in the 1960s (Factory, 1989). Views of t his technologically advanced production system have changed since the 1950s. During that decade, Americans viewed products of such a production system more favorably f or the systems ability to provide convenience foods that saved the housewife time and money (Levenstein, 2003) The 1950s were a time in which consumers were focused on the re lentless pursuit of convenience (Levenstein, 2003, p. 101) This drove food pr oducers and manufacturers to develop additives to aid in processing and preserving, to develop growth additives for animal feed, and to create concentrated animal feeding operations (again, referred to as factory farms by activists) These were accepted innovations in agriculture because it meant cheaper and more convenient foods, like TV

PAGE 18

18 dinners and meat with every meal (Hooker, 1981; Levenstein, 2003). Perhaps it was natural that, in an era when Americans brimmed with confidence in the superiority of t heir political, economic, military, and even cultural institutions, they should feel similarly about their food and those who produced it ( Levenstein, 2003, p. 118). Acceptance and confidence in large, conventional agriculture production waned in the la te 1960s. In 1969 and 1970, calls for a return to natural foods resonated far from the hippie enclaves, striking sympathetic chords among the kind of thoughtful middleclass Americans (Levenstein, 2003, p. 195). In 2002, the market for organically produc ed meat and producewhich is viewed as more natural (Abrams, Meyers, & Irani, 2009) increased dramatically with the creation of the United States Department of Agriculture (USDA) certified organic label. Today, people tend to view natural foods more favorably than those produced with human or technological intervention (Rozin et al., 2004). While U.S. consumers preference for natural and organic foods continues to grow stronger, agriculture is still intensifying production practices with selective br eeding, medical and feed technologies, and advanced mechanical systems (Fitzgerald, 2003). With th ese continued changes has come increased concern about the healthiness and safety of meat, poultry, eggs and dairy for human consumption; environmental impac ts; and animals welfare in such an agricultural system These concerns have steadily risen in the U.S., as demonstrated through changes in legislation (state s banning cages and crates in hogs, layers, and veal) food labeling (certified organic label in 2 002, naturally raised claim in 2009) and growth of the market for products promoting the absence of those perceived risks (Greene et al., 2009).

PAGE 19

19 Types of Food Labeling A variety of labeling claims are used by food marketers to differentiate products and c ommunicate quality and value to consumers. Among the mix of labels are those addressing consumer concerns regarding food production practices. Some examples include the organic label, natural or all natural, free range, humanely raised, ecofriendly, no hormones, and no antibiotics. All of these are practices and labeling claims are voluntary Production labeling claims, which refer to how the food was produced pre harvest or how the animal was raised, are regulated by the USDA Food Safety and Inspection Service (FSIS). This entity develops and provides labeling guidance, policies and inspection methods and administers programs to protect consumers from misbranded and economically adulterated meat, poultry, and egg products [to] ensure that all labels are t ruthful and not misleading (USDA FSIS, 1, 2009). Processing labeling claims, which refer to how the food is altered post harvest or post slaughter (i.e., additives, preservatives, coloring), are also regulated by the USDA FSIS. Perhaps one of the more extensive set of policies and inspection programs created by the USDA is the National Organic Program. An increase in consumers interest and confusion about organic products during the 1990s led to the institution of the USDA National Organic Program in O ctober 2002 (California Institute for Rural Studies, 2005). These standards were established to assure consumers that solabeled products are produced, processed, and certified to meet the consistent national organic regulations (National Organic Program, 2002). The standards provide a set of guidelines for food to be labeled organic that affect the growing, handling, and processing of the food. For organic meat production, the standards prohibit the use of antibiotics and growth hormones, require animals to be fed 100% organic feed, and

PAGE 20

20 require animals to have access to outdoors and access to pasture for ruminants. The organic label is considered a certified label because an inspector visits the farm yearly and on an unannounced basis to certify that the farms practices are meeting the USDA organic standards The separate certification process is what makes the organic label different from most production claims. The USDA FSIS has created some claims with specific guidelines, including free range, free r oaming, natural, no hormones, and no antibiotics. Other claims producers and marketers want to put on their products which may be entirely different or variants on those claims (i.e., raised outdoors as opposed to free range) may be submitted for approval as well. Production claims are upheld by USDA FSIS policies, but are approved differently than those that are certified, like the organic label. The producer or marketer submits the claim to FSIS, along with supporting documentation (operational protocol, affidavits and testimonials), and the request to use the labeling claim is either approved or denied; however, the operation is never physically inspected before approval or denial (USDA FSIS, n.d.). When USDA inspectors conduct an annual random inspection of the entire operation for adherence to required practices, they will check to ensure the operational protocol meets what they agreed upon to qualify for the voluntary production claim. The USDA is not the only entity involved in labeling claim creations and operation inspections. Third party organizations (i.e., Certified Angus Beef and America Grassfed Certified ) that operate a product or service certification system can be approved by the USDA International Organization for Standardization Guide 65 (ISO Guide 65) Program to certify operations for meeting the third partys voluntary production or product

PAGE 21

21 standards to qualify for their label. The ISO Guide 65 Program e nsures that thirdparty certification programs are applying their standards in a consistent and reliable manner (USDA Agricultural Marketing Service, 2008). One example of such a program is Humane Farm Animal Care, which provides certification for their Certified Humane label (see Figure 12 ). Their standards are highly specific and vary by species (layer hens, broilers, dairy cows, etc.) and apply from birth through slaughter. In general, this label means the animal had ample space, shelter, and gentle handling to limit stress; no hormones or antibiotics in their feed; were not kept in cages, crates or tie stalls; and were free to engage in natural behaviors such as dust bathing for chickens and rooting for pigs (Humane Farm Animal Care, n.d.). Many consumers and researchers alike tend to lump all food with production and/or processing labeling claims into the category of organic and natural foods for ease of communication, even though the USDA has specific definitions of the terms organic and natural More recently, when researchers discuss organic food in the United States, they are referring to food produced according to the National Organic Program standards with the USDA label, but the term natural in reference to food st ill tends to be a catchall term referring to both the processing of the meat and how the livestock were raised (Abrams et al., 2009; USDA FSIS, 1999). This is important to aid in the interpretation of the research regarding the market for such foods. Organic and Natural Foods Market The $ 21.1 billion organic food industry in the United States (Greene et al., 2009) is growing due to consumer concerns about food safety, particularly regarding pesticides, antibiotics, growth hormones, and genetic modification (Hwang, Roe, & Teisl, 2005). The organic label distinguishes foods as free of those perceived risks, while other

PAGE 22

22 products attempting to appeal to these consumer concerns use production and processing claims such as all natural, no antibiotics, no hormones, and free range. These products often have a no labeling theme or communicate essentially the same message by saying the livestock or poultry were produced without one of these perceived risks (Abrams et al. 2009). In July 2009, a large market research company report indicated that organic food will be the fastest growing food trend over the next decade with a growth rate of 41% ( The NPD Group, Inc. 2009). Organic meat and poultry is one of the fastest growing segments of the organic food market and is predicted to grow 27% annually through 2010 (Storck, 2008). A nationwide survey conducted by the American Meat Institute (AMI) & Food Marketing Institute (FMI) (2008) found 19% of shoppers had purchased organic or natural meats in the past three months; however, many were not sure if the meat they purchased was organic or natural. The rapid growth of the market for these types of products suggests that a large segment of consumers have come to value production and processing attributes of the food they buy (Caswell, 1998; Thompson & Troester, 2002). These value perceptions have led to increased market diversity as more niche products emerge to fulfill consumers needs. Absent of brand differentiation, consumers were willing to accept more uncertainties and a lack of understanding about production characteristics of food products to a greater degree (Levenstein, 2003) Consumers now have a heightened awareness of food production and health factors as a result of recent food scares (e.g., 2008 beef recall due t o downer cattle, 2009 salmonella in peanuts), increased media

PAGE 23

23 coverage of these topics (Craven & Johnson, 1999), and societal shifts in values (Caswell & Mojduszka, 1996). Consumers Considerations in Purchasing Food with Credence Attributes Today a cons umer may pick up a generic package of fresh chicken and notice the one next to it advertising no antibiotics, no hormones, and free range. Consumers with different preferences, including different risk preferences, will rationally choose different bundles of attributes in foods. C onsumers will buy products that give them the most value in terms of costs and benefits as long as they are able to accurately judge the quality attributes (Caswell & Mojduszka, 1996). Quality cues at the point of purchase are mos t often extrinsic qualities, such as brand, labeling, and price. During handling (i.e., cooking, preparing) or consumption is when quality cues are intrinsic attributes (Ziethaml 1988). Consumers make assumptions about intrinsic values based on information from extrinsic cues (Olson, 1978). For example, an organic label on a bag of potato chips may generate the belief that it is probably healthier for me. Consumers often rely on extrinsic attributes in initial purchase situations when they cannot evaluate the relevant intrinsic attributes of a product or when evaluation of intrinsic cues requires more input than the consumer perceives i s worthwhile (Ziethaml, 1988). In most cases, production and processing claims are quality attributes that cannot be assessed by the consumer before or after use. Darbi and Karni (1973) call these credence attributes. Whether or not credence attributes signal high quality depends on consumers trus t of the claims. If consumers do not trust production or processing claims, then they do not signal good quality. For example, a claim on a package of beef indicating from cattle raised under environmentally friendly practices would be considered a crede nce attribute; the claim, therefore, is referred to as a credence

PAGE 24

24 attribute. Unless the consumer goes to the farm where that cow came from to determine the environmentally friendliness of the practices, he or she will have to trust the claim to derive util ity from it. Several studies have found themes of consumer distrust in credence claims (Abrams et al., 2009; Bruns Fjord, & Grunert, 2002; Padel & Foster, 2005; Yiridoe, Bonti Ankomah, & Martin, 2005). Despite some consumer distrust of the labeling clai ms c redence attributes are becoming more important in the set of considerations consumers make when trying to determine food quality because the perceived benefits outweigh the trustworthiness element The total food quality model developed by Bruns, Fjord, and Grunert (2002) states that consumers evaluate expected quality on four levels: taste, health, convenience, and process. Process refers to how the animal was raised and also how the meat was processed (presence of additives, preservatives, etc.) and it is the most relevant component of the model to this study. In the model, quality is not an aim in itself, but is desired because it helps satisfy purchase motives or values. The model therefore includes motive or value fulfillment, i.e. how food products contribute to the achievement of desired consequences and values ( Bruns et al., 2002, p. 9 10). BechLarson, Grunert, and Poulson (200 1 ) conducted a study assessing consumer choice of food products marketing different health attributes, such as Omega 3s, as simply present in the product or combined with the specific health benefits of that attribute. They found that when food products are marketed based on these credence attributes, quality perception becomes a function of communication effectiveness The effectiveness of that communication depends on the credibility consumers assign to it consumers motivation to process the information, and their ability to understand it

PAGE 25

25 (Bech Larson et al., 2001 ; Bruns et al., 2002). Thus, what the label communic ates and how it fits with the shoppers goals is critical to persuading him or her to purchase the product. Previous research has shown consumers tend to prefer organic foods, foods that were produced in a way that attenuates perceived risks, and foods that appeal to certain value sets (Loureiro, McCluskey, & Mittelhammer, 2005; Yiridoe et al., 2005); however, price is usually the primary barrier between attitudes and purchasing behavior ( Padel & Foster, 2005). If organic meat was the same price as conventi onal meat, the large majority of consumers (95.3%) would purchase it (AMI & FMI, 2008). Outside of price, convenience, and habit, when purchasing these types of meat products, several levels of outcomes are considered; among them are personal health, the environment, and animal welfare (Yiridoe et al., 2005) The latter two are ethical considerations. T he environment could also be a health consideration, depending on individual environmental values Ethical considerations, such as the confinement of livestock in crates and cages, are gaining importance but typically rank below health and meat safety risk perceptions; however, pork and poultry come up most often when consumers perceive risks to animal welfare (Verbeke & Viaene, 1999). One recent survey found consumers purchase organic and natural meat for a variety of reasons, the top three being: 1) positive long term personal health effects (47.2%), 2) better nutritional value (47.2%), and 3) better health and treatment of the animal (40.4%). The reduced environmental impact was ranked sixth with nearly 31% of consumers indicating it as a motivation to buy organic or natural meat (AMI & FMI, 2008). Surveys and polls like this do not shed light on the intri cacies of the consumer

PAGE 26

26 decisionmaking process becaus e researchers determine how the choice items are chosen and written. Looking at the response items in the AMI and FMI (2008) study, positive long term personal health effects is framed in a way that suggests this goal for eating organic and natural meat is an attempt to approach a positive outcome for their health. The response item reduced environmental impact is framed in a way that suggests the goal is to avoid a negative outcome for the environment. The way researchers choose to frame their questions and response items can affect how respondents answer them ( Levin, Schneider, & Gaeth, 1998) Motivators of Green Food Consumerism The environmental ethic that gained worldwide prominence with Earth Day 1990 placed emphasi s on individual responsibility for personal health and social action on environmental quality and animal welfare (Yiridoe et al., 2005, p. 196) In the midst of a strong environmental movement (Dunlap & Mertig, 1992; Gottlieb, 2005), a health foods craze (Dubisch, 2004; Nestle, 2007), and a powerful animal rights movement (Rollin, 1990, 2003), meat seems to represent a consumer commodity and issue through which people can demonstrate their values and goals for their health, the environment, and food animal s with little direct involvement in a movement or overt campaign. Personal Health First and foremost, food safety is the top concern fueling peoples positive attitudes toward organic and natural foods. Consumers want to be assured their food is safe and organic food is often equated with safer food. Perceptions of food safety and risk typically relate to concern about food production technologies. In the United States, concern is highest for pesticides and hormones, followed by antibiotics, genetic modif ication, and irradiation (Hwang, Roe, & Teisl, 2005). Labels and claims are used

PAGE 27

27 by marketers to appeal to those concerns. Food safety is a concern because unsafe food could potentially negatively affect a consumers personal health ( Bruns Fjord, & Grunert, 2002). In other words, personal health concerns can be a function of food safety or nutrition. Environmentalism Eighty three percent of Americans would agree that global warming is a serious problem and 81% feel it is their responsibility to reduce the impacts of global warming (Yale Center of Environmental Law and Policy's Environmental Attitudes and Behavior Project, 2007). Environmental sentiments have been on the rise, but clearly, not all Americans hold the same levels of environmentalism. Researchers have attempted to clarify different value orientations toward the environment. Kempton, Boster, and Hartley (1995) found environmental values are already intertwined with core American values, such as religion and parental responsibility (p. 13). Kempton et al. (1995) found environmentalism is built upon cultural models of how nature works and how humanity interacts with it, and is motivated by environmental values. Americans tend to idealize the environmentalism of simpler times and desire to return to that more natural way of life. Environmental values include humanitys utilitarian need for nature, obligations to future generations the spiritual or religious value of nature, and for some, the ri ghts of nature in and of itself ( Kempton et al., 1995 ). Because most Americans feel some sense of responsibility to the environment, marketers have begun to recognize the need of environmental or green marketing (Sheth & Paravatiyar, 1995; Grant, 2008).

PAGE 28

28 Green issues and marketing can work against each other. One wants you to consume less, the other more. One rejects consumerism, the other fuels it. But they arent always opposed. Marketing can help sell new lifestyle ideas. Its a much needed function t oday, when we all need to act fast to mitigate the effects of climate change (Grant, 2008, Chapter 1: 1) Purchasing meat with credence attribute marketing claims regarding the environment (i.e., environmentally friendly, good for the environment) is a rel atively simple behavior for consumers to reinforce environmental values. While consumers generally have positive attitudes toward such foods, the difficulty in persuading people to purchase them is that they often are priced at a premium and consumers are hesitant to believe their purchase will have an impact (Vermier & Verbeke, 2006). Marketing can help close the gap between attitudes and behavior if the right messages are used. Green marketing often sounds like a good idea, but if it is not based on good intentions and a deep understanding of consumer decision making, then it will not work (Grant, 2008). Animal Welfare Animals are often seen as a part of nature or at least similar to the natural environment, especially in how people view their purpose. Li ke nature, animals have some intrinsic value, but generally a utilitarian value, especially when it comes to livestock. In the United States, people desire some protection of farm animals, whether that be based on their intrinsic or utilitarian values. Munro (2005) distinguished between animal welfare and animal rights. Animal rights refers to the idea that animals, like humans, have innate rights and interests that should not be compromised for human benefit. Animal rightists do not believe humans should f arm animals at all and often promote vegetarianism and veganism (Munro, 2005). Animal welfare represents a balance between human and animal interests and refers to the idea that animals should

PAGE 29

29 not be treated cruelly or in a way detrimental to their health and well being (Munro, 2005). The notion of animal rights is often seen as too extreme for most Americans; however, most support the notion of animal welfare (Garner, 1993). Support or activism in animal welfare and animal agriculture issues (among others ) can occur at many levels, from participation in an animal protection group to private behavior such as consumption choices ( Seguin, Pelletier, & Hunsley, 1998) In the sphere of individual behavior, a consumer will likely choose a product associated with improved animal welfare or production if they somehow feel responsible and/or that their choices will make a difference (Blandford, Bureau, Fulponi, & Henson, 2002; Vermeir & Verbeke, 2006). Political Actions Affecting Meat Production If some consumers ar e not already voting with their dollar to voice support for alternative livestock production practices, they are supporting state legislation in the voting booth on initiatives advocated by the Humane Society of the United States and other well funded op ponents of conventional practices. In Florida, Arizona, and California, voters have overwhelmingly supported a policy banning common methods of animal confinement for pregnant pigs, egg laying hens, and veal calves. The animal agriculture industry tends to blame animal agriculture opponents, such as People for the Ethical Treatment of Animals ( PETA) and the Humane Society of the United States ( HSUS ) for misleading consumers, voters, policymakers, and the media on issues regarding animal welfare, the healthiness of meat products, and the environmental impacts of conventional practices ( Crowell, 2009; Downing, 2009; Gabbett, 2008; Goodwin & Rhoades, 2009; Smith, 2009). The HSUS Factory Farms

PAGE 30

30 campaign website has 31 secondary research reports on the industrys detriments to animal welfare, eight on environmental impacts, and 13 on human health that it widely distributes to policymakers and corporations (HSUS, 2008; E. Williams, personal communication, December 4, 2008) These reports are not necessarily m isleading but show that these organizations are attempting to implicate animal agriculture in detrimental ly affecting human health, the environment, and animal welfare. As with any controversial topic, each side in the debate carefully selects sources and evidence that supports their perspective on the issue. The HSUS is known for campaigning heavily for animal agriculture industry reform, using emotional appeals and more persuasive message strategies than the industry groups like Farm Bureau and the Animal Agriculture Alliance (Abrams & Meyers, 2009; Goodwin & Rhoades, 2009). Answering whether the publics support of policy initiatives on livestock care is evidence of the animal agriculture opponent groups successful campaigning or Americans evolving valu e systems regarding livestock production would be like answering the chicken or the egg conundrum. It is likely that animal agriculture opponents are more successful as a direct result of changing values and less familiarity with farming, especially livest ock production. In the midst of a strong environmental movement (Dunlap & Mertig, 1992; Gottlieb, 2005), a health foods craze (Dubisch, 2004; Nestle, 2007), and a powerful animal rights movement (Rollin, 1990, 2003), meat and livestock production seem to r epresent a consumer commodity and issue through which people can demonstrate their values and goals for their health, the environment, and food animals. The environmental ethic that gained worldwide prominence with

PAGE 31

31 Earth Day 1990 placed emphasi s on indivi dual responsibility for personal health and social action on environmental quality and animal welfare (Yiridoe et al., 2005, p. 196) Within the industry, segments and individuals regard organic agriculture as another foe of the conventional industry (Oba ch, 2007) because organic products are often touted as better in many dimensions including taste, nutritional value, and sustainability (Organic Trade Association, 2008). However, whether organic food actually delivers on these desires and beliefs is cont roversial and the subject of a scientifically inconclusive debate (Obach, 2007). A review of 162 studies conducted over 50 years found that organic food had no nutritional or health benefits over conventional food (Dangour et al., 2009). A USDA publication reviewing several studies comparing organic to nonorganic agriculture production did find that, generally (with a few exceptions) organic agriculture has several environmental advantages in a) maintaining or building soil quality, b) lessening ground a nd surface water contamination, c) reducing greenhouse gas emissions, d) encouraging biodiversity, e) conserving water and energy res ources, and, f) recycling waste (Gold, 2010, Find Out More. Issues and Ref erences, number 3, 1). Despite the scientifi c debate, c onsumers have come to believe in the superiority of organic and more naturally produced foods The Harris Poll found that more than threequarters of the U.S. public believes organic food is safer for the environment (79%) and healthier (76%) than conventional foods (Harris poll results, 2007). The price and intense marketing of organic and other valueadded animal products likely communicates to the consum er that they are indeed better than their conventional counterparts ( Klonsky & Tourte, 19 98). Higher p rice s and level s of advertising often

PAGE 32

32 trigger a placebo effect in which consumers believe those products are of higher quality, and subsequently, they have better experiences with the products than those less advertised and/or with lower prices (Shiv, Carmon, & Ariely, 2005). While some in the agriculture industry may still see organic agriculture as a detriment to the conventional industry, today, many producers and companies have embraced this niche market. This resulted in diversif ied production practices and purchas es of organic farms and brands to capture a piece of the premiums consumers are willing to pay for these products and the positive corporate reputation that comes from being attached to an initiative that is supposedly bett er for animal welfare, the environment, and human health (Guthman, 2004) Although the industry often points to animal agriculture opponent groups for the shift in peoples thinking about what is acceptable in livestock production practices in the United S tates, marketing organic and more naturally produced products as better than unlabeled ones may have unintended consequences The messages consumers receive in the grocery store week after week are likely far more memorable and pervasive than what the HSUS puts in a video on YouTube or in an ad before a vote on a ballot initiative Consumers receive multiple exposures, which are more salient than a single or few exposures to TV or Web ads/videos, to messages about meat production through package labeling cl aims in the grocery store. A 2009 Nielsen poll found 61% of consumers read food labels. Interestingly, shoppers at Whole Foods, Trader Joes, Publix, Costco, and Safeway were mostly likely to read labels (Hale, 2009). Jauregi and Ward (2006) surveyed a litt le over 14,500 households and found 57% check labels for harmful ingredients and 60% base their food purchase on using the labels.

PAGE 33

33 With less than 1% of the U.S. population involved in production agriculture (Hurt, 2002), most consumers may only learn of c ertain production inputs from reading food labels. The question becomes, are production claims on meat labels affecting what people believe about the unlabeled product? Limited research has been done to examine the effects of production labeling claims on consumers attitudes toward those that do not carry such claims. Empirical research is needed to determine the effects of production claims on consumer beliefs about the conventional meat product in the United States. Such research may shed light on politi cal actions that affect livestock production, revealing why many consumers are unwilling to pay for product attributes they perceive to be better, but are willing to support policy that would make such attributes required of all animal products. Loss Aversion Food labels are intended to be persuasive communication to convince consumers that the product is different and better than others to cause them to purchase the product (Golan et al., 2001). Many communication and psychology scholars have found numerous conditions under which persuasive communication is more effective. Persuasion, or what convinces people to change beliefs, attitudes, and behaviors, is a function of how our minds naturally want to think. Persuasive communication appeals to those cognit ive preferences. One of those cognitive preferences and perhaps the most successful and widely used explanatory construct in behavioral decision research (Brenner, Rottenstreic h, Sood, & Bilgin, 2007, p. 369) is loss aversion. Loss aversion is one of the main components of Kahneman and Tv erskys (1979) prospect theory; it shows that losses have a steeper value function than gains. In other words, losses loom larger than equivalent gains. The concept of loss aversion does not necessarily imply

PAGE 34

34 that people pay more attention to losses over gains but the hedonic reaction to a loss is stronger than the reaction to a gain (Brenner et al., 2007). Most studies examining nonloss versus gainmessage framing used quantitative descriptors (Boettcher, 2004; Idson et al., 2004; Liberman et al., 2005; Kahneman & Tversky, 1979; McDermott, 2004; Tversky & Kahneman, 1981), but limited research has tested whether the predictions of loss aversion hold for qualitatively defined frames/descriptors of equivalent gains and nonlosses. Also, a dditional research is needed to test this cognitive preference when the implications of the decision are more removed or not immediately known, such as consequences for personal health in the long term, the environment, and animal welfare. The present study intends to test the loss aversion theory in a context in which the consequences are more removed and not immediately obvious. These are perceived consequences for animal welfare, personal health, and the environment as marketed in production and processing claims on meat labels. Most consumers will never experience the impact of their purchase on those aspects first hand or be able to directly attribute potential changes to it but may be more persuaded by labeling claims that better match their goals in such purchase decisions. If they are indeed loss averse, then conclusions will reflect that loss aversion is a powerful cognitive preference that permeates consumer decisionmaking even when consequences are more removed and qualitatively descr ibed. These findings will offer suggestions for marketers of organic and natural foods, for environmental educators and communicators seeking environmentally responsible behavior change in their audiences, and for extension agents.

PAGE 35

35 Regulatory Focus Framework Regulatory focus theory (Higgins, 1998) offers another explanation of how consumers decision making and behavior operate that adds to Kahneman and Tverskys ( 1979, 1981) loss aversion concept. This theory states that whether negative information is attended to more and weighted more heavily than positive information depends on peoples goals in the decision. The theory posits that individuals function according to two dif ferent types of motivations based on mental depictions of an endstate that will result from committing to a particular decision. Individuals using a promotion focus view their goals as accomplishments, hopes, and aspirations (ideals or maximal goals), and are sensitive to the presence or absence of positive outcomes, or gains and nongains. In contrast, individuals using a prevention focus are more concerned with safety, responsibilities, and obligations (oughts or minimal goals), and are sensitive to the absence or presence of negative outcomes, or nonlosses and losses. One of the fundamental predictions of regulatory focus theory is that individuals attend to information that is relevant to the activated regulatory focus and that they weigh attributes c ompatible with this focus more carefully. This focus can be activated by persuasive messages such as advertising and product labels. Previous research has also shown some people are chronically self regulating using either the prevention focus or promotio n focus when making decisions (Higgins & Silberman, 1998). Essentially, it can be a cognitive style ; however, cues in the environment may cause a shift from one regulatory focus to another. Situational variables can cause changes in sensitivity, emotions, and strategic inclinations that can activate a promotion or prevention focus (Higgins, 1998).

PAGE 36

36 Idson, Liberman, and Higgins (2000) added to the theory when they found that the pleasure of a gain (promotion success) is stronger than the pleasure of a nonloss (prevention success), while the pain of a loss (prevention failure) is stronger than the pain of a nongain (promotion failure). This is because success in the promotion focus is achieving a maximal goal, whereas success in prevention is failure to achieve a minimal goal. This is a different perspective than the predictions of loss aversion, which explains losses loom larger than corresponding gains. Further research is needed to test how consumers attend to nonlosses versus gains as previous contradictory evidence has been found by others as well. Levin, Schneider, and Gaeth (1998) conducted a comprehensive literature review and found conflicting evidence for loss aversion in studies testing goal framing effects (gains vs. nonlossses and losses vs. nongains ) and called for more systematic research in this area. In the context of the present study, the production labeling claims on meat products often promote the avoidance of negative outcomes with a no labeling theme or by saying how the raising of the lives tock or poultry is absent of perceived risks/negative outcomes (i.e., no antibiotics, no hormones), while others may have claims focusing on the positive outcomes that will result from purchasing the product (i.e., eco friendly, great care). This experiment will test how participants react to gain versus nonloss messages via attitudinal measures in a choice situation in which other product features ( price, appearance, cut, and weight) are controlled. Purpose and Objectives Food marketers and regulators use labels and claims to differentiate products and inform consumers about their options. Government food regulators must consider the effects of food labeling to ensure the policies, standards, and guidelines for such labels

PAGE 37

37 are balancing the market for agric ultural products and not misleading consumers (Golan, Kuchler, & Mitchell, 2001). While previous research shows there is a preference for products market ed as improv ing animal care, personal health benefits and environmental impacts (also called credence attributes) no empirical research has examined the effects of marketing these credence attributes on consumers attitudes toward products without such attributes Messages about these credence attributes can be presented (framed) in different ways potentially resulting in varying persuasive effects based on biases in peoples cognitive processing. The purpose of this study wa s to compare the persuasive effects of gainand nonloss framed labeling claims The research objectives we re: Objective 1 : To compare the attitudinal effects of nonloss framed claims to gainframed labeling claims using qualitative descriptors Objective 2 : To determine whether and how consumers attitudes toward products with no claims are influenced by production labeling claims Objective 3 : To assess whether consumers voting intention on animal welfare policy is affected by credence attribute labeling claims.

PAGE 38

38 Figure 11 Cover of Time magazine on August 21, 2009. Figure 12. Humane Farm Animal Care s label.

PAGE 39

39 CHAPTER 2 LITERATURE REVIEW The previous chapter established the need for the current study and the research objectives which are to determine the effects of differently framed labeling claims on consumers attitudes toward the credence attribute product, the conventional product, and voting intention This chapter provides an overview of research related to consumer decisionmaking research with specific emphasis on cognitive biases, prospect theory, loss aversion, framing effects, and regulatory focus theory. Following this is an explanation of the studys context in sustainable agriculture food products Cognitive Biases in the Construction of Choice In explaining how people arrive at the choices they make, the rational choice theory suggests a straightforward path in which people make logical calculations based on an individually defined preference ordering system with respect to 1) the benefi ts of each alternative, 2) the costs of each alternative in terms of utilities foregone, and 3) the best way to maximize utility (Simon, 1955). Therefore, actions or choices are a function of knowledge and a cost/benefit analysis, in which costs are minimi zed and benefits maximized. However, we now know that people do not always act or think rationally. Numerous studies in decisionmaking have revealed that people can make unexpected choices based on automatic heuristics and biases and how information and m essages about choice options are framed rather than careful reasoning (for a review see Gilovich, 1991; Gilovich, Griffin, & Kahneman, 2002). While these simplifying heuristics often lead to accurate judgments, they also yield systematic error. Scholars of ten break down information processing into two parts or two paths: one involving more systematic, careful processing and the other a more superficial or

PAGE 40

40 heuristic processing of available information. One model in this area is the Heuristic Systematic Model (HSM) (Chaiken, 1987; Todorov, Chaiken, & Henderson, 2002). The HSM describes two cognitive mechanisms called systematic processing and heuristic processing. When engaging in systematic processing, people scrutinize the available information to evaluate t he validity of the message and form a judgment based on their elaborations. The systematic mode requires sufficient cognitive resources which is affected by distraction, message repetition, time pressure, communication modality, and knowledge and expertiseand motivation which is affected by personal relevance, need for cognition, task importance, accountability for ones attitudes, and exposure to unexpected message content (Todorov, Chaiken, & Henderson, 2002). When people are not sufficiently motivated or do not have the necessary cognitive resources, they use heuristic processing. This mode is more nonanalytic in nature and is characterized by the use of simple decision rules or heuristics to form a judgment. One common example is the availability heuris tic, which refers to the ease with which a circumstance comes to mind. Vivid scenarios such as shark attacks tend to be more topof mind than more objectively threatening ones like heart attacks. People may be completely unaware of their heuristic processi ng and may even deny that they were influenced by peripheral informational cues (Todorov, Chaiken, & Henderson, 2002). Underlying heuristics are cognitive biases. Biases are defined as deviations from some true or objective value or as violations of basic laws of probability (Gilovich & Griffin, 2002). Tversky and Kahneman (1983) demonstrated how human judgments depart from probability theory or simple logic in their famous Linda problem. In this experiment, subjects read the following personality description: Linda is 31 years old,

PAGE 41

41 single, outspoken, and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice and participated in anti nuclear demonstrations. They were then asked to determine which of two options was more probable: (a) Linda is a bank teller or (b) Linda is a bank teller and active in the feminist movement. Between 80% and 90% of participants tended to select (b) as the more probable option, even though probability theory dictates a conjunction cannot be more likely than either of its parts (Tversky & Kahneman, 1983). Although resulting in systematic errors, cognitive biases allow people to make decisions efficiently and can lead to correct decisions (Haselton, Nettle, & Andrews, 2005). Prospect Theory Prospect theory (Kahneman & Tversky, 1979) was one of the first influential theories offering a descriptive model of how people make decisions that differed from normative and rational choice theory. Kahneman and Tverskys (1979, 1981) studies were able to show (1) that how the outcomes of a decision are framed can affect the ultimate choice, and (2) that the decision makers evaluation under uncertainty works on a value function with three characteristics: diminishing sens itivity, reference dependence, and loss aversion. The theory sheds light on the interaction between the person and the situation in decisionmaking environments (McDermott, 2004, p. 293). Before explaining the constructs of prospect theory, a description of the original work and later tests of the theory will provide some context. In the original work by Kahneman and Tversky (1979), they presented findings indicating people treat losses differently than gains in the context of risky choice decisions invol ving monetary outcomes. They found 80% of participants preferred a certain outcome of $3000 to an 80% chance of $4000 and 20% chance of nothing. This

PAGE 42

42 showed, with respect to gains, people tend to be risk averse. When they reversed the prospects, 92% of par ticipants preferred to gamble on an 80% chance of losing $4000 and 20% chance of losing nothing to a certain loss of $3000. With respect to losses, people tend to be risk acceptant. In other words, if a person stands to gain something for certain, they are less willing to take a risk to gain something more; however, if a person stands to lose something for certain, they are more willing to take a risk to lose more (Kahneman & Tversky, 1979). In later studies advancing prospect theory, Tversky and Kahneman (1981) further examined how the framing of choices/information affects the decision made in the often cited Asian disease problem. Subjects given the positively framed version of a sure saving of onethird of the lives versus a onethird chance of saving all the lives and a two thirds chance of saving no lives chose the option with the certain outcome. Subjects given the negatively framed version of a sure loss of twothirds the lives versus a onethird chance of losing no lives and a twothirds chance of losing all of the lives chose the risky option. The outcome of saving onethird of the lives is the same as losing twothirds of the lives, but in the positively framed version, that was an acceptable choice, whereas in the negatively framed version, it w as unacceptable (Tversky & Kahneman, 1981). Boettcher (2004) found support for prospect theory when individuals evaluated a political decision in the context of a terrorist hostage situation at a U.S. embassy. In the gain frame, hostages were rescued or not rescued, and in the loss frame, hostages died or did not die. However, when subjects came together in a group to make a decision, the preference reversal did not occur; therefore, support was not found for prospect

PAGE 43

43 theory in group decision making. In a group context, people were more risk acceptant regardless of the frame (Boettcher, 2004; McDermott, 2004). Numerous studies, however, have indicated support for prospect theory in a wide variety of domains including politics, business, finance, management, and medicine (see Maule & Villejoubert, 2007, for a review). The implications of prospect theory for the present study suggest individuals may consider messages about credence attributes of meat products differently depending on how the information is presented to them. A persons attitudes toward the product may be stronger if the message about environmental impacts, for example, is framed as avoiding a loss than if it is framed as achieving a gain for the environment. Prospect theory (depicted in Figure 21 ) proposed an explanatory model of choice that deviated from rational choice theory in four important ways, resulting in the main constructs of the theory: diminishing sensitivity, reference dependence, loss aversion, and framing effects. What follows is a brief explanation of the constructs relevant to this study and a deeper explanation of the last two, which will be tested. Kahneman and Tversky proposed that people make choices regarding gains and losses in terms of deviations from a reference point, whic h is usually the status quo, but can be deviations from an aspiration level or some other reference point (Heath, Larrick, & Wu, 1999). Soman (2004) explained that values are coded as gains and losses relative to a reference point, meaning the decision is reference dependant Looking at the Asian disease problem in Tversky and Kahnemans 1981 study, in one condition, the outcomes are framed in terms of saving lives; the potential disaster of losing all the lives becomes the neutral reference point (Soman, 2004, p. 383).

PAGE 44

44 The frame, therefore, can change the perceived reference point of the question. A frame refers to how information is described and interpreted (Bottecher, 2004). Credence attribute messages on meat labeling can suggest a reference point of personal health deterioration (when framed as nonloss), thereby causing people to want to avoid the potential loss. These framing effects demonstrated the cognitive bias of loss aversion, in which the notion of a certain loss is more aversive, causing people to accept a risk (with poor odds to gain) to potentially avoid that loss. Following, is an indepth explanation of loss aversion and framing effects, which are the most relevant to the present study. Loss Aversion: A Bias and Construct in Prospect Theory Loss aversion is perhaps the most successful and widely used explanatory construct in behavioral decision research (Brenner et al., 2007, p. 369). As one of the main components of Kahneman and Tv erskys (1979, 1981) prospect theory, it shows that los ses have a steeper value function than gains. Examining Figure 21, losses and nonlosses are measured against the steep loss part of the value curve, whereas gains and nongains are measured against the shallow part of the value curve. In other words, los ses loom larger than equivalent gains. Loss aversion was originally proposed as an explanation for the endowment effect, which is what explains peoples tendency to place a higher value on an item that they own than on an identical one that they do not own (Kahneman, Knetsch, & Thaler, 1990). The concept of loss aversion does not necessarily imply that people pay more attention to losses over gains but the hedonic reaction to a loss is stronger than the reaction to a gain (Brenner et al., 2007). Hedonics re fer to the basic human emotions of pleasurable and unpleasurable states of consciousness (Kahneman, Diener, &

PAGE 45

45 Schwartz, 2003). Imagine the reaction of an elementary school student that loses a gold star from their publicly displayed achievement tally. It i s likely much more of a reaction than when they received that gold star. Just as happy marriages can be easily knocked off line, but it takes an enormous amount of time, effort, and commitment to repair a marriage that is breaking apart, and in many instances even that cannot fix what has already broken (McDermott, 2004, p. 298). This cognitive bias is why gas stations advertise a lower price for paying cash or using their credit card, rather than an advertised surcharge for using a credit or debit card; people are more willing to forego gains than to accept losses (Thaler, 1980). Although seemingly irrational in the context of business and market transactions, it has roots in lower level psychological laws that seem adaptive to basic environmental demand s. Thus the asymmetry of peoples reactions to pain versus pleasure is eminently sensible in a world that punishes those who ignore danger signs more than it rewards those who pursue signs of pleasure (Newell, Lagando, & Shanks, 2007, p. 119). Many studies have found support for loss aversion (see Levin et al., 1998 for a review). One of the more recent studies testing loss aversion found a greater hedonic reaction to losses than to nongains (supportive of loss aversion) using three different experimental scenarios, including the following one regarding labor union contract negotiations: Gain/nongain condition: one of the conditions listed in the proposal is an increase in employee benefits of approximately $200. Loss/nonloss condition: one of the con ditions listed in the proposal is a decrease in employee benefits of approximately $200. Participants in the gain and loss (nongain and nonloss) conditions answered the question How would you feel if the condition is written (is removed and not writte n) into the new contract? on a scale ranging from good). (Liberman, Idson, & Higgins, 2005, p. 531)

PAGE 46

46 The authors found that gains were perceived as more intensely positive than nonlosses, a result that is opposite to the prediction derived from loss aversion. The figure used to depict prospect theory and loss aversion in their study includes a depiction of nongains and nonlosses (see Figure 22 ), which makes this alternative finding more visible. Similar findings were presented in Idson et al. (2000). Both studies offered regulatory focus theory as a possible explanation for the findings, although the authors made no mention of measuring regulatory focus. Regulatory focus theory explains that how people ev aluate and react to gains and losses is controlled by how they envision the goal or outcome of the decision (Higgins, 1998). The mechanism that moderates this is called the regulatory focus. These findings suggest an area for continued study in other contexts before conclusions can be made regarding the explanatory strength of loss aversion with respect to gains versus nonlosses. In summary, l oss aversion theory states that people have stronger reactions to potential losses than potential gains. The theory predicts the same should hold true for nonlosses (avoiding a loss) relative to gains (achieving a gain); however, some scholars (Idson et al., 2000; Liberman et al., 2005) have found inconsistencies with this prediction. More research is needed to compare the persuasive effects of gain versus nonloss framed messages. Also, most studies examining nonloss versus gainmessage framing have used quantitative descriptors (Boettcher, 2004; Idson et al., 2004; Liberman et al., 2005; Kahneman & Tversky, 1979; McD ermott, 2004; Tversky & Kahneman, 1981). Additional research is needed to test whether the predictions of loss aversion hold for qualitatively defined frames/descriptors of equivalent gains and nonlosses.

PAGE 47

47 Framing Effects What Kahneman and Tversky (1979, 1981) and other previously mentioned researchers have found is that people will arrive at different decisions depending on how the choice information is framed. Framing, at a basic level, refers to the process through which individuals or groups make sense o f their environment; frames are cultural structures that organize understanding of social phenomena. Packets of incoming information pass through various cognitive, affective, and/or social filters to produce a perception of the outside world. This cons truction of reality then drives judgment and decision making and ultimately behavior (Bottecher, 2004, p. 332333). Although this may be an internal process, it is often constructed by some external actor either deliberately or unintentionally (Bottecher 2004). The focus of framing effects in psychology and marketing is slightly different from media framing and political issue framing. Agricultural communication and general communication researchers often use a different conception of framing, so it is important to understand the differences in how framing effects can be operationalized and studied. The psychology literatures definition of a framing effect is when two logically equivalent (but not transparently equivalent) statements of a problem lead decision makers to choose different options (Rabin, 1998, p. 36; Tversky & Kahneman 1981). This is called equivalency framing. In media and issue framing effects literature, an alternative explanation is used. Druckman (2001) described media or issue fram ing effects as when a speakers emphasis on a subset of potentially relevant considerations causes individuals to focus on these considerations when constructing their opinions (p. 1042). In media and issue framing effects studies, the frames are rarely logically equivalent. They are often qualitatively different yet potentially relevant

PAGE 48

48 considerations (Druckman, 2004, p. 672). Issue and media framing involve the selection of some aspects of a situation making them more salient through communicating text with the idea of advocating a particular solution or interpretation of the topic (Entman, 1993). For example, the debate on universal health care is typically framed by one side as health care is a basic human right and by the other side as not the g overnments or taxpayers responsibility. This study deals exclusively with equivalency framing and its effects on decision makers, meaning two different, but logically equivalent frames are used. Levin and Gaeth (1988) offer a good example. They found v ariation in quality preferences regarding beef depending on whether a beef product was labeled as being 75% lean or 25% fat. The ground beef was evaluated by subjects as better tasting and less greasy when it was labeled in the positive light (75%) lean. T he common adage that pessimists see the glass half empty, and optimists see it half full, demonstrates a complimentary description of the same object that is viewed in two different ways. Objectively, a glass half empty is a glass half full, but people wil l make different decisions about that object depending on how it is presented to them. Throughout the literature, valence framing effects wherein the frame casts the same critical information in either a positive or a negative light, are often treated as a relatively homogeneous set of phenomena explained solely by prospect theory (Levin et al., 1998, p. 150; emphasis in original). Levin et al. (1998) organized and interpreted past literature on framing effects to explain contradictory and weak support o f prospect theory, thereby demonstrating the existence of different types of framing effects with different underlying mechanisms and consequences.

PAGE 49

49 Risky Choice Framing Effects Framing, as defined by Tversky and Kahneman (1981), is a the decisionmaker s conception of the acts, outcomes, and contingencies associated with a particular choice (p. 453). This means the choice involves their perceptions of the courses of action, the outcomes associated with the alternative, and the likelihood associated with the outcomes. To study framing effects in the vein of prospect theory, one would set up an experiment as outlined in Figure 23; however, many recent studies of framing effects have deviated greatly from the operational definitions and theoretical concepts used in the original studies (Levin et al., 1998, p. 151). The framing effects studied in prospect theory are what they call the risky choice framing paradigm ( Figure 23 ) in which the outcomes of a potential choice involving options of differing risk levels are described/framed in different ways. In this type of framing, risk preference is affected as seen in Kahneman and Tverskys (1979, 1981) original studies. Overall, the evidence from multiple studies on framing effects in the risky choice paradigm show a relatively consistent tendency for people to be more risk acceptant when the options are framed to focus attention on the chance to avoid losses than when options foc us on the chance to realize gains (Levin et al., 1998). Goal Framing Effects Goals themselves can govern or frame what people attend to, what knowledge and attitudes become cognitively most accessible, how people evaluate various aspects of the situati on, and what alternatives are being considered (Lindenberg & Steg, 2007, p. 119). Goal framing effects refer to the impact of persuasion depending on how a consequence or implied goal of a behavior is framed (see Figure 24 ). What is different about goal framing is that both frames should enhance the evaluation of the issue. It is a

PAGE 50

50 matter of determining which type of goal to avoid a loss or achieve a gain/benefit (Levin et al., 1998). Levin, Gaeth, Evangelista, Albaum, and Schrei ber (2001) directly tested goal framing effects in the context of reducing red meat consumption with American and Australian subjects. The manipulation was: Positive frame condition: If you discontinue eating red meat you will be able to reduce the level of cholesterol in your blood. Thus, you will significantly decrease the likelihood of the early onset of heart disease. Negative frame condition : If you continue eating red meat you will not be able to reduce the level of cholesterol in your blood. Thus, you will fail to significantly decrease the likelihood of the early onset of heart disease. Participants in each condition were then asked to write a number between 0 and 100 to indi cate how likely they are to eliminate red meat from their diet, and to write a number between 0 and 100 to indicate how likely they are to reduce by at least 1/3 the amount of red meat in their diet. (Levin et al., 2001, p. 66) American subjects rated the complete elimination of red meat and the reduction of red meat significantly higher in the negative frame condition; however, this effect was not significant for Australian subjects. Several other studies have generally found a similar loss aversion bias i n which avoiding a loss is greater than the desire to obtain a gain of So for example, a meat product with credence attribute claims framed as avoiding loss or damage to personal health, animal welfare, and the environment may create a different attitudinal response in comparison to claims framed as achieving gains or repairing personal health, animal welfare, and the environment. Later research found this loss aversion bias in goal framing effects can be mitigated by age (Shamaskin, 2009), involvement (Maheswaran & Meyers Levy, 1990; Miller & Miller, 2000), and culture (Levin et al., 2001) such that that those higher in age and those higher in involvement are more influenced by positive frames.

PAGE 51

51 Levin et al. (1998) found the evidence for goal framing is les s homogenous than for risky choice and attribute framing and called for more research in this area of framing effects. Furthermore, the findings explained earlier from Idson et al. (2000) and Liberman et al. (2005) revealed inconsistencies with loss aversi on with respect to gains versus nonlosses. These scholars and others have developed theory to explain a cognitive style underlying how people process information and their goals called regulatory focus. Regulatory Focus Theory Regulatory focus theory (H iggins, 1998) offers another explanation of how consumers decision making and behavior operate that adds to Kahneman and Tverskys (19 79, 1981) loss aversion concept. This theory states that whether negative information is attended to more and weighted more heavily than positive information depends on peoples goals in the decision, which is controlled by the individuals regulatory focus. The theory posits that individuals function according to two different types of motivations based on mental depictions of an endstate that will result from committing to a particular decision. Those with a prevention focus will regulate their behaviors away from negative outcomes, while those with a promotion focus will regulate their behaviors toward positive outcomes ( Higgins, 1998). Individuals using a promotion focus view their goals as accomplishments, hopes, and aspirations (ideals or maximal goals), and are sensitive to the presence or absence of positive outcomes, or gains and nongains. When the endstate is des ired/positive, individuals are said to have an approach goal. Approach goals are achieved by maximizing the presence or minimizing the absence of positive outcomes (Higgins, 1998; Aaker & Lee, 2001). For example, an environmentally conscious consumer may

PAGE 52

52 w ish to improve the environment (desired end state) and purchase meat with credence attributes (strategy that maximizes the presence of a positive outcome). In contrast, individuals using a prevention focus are more concerned with safety, responsibilities, and obligations (oughts or minimal goals), and are sensitive to the absence or presence of negative outcomes, or nonlosses and losses. When the endstate is undesired/negative, individuals are said to have an avoidance goal. Avoidance goals are achieved by minimizing the presence or maximizing the absence of negative outcomes. For example, an environmentally conscious consumer may wish to avoid damaging the environment (undesired endstate) by purchasing meat with credence attributes (strategy that minimiz es the presence of a negative outcome). The examples used with meat purchasing demonstrate that people can envision their goals for the environment slightly differently (improve or repair vs. avoid damage) but still use the same strategy (purchase meat wit h credence attributes) to attain the goal. Products, like meat with credence attributes, can be regarded as a means to approaching a positive outcome or avoiding a negative one (Florack, Scarabis, & Gosejohann, 2005, p. 240). Sources of Regulatory Focus Regulatory focus is affected by three sources: chronic regulatory focus of the decision maker, contextual priming during or before the decision task, and the decision task itself (Florack et al., 2005). A chronic regulatory focus is determined by caretaker child interactions. A childs behavior regulated by positive reinforcement increases their sensitivity to promotion goals, whereas negative reinforcement increases their sensitivity to prevention goals (Higgins & Silberman, 1998 ; Higgins, 1998 ) Like oth er motivational orientations, regulatory focus may vary between individuals not only dispositionally, but also momentarily and independently of the chronic focus (Florack et

PAGE 53

53 al., 2005, p. 237). Situational variables can cause changes in sensitivity, emoti ons, and strategic inclinations that can activate a promotion or prevention focus (Higgins, 1998). This is shown in studies through which subjects are primed to adopt a promotion or prevention focus. The priming is typically done through having individuals complete a thought listing activity. Freitas and Higgins (2002) offer the following induction script: Promotion: Please think about something you ideally would like to do. In other words, think about a hope or aspiration that you currently have. Please list the hope or aspiration below. Prevention: Please think about something you think you ought to do. In other wo rds, think about a duty or obligation that you currently have. Please list the duty or obligation below (Freitas & Higgins, 2002, p. 3). A decision task or consumer good can also be associated with a certain regulatory focus. Zhou and Pham (2004) found par ticipants who made preventionrelated investment decisions were more likely to have adopted a prevention focus as indirectly measured through participants choosing a product with prevention claims. In examining regulatory fit effec ts, Florack and Scarabis (2006) discovered that sunscreen is a product category that prompts a prevention focus. Limited research has been done to determine what other decision tasks and consumer goods can be associated with a particular regulatory focus. Regulatory Focus in Cons umer Decisions Regulatory focus theory has been tested in a number of consumer decision making contexts to determine the robustness of the theory and continue its expansion. What follows is a review of the key studies in this area. Regulatory fit effect O ne of the fundamental predictions of regulatory focus theory is that individuals attend to and have stronger hedonic reactions to information that is relevant to the

PAGE 54

54 activated regulatory focus and that they weigh attributes compatible with this focus more carefully (Higgins, 2002). This is called the regulatory fit effect. Florack and Scarabis (2006) found a relationship between an advertising claim and consumers regulatory focus had an impact on product preferences. Their study compared participants that were primed to adopt a promotion focus or prevention focus, and their preferences for sunscreen, based on the packaging claims on the bottle. One set of claims emphasized a promotion focus (Enjoy the sun) while the other emphasized a prevention focus (G ive sunburn no chance). Participants who were asked to think about negative (positive) things they try to avoid (pursue) while on vacation showed a stronger preference for a brand with a preventionfocused claim (promotionfocused claim). The product was a means to achieve the goals of approaching a positive outcome (getting a tan) or avoiding a negative outcome (sunburn) (Florack & Scarabis, 2006). As mentioned in the discussion on sources of regulatory focus, they found sunscreen is a product category that prompts the prevention focus. Their manipulation checks, however, did show successful priming. One explanation for the regulatory fit effect is that the fit of the message with a persons regulatory focus leads to enhanced persuasion because individual s evaluate messages more positively when they are in line with their attitudes, motivations, and needs (Florack et al., 2005, p. 244). The perception of fit may be used as a heuristic and lead to biased message processing. Regulatory fit evokes a feeling of importance or feeling right which gets interpreted as a positive evaluation (Higgins, 2002). Furthermore, it increases the recipients engagement with the message and is perceived as more persuasive (Cesario, Higgins, & Scholer, 2008). If the consumers

PAGE 55

55 attention to meat with credence attributes prompts a particular regulatory focus, then the messages that fit that focus would be the most appealing. Moderators of regulatory focus effects Wang and Lee (2006) looked at how regulatory focus theory affect s consumers evaluations of products to determine if advertising promotion and prevention messages simultaneously would enhance or diminish persuasion. In addition, they examined the effects of involvement on the regulatory fit effect. They found subjects in the low involvement condition place more weight on features that fit their regulatory focus when reviewing both fit and nonfit product feature claims. Timing subjects evaluations of the product feature claims showed those primed with a prevention focus spent more time looking at the prevention claims, while those with a promotion focus spent more time on promotion claims. Again, this only occurred in the low involvement condition. In the high involvement condition, subjects spent about the same amount of time on both types of claims. They did show that their evaluation of the products was driven more by their perceived attractiveness of the features than by the extent of processing. That is important because one could have argued that their preference f or the product with feature claims that fit their regulatory focus was a function of (mediated by) time spent processing that information. In sum, people rely on their regulatory focus as a filter to process information selectively to construct their pref erences when cognitive resources are limited (Wang & Lee, 2006, p. 36). The research on regulatory fit effect demonstrates that it is more reflective of heuristic versus systematic processing. Evan and Pettys (2003) finding that the regulatory fit effect is moderated by need for cognition is consistent with Wang and Lees (2006) findings regarding involvement.

PAGE 56

56 In a study aptly titled I seek pleasures and we avoid pains: The role of self regulatory goals in information processing and persuasion, Aaker and Lee (2001) found individuals who viewed themselves as independent were more persuaded by promotionfocused product information, whereas, those who viewed themselves as interdependent were more persuaded by preventionfocused product information. Subjec ts evaluated messages on the Welchs Grape Juice website. The independent variable of self view was manipulated in two ways: (1) picture focusing on an individual or family, and (2) text that emphasized the individual (you, your) or the interdependent self (family). The independent variable of regulatory focus was manipulated through product claims about Welchs Grape Juice. The promotion focus emphasized messages its ability to increase energy. The prevention focus messages emphasized its ability to prevent cancer and heart disease (Aaker & Lee, 2001). Figure 2 5 illustrates their findings that promotion information appeals more to the independent self view. Aaker and Lees (2001) findings were important because they demonstrated that accessible self view moderates the persuasiveness of promotion/preventionfocused messages that could easily be manipulated through advertising and a strategy to approach different audience segments (families vs. individuals). Implications for Loss Aversion Other researchers (Idson et al., 2000; Idson, Liberman, & Higgins, 2004; Liberman et al. 2005) added to the theory when they found that the pleasure of a gain (promotion success) is stronger than the pleasure of a nonloss (prevention succes s), while the pain of a loss (prevention failure) is stronger than the pain of a nongain (promotion failure) (see p. 44 of this document for a complete description of the experiment).

PAGE 57

57 Regulatory focus theory predicts that because promotion success (gain) is success in achieving a maximal goal (a standard one hopes to achieve), it should be experienced more intensely than prevention success (nonloss), which is success in achieving a minimal goal (a standard one must achieve) (Liberman et al., 2005, p. 269) The pleasure of a gain being stronger than the pleasure of a nonloss is a different perspective than the predictions of loss aversion, which explains losses loom larger than corresponding gains. Idson et al. (2000), however, did not directly examine the predictions derived from loss aversion, where as Idson et al. (2004) and Liberman et al. (2005) did. They suggested that more caution is needed in using [loss aversion] to explain decision making phenomena in economics, political science, and social psyc hology, particularly when examining gains versus nonlosses (Liberman et al., 2005, p. 534). Furthermore, they called for more research to determine whether regulatory focus effects overwhelm the predicted loss aversion effect for gains versus nonlosses, which is what the present study intends to address. Summary of Regulatory Focus Theory The studies presented on regulatory focus theory demonstrate that decision makers evaluate information that fits their focus more favorably than information that does not. People tend to elaborate, better understand, and feel right when presented with information in tune with their regulatory focus (Cessario et al., 2008). R egulatory focus theory offers implications for the design of messages as evidenced in Aaker and Lees (2001) self view study and Liberman et al.s (2005) study of gains versus nonlosses. Figure 26 provides a conceptual model of regulatory focus theory. Measuring Framing Effects through Attitudes Studies in agricultural economics often examine different ways of labeling food to determine peoples preferences (Hu, Woods, & B astin, 2009). They determine

PAGE 58

58 preference by measuring peoples willingness to pay (WTP) for a product with certain attributes. The concept of a preference is, in some ways, the counterpart in economics to the concept of an attitude in psychology, but the l ogic of attitudes and the logic of preferences are quite different (Kahneman & Sugden, 2005, p. 164). Preferences are subjective, but their logical structure is objective. If a consumer prefers a ground beef product that is 25% fat, they should prefer a p roduct that is 75% lean. Attitudes are not objective in structure; therefore, a consumer might have a negative attitude toward a grou nd beef product that is 25% fat but a positive attitude toward one that is 75% lean. The occurrence of framing effects does not violate the logic of attitudes as it does the logic of preference (Kahneman & Sugden, 2005). Preferences are best measured by making people choose between two options, while attitudes are best measured by affective responses to a single object. Attitu des have a reasonable amount of stability. This stability of attitudes lends some stability to the choices that people make, but attitudes are also susceptible to a lot of manipulations that are not allowed to have any effect in a rational theory of preferences (Kahneman & Sugden, 2005, p. 165). Therefore, a framing effect should yield a change in attitude. An attitude is defined as an association between an object of thought and a valence evaluation with three components: cognitive, emotional, and behavi oral ( Ostrom, Bond, Krosnick, & Sedikides 1994). Cognitive responses are based on beliefs, inferences, knowledge, and assumptions about the attitude object. Emotions are the feelings connected to thinking or experiencing an attitude object. Behavior is the action or actions taken in response to the attitude object (Ostrom et al., 1994). Similarly to Ostrom et al. (1994), Batra and Ahtola (1991) state that consumer attitudes have

PAGE 59

59 distinct hedonic and utilitarian components (p. 168). The hedonic component refers to affective/emotional gratification from consumption behavior. The utilitarian component refers to the instrumental, practical reasons. Attitude, therefore, can be measured through utilitarian and hedonic descriptors Context of the Theoretical R esearch The term sustainable agriculture is often used to incorporate the dimensions of personal health (food safety), the environment, and animal welfare. It is difficult to define because both conventional and organic agriculture attempt to frame their practices as sustainable. Definitions of sustainable agriculture vary widely. A basic, conservative definition is: The primary goals of sustainable agriculture include: (1) providing a more profitable farm income; (2) promoting environmental stewardship, including protecting and improving soil quality, reducing dependence on nonrenewable resources, such as fuel and synthetic fertilizers and pesticides, and minimizing adverse impacts on safety, wildlife, water quality and other environmental resources; (3) promoting stable, prosperous farm families and communities (Sustainable Agriculture Research & Education [SARE], n.d. 3) It is also defined as a way of raising food that is healthy for consumers and animals, does not harm the environment, is humane for workers, respects animals, provides a fair wage to the farmer, and supports and enhances rural communities (Sustainable Table, n.d., 1). Even those using conventional agricultural practices could argue that they are sustainable whether they ascribe t o either definition. These two definitions may lead one to conclude that the devil is in the details and sustainability is in the eye of the beholder. Regardless, most people have strong, pleasurable associations with the idea of sustainable agriculture (Williams & Wise, 1997); therefore, products marketed on dimensions of sustainability benefit from those associations.

PAGE 60

60 The problem with the marketing of these food products is that it could suggest the unlabeled or conventionally produced foods are inferior and from unsustainable agricultural systems. The United States government frames the organic label as a marketing label, and rejects the idea that organic food production would have relative advantages to the environment, health or food quality ( Bostr m & Klintman, 2003). The organic label and production claims are not meant to differentiate the food as safer, but unintentionally, they may have. Government regulations have typically been used to distinguish between safe and unsafe foods; therefore, organic standards could give consumers the impression that conventionally produced foods are unsafe (Klonsky & Tourte, 1998). In addition, t he price and intense marketing of organic and other valueadded animal products likely communicates to the consum er that they are indeed better than their conventional counterparts ( Klonsky & Tourte, 1998). Higher p rice s and level s of advertising often trigger a placebo effect in which consumers believe those products are of higher quality, and subsequently, they have better experiences with the products than those less advertised and/or with lower prices (Shiv et al., 2005). Food regulators need to have an understanding of how production claims labeling affects consumers beliefs about meat in order to balance the market for such products and avoid misleading consumers. R esearch investigating whether consumers attitudes toward conventionally produced products are affected by production claims and how these attitudes might translate into behavioral intent (e.g., intent to su pport an animal welfare ballot initiative) has yet to be done. Such research may also shed light on political actions that affect livestock production, revealing why many consumers are unwilling to

PAGE 61

61 pay for product attributes they perceive to be better, but are willing to support policy that would make such attributes required of all animal products. The attitudes toward products from sustainable agricultural systems are typically positive. Consumers reasons for preferring meat products from such agricultural systems are: (1) health and nutritional benefits, (2) improved animal welfare, and (3) decreased environmental impact (AMI & FMI 2008, Yiridoe et al., 2005). W hat is unknown is how consumers envision these three goals: are they trying to approach posi tive outcomes, or avoid negative outcomes? Testing the theories of loss aversion, and regulatory focus in the context of food labeling offer s implications for marketing sustainable agricultural products and the market for all agricultural products Further more, this application context can be used to test whether regulatory focus effects can overwhelm the predictions of loss aversion, and how the predictions of loss aversion hold when communicating gains and nonlosses qualitatively Summary The review of the literature outlined in this chapter provided an overview of biased information processing with an emphasis on loss aversion, framing effects, and regulatory focus theory. Current gaps in the literature illustrate the need to further explore message framing effects of gains versus nonlosses communicated qualitatively and the effects of credence attribute labeling on consumers attitudes toward products without such claims and voting intention on an animal welfare ballot initiative.

PAGE 62

62 Fi gure 21. Value function as proposed by prospect theory. Obtained from Jacob and Ehret (2006). Figure 22. Value function under prospect theory with reference to gains/nongains and losses/non losses. Obtained from Liberman et al. (2005).

PAGE 63

63 Figure 23. Risky choice framing paradigm (Levin et al., 1998) Figure 24. Goal framing paradigm (Levin et al., 1998).

PAGE 64

64 Figure 25. W ebsite evaluations as a function of situational prime and regulatory focus (Aaker & Lee, 2001).

PAGE 65

65 Figure 26. Regulatory focus theory conceptual model.

PAGE 66

66 CHAPTER 3 METHODOLOGY The purpose of this study wa s to compare the persuasive effects of gainand nonloss framed labeling claims The objectives of this study were to determine the effects of differently framed labeling claims o n consumers attitudes toward the credence attribute product, the conventional product, and voting intention. The theories of loss aversion (Tversky & Kahneman, 1981) and goal framing effects (Levin et al., 2001) predict that losses and potential losses garner a stronger hedonic reaction than gains; therefore, avoiding a loss should yield a stronger response than achieving a gain. Although two studies have suggested gains are reacted to more strongly than nonlosses (Idson et al., 2005; Liberman et al., 2005) and offer the regulatory focus theory as an explanation, the literature testing and supporting the predictions of loss aversion is far more extensive. However, to ensure regulatory focus is not affecting the attitudinal response, subjects chronic regulatory focus will be measured and controlled statistically. Subsequently, the following hypothes i s is offered : H1: When controlling for regulatory focus, subjects exposed to nonloss framed claims will have more positive attitudes toward the product with pro duction claims than those exposed to gain framed labeling claims or control group claims The literature has suggested the intense marketing of sustainable agriculture products or food products with production claims could communicate that the unlabeled or conventionally produced foods are inferior and from unsustainable agricultural systems (Klonsky & Tourte, 1998Therefore, in examining the effects of production claims on attitudes toward conventional products that do not have production cl aim labeling and subsequent voting behavior on an animal welfare ballot initiative, the following hypotheses are:

PAGE 67

67 H 2 : When controlling for regulatory focus, subjects exposed to nonloss framed claims will have less positive attitudes toward the product without production claims than those exposed to gainframed labeling claims or control group claims. H 3 : Subjects exposed to a food product with production claims and a product without such claims will have less positive attitudes toward the product without th e claims than those who do not see a food product with production claims. H 4 : Subjects exposed to a food product with production claims will be more likely to have intentions to vote yes for an animal welfare ballot initiative than those who do not see a food product with production claims. The experiment focuse d on determining how an individuals attitudes toward meat products are influenced by differently framed credence claims and how voting behavior on animal welfare initiatives is influenced. A subjects regulatory focus (prevention, promotion) and the two independent variables (presence or absence of production labeling claims, frame of production labeling claims) should lead to different effects on attitudes toward the product with the c laims and toward the product without the claims and the subjects intent to support an animal welfare ballot initiative. See Figure 31 Research Design This study used a 2 ( production claims: present and not present) x 3 (claim frame: nonloss, gain, and neutral) betweensubjects incomplete factorial design. This design was chosen to determine (1) the effects of gain framed claims and nonloss framed labeling claims regarding animal welf are and environmental impact on attitude toward the product and (2) the effects of production claims on attitudes toward products without production labeling claims and voting intention. Factorial design s allow the determination of the effect of two manipulated independent variables on the dependent variables and the interaction among the variables The design of th e stud y is depicted in Table 3 1 and was implemented as follows: R= random assignment, X= treatment (independent variable), O= dependent variable

PAGE 68

68 XA 1= Exposure to production claims and product without production claims XA 2= No exposure to production claims (control) XB 1= Environmental and animal welfare nonloss framed claims XB 2= Environmental and animal welfare gainframed claims XB 3= Neutral framed general product claims (control) O1= posttest measure of attitude toward product with claims O2= posttest measure of attitude toward product without claims O3= posttest measure of voting on animal welfare ballot initiative An incomplete factorial design is used when some combinations of values of factors are nonsensical or not of theoretical interest ( Shadish, Cook, & Campbell 2002). The cells of XA 2B1 and XA 2B2 are considered nonsensical because when the production claims are not present, they clearly cannot also have a frame. XA2 and XB3 serve as the control levels for each factor. The XA 1B3 cell is of no theoretical interest. Controlling Threats to Internal and External Validity The research design account ed for a number of threats to internal and external validity. The threat of selection to internal validity was controlled by using random assignment to conditions using a random number generator in addition to measuring some antecedent and intervening variables that could be controlled for statistically if necessary. Attrition/mortality was not a major concern in this design, however, extensive pretesting helped determine the ease and length of time it takes to complete the experiment aided in preventing attrition and fatigue. The threat of instrumentation (the instrument changing from personto person) wa s a concern in this study given the reliance on technology to administer the treatment and collect data. The online survey tool was extensively pretested on various computers, Internet browsers, and operating

PAGE 69

69 systems to protect against this threat Furthermore, subjects were asked if they could view the images depicting the treatment and automatically skipped the dependent variable measurements if they were not able to view the photos. Finally, other extraneous variables related to the treatment that would potentially interfere with the internal validity were controlled across conditions, including the chicken product, lab el design, brand, and price. The labels were designed by the researcher, printed on label paper, and placed directly on a package of boneless, skinless chicken breasts. The labels were swapped for the different treatment groups on the same package of chicken, which was photographed in a controlled studio environment by a professional photographer to ensure reliability between the treatment groups. Construct validity threats were controlled through pretesting, pilot testing, and manipulation checks in the ex periment and also by ensuring the constructs were well defined and measured using multiple questions (which controls for nonmeasure bias). Monooperation bias was controlled by using two different treatments (gain and nonloss frame) and a control group that did not receive the treatment. Suspicion/hypothesis guessing was controlled for by not telling participants that they are participating in an experiment. Instead, they were told upfront that it is a survey. This also controlled for compensatory rivalry, since they will not know there may be different surveys (treatment conditions). Administering the experiment online and by not using leading language or leading questions also control l ed for the experimental expectancies threat. The interactions of other treatments on the outcomes were controlled for by measuring some of those potential interactions (previous purchasing behavior, label attention) and using random assignment to conditions

PAGE 70

70 Subjects The convenience sample for this study included students fro m four courses at a large southeastern university : a research and business writing class (N= 192) a public speaking class (N= 176) an introduction to journalism class (N= 138) and an intro duction to mass media class (N= 234) Subjects were offered cours e extra credit to incentivize participation. When the online experiment was sent out, 7 4 0 students were enrolled in these classes. T he courses contained students from a variety of colleges and majors and at varying phases in their program (freshman, sophomore, juniors, and seniors). Students enrolled in more than one of these courses were accounted for and only allowed to participate in the study once; ho wever, students enrolled in more than one of the classes used in the sample were given extra credit in all of them by taking the questionnaire once. Convenience sampling is often used in psychology research, usually with easily accessible college students (Peterson, 2001) This method involves choosing a sample based on what is convenient to access. Cognitive psychologists argue that when examining cognitive mechanisms, like memory attention, or biases, college students are an acceptable sample because th ey will maintain the same neural networks. Making generalizations about consumer behavior from college students may be more difficult. Peterson (2001) reviewed experimental studies using different subject samples and found differences in the direction and magnitude of effect sizes between student and nonstudent samples. He advised that making generalizations about consumer behavior from college student samples to nonstudent samples should be done with caution. However, when examining a theoretically inter esting causal relationship (strictly theory testing), the focus may need to be more on internal validity than external, and,

PAGE 71

71 therefore, using a college student conveni ence sample is appropriate (Kam, Wilking, & Zechmeister, 2007). The nature of the study i s to examine cognitive mechanisms (framing effects, loss aversion, and regulatory focus) that have shown prevalence in multiple nonstudent samples (Druckman, 2001) as well as student samples (Liberman et al., 2005; Tversky & Kahneman, 1981) The theoretical contribution being whether exposure to production claims affects attitudes toward conventional products without such claims and how the frame affects attitudes toward the production attribute product While generalizations cannot be made to all consumers from this convenience sample, providing data about the sample characteristics can aid in external validity conclusions because readers can determine how similar other populations of consideration may be to the sample used in this study It is unlikely this group will have much previous exposure to or knowledge about these kinds of labels; therefore, perhaps this sample is ideal to make a theoretical contribution. In the case of marketing valueadded meat products, another key issue is identifying users or potential users for the product category. Young adults, and specifically college students, are one segment of consumers for food products. Currently, there are over 15.9 million college students in the United States, representing a $9.2 billion market that is viewed by packaged goods marketers as "a meaningful segment" on its own, with distinct characteristics, brand loyalties, and preferences for consumable goods, including food (Ness, Gorton, & Kuznesof, 2002, p. 506). As a segment, traditional 18 to 24year old college students have been shown to differ from their similar aged nonstudent peers, in that they are much more likely to live away from

PAGE 72

72 home, and thus are able to establish an independent lifestyle, including the need to develop life skills such as food shopping and meal preparation (Mintel, 1999). Students may even spend more on food as a percentage of their total living expenses compared with other consumers ( Ness, Gorton, & Kuznesof, 2002). T hey are also more likely to be aware of diet and heal th issues as compared with the population as a whole (Ness, Gorton, & Kuznesof, 2002), which makes them a relevant target for marketing new food products and technologies. Also, r esearch shows that young consumers (1832), and those with a college educatio n are more likely to purchase organic food products (Onyango, Hallman, & Bellows, 2007). Most researchers who use college student samples do so because of cost and convenience factors, and the practice must therefore be viewed as a limitation of the study. In the present study, however, college students were also used because they represent a group of consumer prospects whose attitudes have long been tracked by industry for their ability to influence and predict mainstream consumer trends, and this predicti ve value is particularly significant for attitudes toward meat products Independent Variables R egulatory F ocus Regulatory focus theory suggests that a cognitive mechanism regulates how individuals attend to loss/nonloss and gains/nongains (Higgins, 1998); therefore, subjects regulatory focus was measured to control for these effects. The regulatory focus questionnaire (RFQ) (Higgins et al., 2001) contains 11 items (see Table 32 ) with two subscales Subjects were given the RFQ before the treatment was administered to prevent regulatory focus priming from the framed production claims.

PAGE 73

73 Prete sting of Message Stimuli To determine the labeling claims that would be used as the treatment a threestep process was used. First, the researcher and an assistant visited six grocery stores (three regional chain supermarkets, one national superstore, one local grocer, and one natural and organic foods retailer chain) and recorded all unique meat labeling claims addressing health, animal welfare, and environmental impacts. When the lists were collapsed, 33 unique claims resulted. This first step was taken to improve the studys external validity by using real labeling claims. The second step involved pretesting these 33 claims with three focus groups (tw o faceto face and one online) with a total of 20 college student participants (7 in the first, 8 in the second, 5 in the third). The claims were assessed for basic understanding, claim type (health, animal welfare, or environmental) and perceived frame (gain or nonloss). Participants used nominal group assessment to categorize the claims first as individual s and then discussed discrepancies and claim clarity as a group to resolve differences. Because no clear animal welfare claim with a nonloss frame and no clear equivalent ly framed health claim s emerged from the nominal group assessments, a third step was needed. The third step consisted of pretesting the labeling claims using an online survey with different samples of college students. Ultimately, two online surveys were conducted: the first tested the original 33 claims with the addition of the claim No cages, and the second tested the original 33 claims, No cages, and No fat. The first online survey was done with 23 college students, who were not part of the original sample for the nominal groups. The second one also had 23 college students, who wer e not a part of either of the samples already used. Between the nominal groups

PAGE 74

74 and two online surveys, 66 college students participated in the pretesting of the labeling claims to determine type and frame. The results are presented in Table 33 The claims were chosen based on 1) a Chi square analysis of the combined data, and 2) whether they were equivalent frames. Unfor tunately, no clear equivalently framed (nonloss and gain) health claims were found, and subsequently, the health claim was eliminated from the study. Most studies examining nonloss versus gainmessage framing used quantitative descriptors (Boettcher, 2004; Idson et al., 2004; Liberman et al., 2005; Kahneman & Tversky, 1979; McDermott, 2004; Tversky & Kahneman, 1981) but this study used qualitative descriptors to improve external validity, meet the applied research objectives and further test the limits of loss aversion. The environmental gainframed claim chosen was Good for the environment, and the nonloss framed claim chosen was No negative environmental impacts. These two claims are qualitatively equivalent in that a product produced in a way that does not have negative environmental impacts is good for the environment. In the same line of logic, a product produced in a way that is good for the environment does not have negative environmental impacts. The animal welfare gainframed claim chosen was No cages, and the nonloss framed claim chosen was Free to roam. These two claims are qualitative ly equivalent in that animals raised in a production system with no cages would be free to roam, and animals free to roam are not in cages. The claims were printed on a label, placed on a package of boneless, skinless chicken breasts, and photographed. C hi cken was chosen to ensure reliability of the study because it is a uniform product with little to no differences of product characteristics that are able to be visually detected. In addition, chicken is a product

PAGE 75

75 consumers choose primarily based on color w ith no consideration for marbling or other visual quality cues (Becker, Benner, & Glitsch, 2000). Chicken ranks number one in total meat consumption in the United States (USDA Economic Research Service, 2007). Demerritt (2004) found that organic poultry is a gateway organic food and an important frontline product for the organic industry (as cited in Oberholtzer, Greene, & Lopez, 2005). Because of the standardization of this product, chicken is ideal for experimental purposes to ensure participants are maki ng their decision based on the claim and not on physical quality characteristics. Results from this study are transferable to other meat products with more distinguishing characteristics, like beef or lamb. The same package of chicken was used for both treatment conditions to control for any quality differences. Price, cut, weight, and brand were also held consistent between the conditions. The ballot initiative regarding animal welfare used the same language from Californias Proposition 2 that passed in N ovember 2008. The same language has been consistently used by HSUS in states like Illinois, Ohio, Michigan, Arizona, and Florida as it either attempts or successfully proposes to ban certain animal confinement practices (HSUS, 2009). It read: C alves raised for veal, egg laying hens, and pregnant pigs can be confined only in ways that allow these animals to lie down, stand up, fully extend their limbs, and turn around freely Under the measure, any person who violates this law would be guilty of a misdemean or, punishable by a fine of up to $1,000 and/or imprisonment in county jail for up to six months. ( Prop 2: Standards for confining farm animals 2008) Dependent Variables Attitudinal M easures After viewing the product with claims and product without claims simultaneously subjects attitudes toward each product were measured The scale developed by Batra

PAGE 76

76 and Ahtola (1991) measures the hedonic and utilitarian sources of consumer attitudes using eight semantic differential questions. The scale reliabilities exceeded a Chonbachs alpha of .89 (Crowley, Spangenberg, & Hughes, 1992). The utilitarian components were measured using Batra and Ahtolas (1991) scale by the fivepoint semantic differential items of useful/useless, valuable/worthless, beneficial/harmful and wise/foolish. The hedonic component was measured by the items pleasant/unpleasant, nice/awful, agreeable/disagreeable, and happy/sad. Overall attitudes were measured by using items of good/bad, positive/negative, like/dislike, and favorable/unfavorable. Again, all of these items were measured on a fivepoint semantic differential scale (see Table 34 ). Voting B ehavior A fter completing the attitudinal measures, subjects voting intention was assessed to test H4 To measure voting intention a ballot was presented with the following proposition: On the next ballot in your state, the following initiative regarding the confinement of livestock is being proposed: C alves raised for veal, egg laying hens, and pregnant pigs can be confined only in ways that allow these animals to lie down, stand up, fully extend their limbs, and turn around freely Under the measure, any person who violates this law would be guilty of a misdemeanor, punishable by a fine of up to $1,000 and/or imprisonment in county jail for up to six months. How do you plan to vote? Yes No Attribute Variables To improve the generalizability of the findings, several attribute variables were included in the measures P roviding data about the sample characteristics can aid in

PAGE 77

77 external validity conclusions because readers can determine how similar other populations of consideration may be to the sample used in this study (Ary, Jacobs, & Razavieh, 2002) These variables included chronic regulatory focus, age, gender, organic food purchas ing behavior, attention to meat labeling claims, and personal/family ties to agriculture. Instrumentation The instrumentation for this study was implemented using an online questionnaire tool. Experiments administered onlin e offer several advantages including higher statistical power from larger sample sizes, savings in time, resources, space, and manpower reduction of experimenter bias, and ease of access for the subjects (Reips, 2000) This study followed recommendations from Reips (2000) for conducting experiments online, which included extensive pretesting, using the subjects first name in contacts with them, and sending several reminders The online questionnaire tool provided the abilit y to collect the data completely electronically using photos to represent the treatment and email to administer it Selection bias was not a concern in this study because most college students are comfortable using the Web and tend to be early adopters of new Internet technologies (Jones, JohnsonYale, Millermaier, & SeoanePerez, 2009). Instrument Content Three different online questionnaires were created, one for each treatment group. The only differences between all three questionnaires were the images of chicken packages with different labeling claims. One of the manipulation check questions was different between the two treatment groups and the control. The treatment groups were asked which product was better for animal welfare and the environment, while the

PAGE 78

78 control group was asked which product offered more information. Subjects responded to 65 questions in total. The instrument (Appendix C) was developed as a series of pages to minimize scrolling and was comprised of the following elements: Page 1 included the informed consent. Page 2 included the prompt to type in their unique participant identification number, which was used to avoid collecting names in the data set but still be able to provide the students the extra credit incentive. Page 3 included a checkbox question for the subject to indicate what courses they were enrolled in. This was to ensure that students enrolled in more than one of the courses included in the sample received the extra credit and only took the survey once. Page 4 was a transition page briefly explaining the first set of questions that would be asked. Page 5 included seven items from the RFQ that had the same response items. Page 6 included the last two items from the RFQ that had the same response items. Pag e 7 was a transition page briefly explaining the second set of questions that would be asked. Page 8 displayed the two photos of the packages of chicken and asked whether they could view the images. This was done to ensure the subjects Internet browser wa s displaying the images. If they answered no, the survey tool automatically skipped the treatment pages and questions and took them to the demographics questions to prevent subjects from answering the questions without the ability to view the treatment. Page 9 displayed the two photos again, this time with the 16 semantic differential questions that make up the attitudinal index to measure subjects attitudes toward the product with the claims. The question was I feel that Product A is Page 10 displayed the two photos again, with the attitudinal index to measure subjects attitudes toward the product without the claims. The question was I feel that Product B is Page 11 was a transition page briefly explaining the next screen would contain a potential state law regarding the confinement of livestock to vote on. Page 12 stated On the next ballot in your state, the following initiative regarding the confinement of livestock is being proposed. Following that text, was the language

PAGE 79

79 from the California Pr oposition 2 that appeared on the state ballot in 2008. They were then asked, How do you plan to vote? Page 13 was a transition page explaining the next set of questions would be about demographics and grocery shopping decisions. Page 14 included four dem ographic questions to assess age, gender, community of origin, and connections with livestock production. Page 15 asked Approximately how often do you eat meat (including all meals and snacks) in a typical week? Page 16 included two questions to assess p olitical party affiliation and views. Page 17 asked whether they currently shopped for groceries for themselves or household. If the subject indicated no, it sent them to page 18. If they indicated yes, it sent them to page 19. Page 18 was only seen if the subject indicated they did not do the grocery shopping. This page asked if they helped make the decisions about the food that would be bought. If they indicated yes, it sent them to page 19. If they indicated no, it skipped them to page 21. Page 19 included five yes/no questions regarding their attention to organic and production labeling on poultry or meat products. Page 20 included five questions to assess how often they purchased organic and credencelabeled products. Page 21 was a transition page explaining the final set of questions would be related to their experience taking the survey. Page 22 included one manipulation check question to see if they noticed the difference in the labeling between the two products. Page 23 included one manipulat ion check question to see if they noticed one was intended to be labeled as better for the environment and animal welfare. Page 24 was the thank you and debriefing page explaining the purpose of the study and that the voting situation was only hypothetical Pilot Test The procedure was pilot tested with 30 undergraduate students. After the pilot subjects completed the experimental procedure, they were asked a series of qualitative questions to determine fatigue, technical, and understanding issues. Using SPSS 16.0

PAGE 80

80 for Windows, item analysis statistics were run to determine the construct validity of each of the scales measuring concepts of interest. In the social sciences, a reliability coefficient of 7 0 or larger indicates an index is adequate (Tr aub, 1994). This data analysis indicated one necessary change to the instrument before final distribution. The 16 item attitudinal scale had an alpha reliability of .97 overall, with a reliability of .97 on the 12item index developed by Batra and Ahtola ( 1991) subscale and .88 on the 4item researcher developed subscale. DeVellis (2003) states that, ideally, the Chronbach alpha coefficient of a scale should be above .7; therefore, all items were kept in this scale. The 11item regulatory focus questionnair e contains two sub scales: a 5 item prevention scale and a 6item promotion scale. The prevention scale had an alpha reliability of .77, while the promotion scale had an alpha reliability of .67. Removing the item I have found very few hobbies or activities in my life that capture my interest or motivate me to put effort in them was deleted resulting in a final reliability of .72 for the promotion index. Procedure Subjects (N= 740) were randomly assigned (with the use of a random number generator) to eith er the nonloss framed claims condition, the gainframed claims condition, or the control claims condition to test the hypotheses. The claims and treatment conditions are shown in Table 35 In the gain frame and nonloss frame condit ions, subjects simultaneously viewed a package of chicken with two production claims (animal welfare and environmental impact), cut, weight, and price on the label and a package of chicken with only cut, weight, and price on the label (referred to hereafter as the product without production claims). In the control condition, subjects

PAGE 81

81 simultaneously viewed a product without production claims and a product with general product claims (boneless and skinless, and chicken breasts). Subjects were verbally given a pre notice from their course instructor two to three days before the initial contact e mail with directions and the link were sent out (see Appendix A). The first contact was sent to students as a personalized email (Dear [First name]) to better establis h a connection with the subjects ( Reips, 2000) with a brief explanation of the survey, their unique participant identification number, and a unique link matched with their randomly assigned treatment group (see Appendix B). Subjects had 10 days to complete the questionnaire. Three reminder emails were sent to subjects who had not yet responded: the first was sent four days after the first contact the second three days after that, and the third was sent the morning of the f inal day (see Appendices B 1 and B 2). Data Analysis Data analysis for this study was completed using SPSS 16.0 for Windows PC. Cronbachs coefficient alpha was used as an internal consistency measure of reliability. This measure is used with Likert type questions when a score can take on a range of values (Ary et al., 2002). One way ANOVAs and t tests were used to address the first three hypotheses and a Chi square test for independence was used to address the fourth hypothesis.

PAGE 82

82 Table 31. Incomp lete factorial research design R XA 1B1 O1, O2, O3 R XA 1B2 O1, O2, O3 R XA 2B3 O1, O2, O3 Table 32. Regulatory focus questionnaire (Higgins et al., 2001) Never Seldom Sometimes Often Very often Compared to most people, are you typically unable to get what you want out of life? 1 2 3 4 5 Growing up, would you ever cross the line by doing things your parents would not tolerate? 1 2 3 4 5 How often have you accomplished things that got you psyched to work even harder? 1 2 3 4 5 Did you get on your parents nerves often when you were growing up? 1 2 3 4 5 How often did you obey rules and regulations that were established by your parents? 1 2 3 4 5 Growing up, did you ever act in ways that your parents thought were objectionable? 1 2 3 4 5 Do you often do well at different things that you try? 1 2 3 4 5 Not being careful enough has gotten me into trouble at times. 1 2 3 4 5 Certainly false Somewhat false Neither true nor false Somewhat true Certainly true When it comes to achieving things that are important to me, I find that I dont perform as well as I ideally would like to. 1 2 3 4 5 I feel like I have made progress toward being successful in my life. 1 2 3 4 5 I have found very few hobbies or activities that capture my interest or motivate me to put effort into them. 1 2 3 4 5

PAGE 83

83 Table 33. Results of nominal group assessment Claim Claim Type Frame Health Environment Animal Unclear Nonloss Gain Unclear Earth f riendly 0 66* 0 0 22 39 5 Raised without antibiotics 35 2 19 12 43 21 2 Great care 19 2 31 14 10 32 4 Family farmers in harmony with nature 6 44 14 2 25 38 5 Good for you 66* 0 0 0 3 62* 1 Minimally processed 52* 6 5 3 44 18 4 Lean 61* 0 4 1 15 48* 3 Eco friendly 0 66* 0 0 19 46* 1 Fed v egetarian d iet 21 1 35 9 24 38 4 No hormones administered 26 11 25 13 41 23 2 N o artificial ingredients 65* 0 0 1 48* 18 0 G ood for the environment 0 66* 0 0 9 57* 0 N o animal or poultry products in feed 17 1 41 7 47* 17 2 No cows injected w / rgbh 20 0 38 8 46* 19 1 N o environmental contamination 6 57* 0 3 44 19 3 N o negative environmental impacts 0 66* 0 0 48* 19 0 Raised to reduce environmental impacts 1 61* 1 3 15 51* 0 No antibiotics 41 1 15 9 52* 14 0 No antibiotics administered 33 0 21 11 52* 13 1 Produced without antibiotics, synthetic hormones, or pesticides 37 3 9 17 32 22 12 No hormones 30 0 15 20 37 14 0 Raised without hormones 23 0 29 12 37 28 1 Cage free 2 0 64* 0 22 44 1 No growth stimulants 26 0 24 13 50* 15 1

PAGE 84

84 Table 3 3. Continued Claim Claim Type Frame Health Environment Animal Unclear Nonloss Gain Unclear No preservatives 62* 1 1 2 44 22 0 No fillers 53* 0 9 4 44 22 0 No steroids 35 0 19 12 50* 14 2 Meets humane society standards 5 1 60* 0 26 39 0 No chemical medicines 27 1 23 15 39 12 0 Humanely raised 0 0 66* 0 16 50* 0 No nitrates 41 17 2 6 49* 17 0 No sodium 65* 0 1 0 40 25 1 Free to roam 0 0 66* 0 16 50* 0 No cages (n= 46) 1 1 44* 0 47* 22 0 No fat (n= 23) 23* 0 0 0 6 17* 0 Note : Labels in italics were added later and only tested with online survey. Chisquare analysis p < .05. Table 34. Hedonic/utilitarian sources of attitude scale (Batra & Ahtola, 1991) and researcher developed measures I feel that the product is Useless 1 2 3 4 5 Useful Worthless 1 2 3 4 5 Valuable Harmful 1 2 3 4 5 Beneficial Foolish 1 2 3 4 5 Wise Unpleasant 1 2 3 4 5 Pleasant Awful 1 2 3 4 5 Nice Disagreeable 1 2 3 4 5 Agreeable Sad 1 2 3 4 5 Happy Bad 1 2 3 4 5 Good Negative 1 2 3 4 5 Positive Dislike 1 2 3 4 5 Like

PAGE 85

85 Table 3 4. Continued I feel that the product is Unfavorable 1 2 3 4 5 Favorable Unhealthy 1 2 3 4 5 Healthy Unsafe to eat when cooked 1 2 3 4 5 Safe to eat when cooked From an animal treated inhumanely 1 2 3 4 5 From an animal treated humanely Bad for the environment 1 2 3 4 5 Good for the environment Note : Italicized text indicates researcher developed item. Table 35. Experiment treatment groups Claim Type Nonloss Frame Condition Gain Frame Condition Control Condition Environmental Impact No negative environmental impacts Good for the environment X Animal Welfare No cages Free to roam Control X X Boneless and skinless Chicken breasts

PAGE 86

86 Covariate Chronic Regulatory Focus (prevention or promotion) Message Frame/Treatment Nonloss/No claims Gain/No claims Neutral/ No claims (control) Framing Effect Attitude toward product with credence claims (More Positive) Exposure to production claim effect Attitude toward product w /o production claims (More Positive) Exposure to production claim effect Voting behavior on animal welfare ballot initiative (Support) Framing Effect Attitude toward product with claims (Less Positive) Exposure to production labeling claim effect Attitude toward product without production claims (Less Positive) Exposure to production claim effect Voting behavior on animal welfare ballot initiative (Does not support) Figure 31. Operational framework for the current study

PAGE 87

87 CHAPTER 4 RESULTS With loss aversion and regulatory focus theory as the theoretical framework the objectives of this study were to determine the effects of differently framed labeling claims on consumers attitudes toward the product with production claims the conventional product, and voting intention. The treatment and assessments were delivered online through the use of an online questionnaire tool to college students in four different large university courses. The three independent variables were the presence/absence of production labeling claims, claim frame (nonlos s or gain), and regulatory focus (promotion, prevention). The two dependent variables were attitudes toward the product and voting intention on an animal welfare ballot initiative. This chapter provides an analysis of the data beginning with sample demographics, followed by analysis of the variables of interest. Next is a discussion of the scale reliabilities used to develop the indexes that measure the independent and dependent variables, followed by an overview of manipulation checks. The chapter concludes with a discussion of the tests of hypotheses used in the study. Descriptive Analysis Using an online survey tool for development and administration, the 65item questionnaire was administered to 740 college students from four different large university c ourses. The overall response rate was 89.2% (n= 660). Demographics The demographic characteristics included in the instrument were: age, gender, political party affiliation, political views, ruralurban background, connection with the livestock industry, m eat consumption frequency, and attention to and purchase

PAGE 88

88 frequency of meat and/or poultry with five types of production labeling claims. Subjects who did not grocery shop for themselves or their household, n or help make the food purchasing decisions skipped the questions about attention to and purchase frequency of meat and/or poultry products with production label ing claims Descriptive analysis indicated 459 of subjects were female (69.5%) and 201 were male (30.5%). The undergraduate student population f rom which the sample was chosen contains more females (55%) than males (45%) (University of Florida Office of Institutional Planning and Research, 2009). The age range of the respondents was 18 to 33 years old, with a mean of 21 years old ( SD = 1.69). The m ajority of subjects described the community in which they grew up in as a subdivision in a city or town (n= 491, 74.4%), followed by rural, not a farm (n= 98, 14.8%), downtown in a city or town (n= 47, 7.1%), and farm (n= 23, 3.5%). Most subjects indicated that neither they nor their immediate family work in livestock production (n= 563, 85.3%). Subjects political party affiliation was fairly evenly distributed among the response items with most identifying themselves as Independent leaning Democrat (n= 1 05, 15.9%), followed by Democrat (n= 97, 14.7 %), and Republican (n= 85, 12.9%). When the response items were collapsed, 265 were Democrat (40.2%), 228 were Republican (34.5%), and 78 were Independent (11.8%). Subjects political views leaned slightly more toward Liberal than Conservative. Most indicated they were somewhat Liberal (n= 116, 17.6%), followed closely by Liberal (n= 115, 17.4%), while 90 considered themselves somewhat Conservative (13.6%) and 98 Conservative (14.8%). When the response items wer e collapsed, 277 were Liberal (42.0%), 222 were Conservative (33.6%), and 107 were neither (16.2%).

PAGE 89

89 The majority of subjects consumed meat on a regular basis with most eating it 4 7 times per week (n= 258, 39.1%) and 8 14 times per week (n= 216, 32.7%). O nly 27 (4.1%) indicated that they never eat meat, and 14 (2.1%) indicated they eat it less than once per week (see Figure 41 ). Attention to and Purchase Frequency of Meat/Poultry with Production Claims Before assessing attention to and purchase frequency of meat and poultry products with production labeling claims, subjects were asked if they do the grocery shopping or help make the decisions for food purchases. The majority of subjects (n= 601, 91.1%) do the grocery shopping for themselves or their household and 41 (6.2%) help make the decisions as to what food to purchase. Only 17 (2.6%) indicated they do not purchase nor help make the decisions; therefore, they automatically skipped over the production label attention and purchase frequency questions. Subjects were asked whether they pay attention to five different types of production labeling claims: 1) organic labels, 2) labels that address the way the animal was raised, 3) labels that say no hormones, 4) labels that say no antibiotics, and 5) labels that suggest the product is better for the environment (green). The sample was fairly evenly split (with the exception of claims addressing the way the animal was raised) between yes and no, with slightly more indicating no on all five of the labeling claim types. Table 41 displays the results in entirety. When asked how often they purchase meat or poultry products with these five production labeling claims, most indicated they purchase them never or less than once a month. The means were all less than 2, with the way the animal was raised having the lowest purchase frequency ( M = 1.13, SD = 1.44) and no hormones having the highest purchase frequency ( M = 1.55, SD = 1.71). See Table 42 for the complete results.

PAGE 90

90 Scale Reliabilities Within the loss aversion and regulatory focus framework, three scales were used in this study to measure independent and dependent variables. The regulatory focus questionnaire contained two scales: one to assess promotion focus and the other to assess prevention focus. This was an independent variable in the study. Attitude, a dependent variable, was measured using a scale containing 16 items, which consisted of two subscales. Regulatory Focus Scales Regulatory focus was measured using a questionnaire developed by Higgins et al. (2001). The total questionnaire contained 11items, five prevention items and six promotion items. They are considered t wo separate scales. In the pilot study one of the promotion items was eliminated to improve the reliability alpha, thereby making the scale consist of five items. Subjects indicated their responses using a 5point Likert scale. Each scales total score co uld be 25 at the strongest regulatory focus down to five at the weakest regulatory focus. The promotion scale had a range of standard deviations from .82 to .93, demonstrating a minimal amount of variance in the data. The correct ed item total correlations on the prevention scale ranged from .43 to .67 ( Table 43 ). The alpha reliability coefficient for the entire index was = .79 and would not be improved by removing any item. DeVellis (2003) states that, ideally, the Chronbach alpha coefficient of a scale should be above .7. The grand mean for the promotion scale was 17.88 ( SD = 3.22). The prevention scale had a range of standard deviations from .70 to 1.12 demonstrating some variance in the data. The correction item total correlations on the

PAGE 91

91 prevention scale ranged from .43 to .67 ( Table 44 ). The alpha reliability coefficient for the entire index was = .68 and could not be improved by removing any item. This is slightly lower than what DeVellis (2003) recommen ds, and is therefore, a limitation of the study. The grand mean for the prevention scale was 19.48 ( SD = 2.73). Attitude Scales Attitude was treated as one dependent variable that included product specific attitude and general attitude because the overall r eliability of the scale was strong The total attitude scale with all 16 items had standard deviations ranging from .94 to 1.15 on attitude toward the product without claims and .85 to 1.09 on attitude toward the product with the claims. The alpha reliabil ity coefficient for the entire index was = .96 on attitude toward product without the claims and = .96 on attitude toward product with the claims. Neither would be significantly improved by removing any item ( Table 45 ). Descriptive Analysis of Variables of Interest Regulatory Focus Subjects regulatory focus was measured using Higgins et al. (2001) Regulatory Focus Questionnaire. One item (question 11) from this index was deleted based on the scale reliabilities in the pi lot testing data. This questionnaire contains two scales: one to assess promotion focus and the other to assess prevention focus. The lowest score on each scale was seven and the highest was 25. The mean promotion score was 19.48 ( SD = 2.73) and the mean pr evention score was 17.88 ( SD = 3.22), indicating that this sample tended to be more promotion focused. Table 46 shows the results for each item.

PAGE 92

92 Attitude Toward Product The attitude score ranged from 1 (most negative) to 3 (neutral) to 5 (most positive). The grand mean on attitude toward the products without the claims was 3.53 ( SD = .84) The grand mean attitude toward the products with the claims was higher ( M = 4.04, SD = .74) Overall, attitude toward the product with the claims was more positive than attitude toward the product without the claims. Table 47 displays the results for each item and Table 4 8 displays the results for attitude between treatment groups. Voting Intention Voting intention was measured using a oneitem measure. Subjects were asked how they would vote on an animal confinement law in their state. The language of the proposed initiative was the same that appeared on Californias 2008 ballot for Proposition 2. The question was posed as follows: On the next ballot in your state, the following initiative regarding the confinement of livestock is being proposed: C alves raised for veal, egg laying hens, and pregnant pigs can be confined only in ways that allow these animals to lie down, stand up, fully extend their limbs, and turn around freely Under the measure, any person who violates this law would be guilty of a misdemeanor, punishable by a fine of up to $1,000 and/or imprisonment in county jail for up to six months. How do you plan to vote? Yes No Subjects were not provided any additional information. Most subjects indicated they plan to vote yes for this law (n= 510, 77.3%), while only 150 (22.7%) indicated no. Manipulation Checks To evaluate the labeling claim stimuli used in the treatment, two manipulation checks were conducted. The labels for the two products in each condition were

PAGE 93

93 designed identically, with the exception of the presence of th e claims. The first manipulation check was designed to determine if subjects noticed the difference between the labels on the packages of chicken. In the gainframe condition, 97.5% (n= 196) noticed the differences between the two labels. In the nonloss fr ame condition, 98.3% (n= 227) noticed the differences. In the control condition, 93.4% (n= 197) noticed the differences. A oneway betweengroups analysis of variance with the Welch correction showed that the differences between the groups was not signific ant F (2, 660) = 2.17, p = .12. The second manipulation check was designed to determine if subjects recognized that the product with the claims was better for animal welfare and the environment. In the control condition, subjects were instead asked which product contained more information. In the nonloss frame condition, 98.3% (n= 232) identified the product with claims as better for animal welfare and the environment. In the gaincondition, 96.6% (n= 201) identified the product with claims as better for animal welfare and the environment. In the control condition, a different question was asked since general product claims were used. In this condition, 96.3% (n= 208) identified the product with the claims as the one containing more information. A oneway between groups analysis of variance with the Welch correction showed that the differences between the groups was not significant F (2, 640) = 1.13, p = .33. Tests of Hypotheses To determine if the theoretical covariate was influencing the dependent variables the relationships between the regulatory focus scores (promotion, prevention) and attitude toward the products were investigated using Pearson product moment correlation. Preliminary analyses were performed to ensure no violation of the

PAGE 94

94 assumptions of normality, linearity, and homoscedasticity. There was a small positive correlation between promotion score and attitude, r = .08, n = 660, p = .04, with a greater promotion focus associated with a more positive attitude. There were no significant correlations between prevention score and attitude. Because the promotion score only helps explain .64% of the variance in subjects scores on the attitude scale, the promotion focus effect was considered negligible. A correlation coefficient less than .09 means ther e is no relationship between the variables (Cohen, 1988). Subsequently, both covariates (promotion score, prevention score) were removed from the data analysis of the hypotheses. H1: When controlling for regulatory focus, subjects exposed to nonloss framed claims will have more positive attitudes toward the product with production claims than those exposed to gainframed labeling claims or neutral general product claims. The covariates for regulatory focus were not included in the analysis because Pearson product moment correlations revealed no relationship between the covariates and the dependent variables. A oneway betweengroups analysis of variance was conducted to compare the different labeling claim framing effects on attitudes toward the product with the claims. The independent variable was the frame of the claim (nonloss, gain, neutral), and the dependent variable was attitude toward the product with the claims. Preliminary assumption testing showed no serious violations noted. There was a significant effect of labeling claim frame on attitudes toward the product with production claims, F (2, 657) = 16.87, p < .001 (see Table 49 ). Planned c ontrasts revealed that subjects exposed to gainframed claims had more positive attitud es toward the product with the claims than those exposed to neutral product claims t (657) = 5.26, p < .001, and those exposed to nonloss framed claims

PAGE 95

95 had more positive attitudes in comparison to the control group as well t (657) = 4.79, p < .001. The difference between gain and nonloss labeling claim frames, however, was not significant t (657) = .64, p = .52 (2 tailed) (see Table 410 and Figure 42 ). H 2 : When controlling for regulatory focus subjects exposed to nonloss framed claims will have less positive attitudes toward the product without production claims than those exposed to gainframed labeling claims or neutral general product claims. The theoretical covariate of regulatory focus was not included in this analysis either. A one way betweengroups analysis of variance was conducted to compare the different labeling claim framing effects on attitudes toward the product without the claims. The independent variable was the frame of the cl aim (nonloss, gain, neutral), and the dependent variable was attitude toward the product without the claims. Preliminary assumption testing was conducted with no serious violations noted. There was a significant effect of labeling claim frame on attitudes toward the product without production claims, F (2, 657) = 6.41, p = .002 (see Table 411). Planned c ontrasts revealed that subjects exposed to gainframed claims had less positive attitudes toward the product without the claims th an those exposed to neutral product claims t (657) = 2.12, p = .035, and those exposed to nonloss framed claims had less positive attitudes in comparison to the control group as well t (657) = 3.56, p < .001. The difference between gain and nonloss labeling claim frames, however, was not significant t (657) = .1.37, p = .17 (2 tailed) (see Table 412 and Figure 43 ). H 3 : Subjects exposed simultaneously to a food product with production claims and a pr oduct without such claims will have less positive attitudes toward the product without the claims than those who do not see a food product with production claims. While the data analyses for H1 and H2 offered insight into this hypothesis, a specific analys is was conducted to offer a complete picture. The treatment groups (nonloss, gain, control) were recoded into a new independent variable that grouped

PAGE 96

96 together subjects in the nonloss and gain conditions because they were the groups that saw the production labeling claims, whereas the control group saw general product claims (boneless and skinless, chicken breasts). An independent samples t test was conducted to compare the presence of the production claims effects on attitudes toward the product without the claims. The independent variable was the presence of the production claims (present, absent), and the dependent variable was attitude toward the product without claims. Preliminary assumption testing was conducted with no serious violations noted. The in dependent samples t test showed a significant difference between the two groups t (658) = .3.31, p = .001 (2tailed). An inspection of the mean scores indicated that subjects exposed to the production labeling claims had less positive attitudes toward the product without claims than those who were not (see Table 413 and Figure 4 4 ). H 4 : Subjects exposed to a food product with production claims will be more likely to have intentions to vote yes for an animal welfare ballot initiative than those who do not see a food product with production claims. A Chisquare test for independence indicated no significant association between subjects exposure to production labeling claims and voting decision on the animal welfare ballot initiative. The majority of subjects voted yes (n = 510, 77.3%). Post Hoc Analyses Several post hoc analyses were conducted to explore other relationships among the variables collected in this study that were not included in the hypotheses. These analyses provide a more complete picture of other variables that influence the dependent variables in this study

PAGE 97

97 Attention To and Purchase of Products With Production Claims Subjects were asked 1) whether they pay attention to, and 2) how often they purchase products with five different types of production labeling claims on meat and/or poultry when grocery shopping: 1) organic labels, 2) labels that address the way the animal was raised, 3) labels that say no hormones, 4) labels that say no antibiotics, and 5) labels that suggest the product is better for the environment (green). These items were recoded into a single variable for attention ( Chronbach alpha coefficient of .8 6 ) and a single variable for purchase frequency (Chronbach alpha coefficient of .89) The relationship between these grocery shopping behaviors (attention and purchase frequency) and attitudes toward the products was investigated using Pearson product moment correlati on coefficient The analyses revealed three key findings ( see Tables 41 4 and 4 1 5 for all results ) The first is that attention to production claims had a strong, positive correlation with purchase frequency of these products r = .58, p < .001 The second is that attention (r= .28, p < .001) and purchase frequency (r= .21, p < .001) both had small, negative correlations with attitude toward the product without production claims (see Table 41 5 ). The more production claims subjects indicated they pay attention to and the more frequently they purchase products with these claims, the less positive their attitudes were toward the product without production claims. Community Upbringing and Livestock Background Subjects identified themselves as growing up in a: subdivision in a city or town (n= 491, 74.4%), rural area, not a farm (n= 98, 14.8%), downtown area in a city or town (n= 47, 7.1%), and farm (n= 23, 3.5%). Table 41 6 displays the descriptive statistics for community upbringing and attitude toward the products Examining the attitudinal means between the two products, it appears that those who identified themselves as

PAGE 98

98 growing up on a farm or in a rural area had more positive attitudes toward both products than those from a subdivision or urban area. Means for attitudes toward the products were compared using oneway ANOVAs to determine if there were any differences based on their self identified community upbringing. There was no significant dif ference between the groups on attitudes toward the product with the claims, F (3, 654 ) = 1. 10, p = 35 There was a significant difference between the groups on attitudes toward the product without the claims ( Table 41 7 ) Despite reaching statistical significance, the actual difference in mean scores between groups was quite small given an effect size of .02. Post hoc comparisons using the Bonferroni correction indicated that the mean score for those who grew up in a rural area, not a farm ( M = 3. 76, SD = 8 3 ) had significantly more positive attitudes tha n those who grew up in a subdivision ( M = 3.49, SD = .83), or an urban (downtow n) area ( M = 3. 34, SD = 82). Table 41 8 displays the results of the post hoc comparisons. An analysis of variance was conducted to determine if having a personal or family background in livestock production affected attitudes toward the product. There was no significant effect of planning to or having a background in livestock pr oduction on attitudes toward the product with the claims F (3, 653) = 1.39, p = .25 or attitudes toward the product without the claims F (3, 653) = 2.29, p = .08. Political Affiliation and Voting Intention As mentioned earlier in the chapter, subjects indicated their political party affiliation and political viewpoint. With respect to political party identification, 265 indicated themselves as Democrat (40.2%), 228 Republican (34.5%), and 78 Independent (11.8%). For political viewpoint, 277 considered t hemselves Liberal (42.0%), 222 Conservative (33.6%), and 107 were neither (16.2%). A Chisquare test for

PAGE 99

99 independence did not show any association between political party or viewpoint and voting intention on the animal welfare ballot. Other NonSignifican t Relationships Several other relationships were investigated using either Pearson product moment correlations, Chi square tests for independence, and AN OVAs depending on the type of variables No statistically significant relationships were found between age and regulatory focus, political affiliations and regulatory focus, gender and regulatory focus, age and attitude s toward the products, gender and attitudes, or course and attitudes.

PAGE 100

100 Table 41. Attention to selected production labeling claims on meat /poultry Yes No Type of Labeling Claim n % n % Organic 306 47.6 337 52.4 Way animal was raised 226 35.1 417 64.9 No hormones 319 49.6 324 50.4 No antibiotics 285 44.4 357 55.6 Better for environment 283 44.1 359 55.9 Table 42. Purchase frequency of meat/poultry with select production labeling claims Type of Labeling Claim n M SD Organic 640 1.21 1.45 Way animal was raised 641 1.13 1.44 No hormones 642 1.55 1.71 No antibiotics 638 1.42 1.66 Better for environment 642 1.21 1.42 Note: Scores based on Likert scale with 0= never, 1= less than once a month, 2= once a month, 3= twice a month, 4= weekly, 5= every time. Table 43. Promotion focus scale inter item consistency statistics M SD Corrected Item Total Correlation Alpha if I tem Deleted RF2 3.51 .9 2 .67 .7 2 RF4 3.34 .9 3 .5 8 .75 RF 5 4.13 .8 2 .55 .7 6 RF6 3.53 .85 .63 .73 RF8 3.37 .84 .4 3 .79 Table 44. Prevention focus scale inter item consistency statistics M SD Corrected Item Total Correlation Alpha if Item Deleted RF 1 3.73 .72 .4 4 .6 4 RF 3 3.83 .76 40 .65 RF7 3.91 .75 40 .65 RF 9 3.58 1.12 .49 .62 RF 10 4.43 .70 .5 3 .60

PAGE 101

101 Table 45. Total attitude scale inter item consistency statistics Product Without Claims Product With Claims M SD Corrected Item Total Correlation Alpha if Item Deleted M SD Corrected Item Total Correlation Alpha if Item Deleted Useless: Useful 3.98 .9 8 .70 .96 4.18 .87 .71 .96 Worthless: Valuable 3.86 .9 7 .75 .96 4.14 .85 .75 .96 Harmful: Beneficial 3.62 1.08 .84 .96 4.17 .8 5 .82 .9 6 Foolish: Wise 3.42 .94 .79 .96 3.82 .96 .75 .96 Unpleasant: Pleasant 3.49 1.04 .84 .96 3.93 .9 3 .82 .9 6 Awful: Nice 3.49 .96 .87 .96 3.91 .91 .84 .9 6 Disagreeable: Agreeable 3.54 .99 .85 .96 3.90 .93 .82 .9 6 Sad: Happy 3.23 .99 .79 .96 3.71 .9 2 .78 .96 Bad: Good 3.53 1.0 8 .88 .96 4.01 .9 2 .84 .9 6 Negative: Positive 3.41 1.05 .87 .96 4.04 .9 1 .85 .9 6 Dislike: Like 3.57 1.13 .86 .96 4.05 .93 .84 .9 6 Unfavorable: Favorable 3.42 1.1 5 .85 .96 4.08 .9 5 .85 .9 6 Unhealthy: Healthy 3.70 1.09 .78 .96 4.27 .84 .75 .96 Unsafe to eat when cooked: Safe to eat when cooked 4.25 .97 .51 .9 7 4.46 .85 .63 .96 From an animal treated inhumanely: From an animal treated humanely 2.92 1.1 5 .60 .9 7 3.95 1. 09 .63 .96 Bad for the environment: Good for the environment 3.08 1.0 4 .66 .96 3.96 1.0 2 .68 .96 Note: Scores based on semantic differential scale from 1= useless to 5= useful. *Researcher developed item. Table 46. Regulatory focus questionnaire descriptive statistics n M SD Compared to most people, are you typically unable to get what you want out of life? (promotion item)* 660 3.73 .7 2 Growing up, would you ever cross the line by doing things your parents would not tolerate? (prevention item)* 660 3.51 .9 2

PAGE 102

102 Table 4 6. Continued n M SD How often have you accomplished things that got you psyched to work even harder? (promotion item) 660 3.83 .7 6 Did you get on your parents nerves often when you were growing up? (prevention item)* 660 3. 34 .9 3 How often did you obey rules and regulations that were established by your parents? (prevention item) 660 4.13 .8 2 Growing up, did you ever act in ways that your parents thought were objectionable? (prevention item) 660 3.5 3 .85 Do you often do well at different things th at you try? (promotion item) 660 3.91 .75 Not being careful enough has gotten me into trouble at times (prevention item)* 660 3.37 .84 When it comes to achieving things that are important to me, I find that I dont perform as well as I ideally would like to. (promotion item)* 660 3. 58 1.1 2 I feel like I have made progress toward being successful in my life. (promotion item) 660 4.43 .70 Note: Scores based on Likert scale of 1= never to 5= very often or 1= certainly false to 5= certainly true. *Item was reverse coded Table 47. Attitude toward product (product specific* + general attitude) Attitude Toward Product Without Claims Attitude Toward Product With Claims n M SD n M SD Useless:Useful 660 3.98 .9 8 660 4.18 .87 Worthless:Valuable 660 3.86 .9 7 660 4.14 .85 Harmful:Beneficial 660 3.62 1.08 660 4.17 .8 5 Foolish:Wise 660 3.42 .94 660 3.82 .96 Unpleasant:Pleasant 660 3.49 1.04 660 3.93 .9 3 Awful:Nice 660 3.49 .96 660 3.91 .91 Disagreeable:Agreeable 660 3.54 .99 660 3.90 .93 Sad:Happy 660 3.23 .99 660 3.71 .9 2 Bad:Good 660 3.53 1.0 8 660 4.01 .9 2 Negative:Positive 660 3.41 1.05 660 4.04 .9 1 Dislike:Like 660 3.57 1.13 660 4.05 .93 Unfavorable:Favorable 660 3.42 1.1 5 660 4.08 .9 5 Unhealthy:Healthy 660 3.70 1.09 660 4.27 .84 Unsafe to eat when cooked: Safe to eat when cooked 660 4.25 .97 660 4.46 .85

PAGE 103

103 Table 4 7. Continued Attitude Toward Product Without Claims Attitude Toward Product With Claims n M SD n M SD From an animal treated inhumanely: From an animal treated humanely 660 2.92 1.1 5 660 3.95 1. 10 Bad for the environment: Good for the environment 660 3.08 1.0 4 660 3.96 1.0 2 Note: Scores based on semantic differential scale from 1= useless to 5= useful. *Researcher developed item to measure product specific attitude. Table 48. Attitude toward product grand means among treatment groups Attitude Toward Product Without Claims Attitude Toward Product With Claims Treatment Group n M SD n M SD Gain Frame 208 3.51 0.81 208 4.17 0.67 Nonloss Frame 236 3.41 0.87 236 4.13 0.68 Control 216 3.68 0.81 216 3.80 0.81 Total 660 3.53 0.84 660 4.04 0.74 Note: Scores ranged from 1 (most negative) to 3 (neutral) to 5 (most positive). Table 49 Effects of labeling claim frame on attitudes toward product with claims Source SS df MS F p Claim Frame 17.49 2 8.75 16.87 .000 Error 340.53 657 .518 Total 358.02 659 Table 410. Planned comparisons t test for differences between treatment groups on attitude toward product with claims n M SD t df p Gain Framed Production Claims 236 4.17 0.67 5.26 657 .000 Neutral Framed Product Claims (Control) 216 3.80 0.81 Nonloss Framed Production Claims 236 4.13 0.68 4.79 657 .000 Neutral Framed Product Claims (Control) 216 3.80 0.81 Gain Framed Production Claims 236 4.17 0.67 .64 657 .52 Nonloss Framed Production Claims 236 4.13 0.68

PAGE 104

104 Table 411. Effects of labeling claim frame on attitudes toward product with claims Source SS df MS F p Claim Frame 9.86 2 4.43 6.41 .002 Error 453.90 657 .691 Total 462.76 659 Table 412. Planned comparisons t test for differences between treatment groups on attitude toward product without claims n M SD t df p Gain Framed Production Claims 236 3.51 0. 81 2.12 657 .035 Neutral Framed Product Claims (Control) 216 3.68 0.81 Nonloss Framed Production Claims 236 3.41 0. 87 3.56 657 .000 Neutral Framed Product Claims (Control) 216 3.68 0.81 Gain Framed Production Claims 236 3.51 0. 81 1.37 657 .17 Nonloss Framed Production Claims 236 3.41 0. 87 Scores ranged from 1 (most negative) to 3 (neutral) to 5 (most positive). Table 413. Independent samples t test for differences between subjects exposed to production claims and subjects exposed to general product claims n M SD t df p Production Claims Absent 216 3.68 0.81 3.31 658 .001 Production Claims Present 444 3.46 0.84 Note: Scores ranged from 1 (most negative) to 3 (neutral) to 5 (most positive). Table 41 4 Pearson product moment correlations between grocery shopping behavior and attitude toward product with production claims Attitude toward product with production claims Attention to production claims in grocery store Purchase frequency of products with production claims Attitude toward product with production claims 1 .08 .002 Attention to production claims in grocery store N= 629 1 .59** Purchase frequency of products with production claims N= 629 N= 629 1

PAGE 105

105 Table 41 5 Pearson product moment correlations between grocery shopping behavior and attitude toward product without production claims Attitude toward product with out production claims Attention to production claims in grocery store Purchase frequency of products with production claims Attitude toward product with out production claims 1 .22** .17** Attention to production claims in grocery store 629 1 .59** Purchase frequency of products with production claims 629 629 1 ** p < .001 (2tailed). Table 41 6 Mean attitude toward products between community upbringing Attitude Toward Product With Claims Attitude Toward Product Without Claims n M SD n M SD Farm 23 4.11 .95 23 3.86 .92 Rural, Not a Farm 98 4.14 .77 98 3.76 .83 Subdivision in Town or City 491 4.02 .73 491 3.49 .83 Downtown in Town or City 47 3.95 .65 47 3.36 .82 Table 41 7 Effects of community upbringing on attitude toward product without claims Source SS df MS F p 2 Label Presence 7.63 1 7.63 11.22 .001 .017 Community Upbringing 9.94 3 3.31 4.87 .002 .022 Error 445.04 654 .68 Total 8682.90 659 Table 41 8 Post hoc comparisons between community upbringing 95% CI Comparisons Mean Attitude Difference SE p Lower Bound Upper Bound Farm vs. Rural .095 .191 1.0 .411 .600 Rural vs. Subdivision .274* .091 .017 .032 .516 Subdivision vs. Urban .128 .126 1.0 .206 .461 Urban vs. Farm .496 .210 .11 1.052 .059 Rural vs. Urban .402* .146 .037 .014 .789 Farm vs. Subdivision .369 .176 .220 .097 .834 Note : p values reflect Bonferroni adjustment for multiple comparisons

PAGE 106

106 Figure 41. Subjects weekly meat consumption

PAGE 107

107 Figure 42. Means between the attitudes toward product with claims in each treatment group. Total attitude includes product specific and general attitude.

PAGE 108

108 Figure 43. Means between the attitudes toward product without claims in each treat ment group.

PAGE 109

109 Figure 44. Means between the attitudes toward product without claims in each treatment group.

PAGE 110

110 CHAPTER 5 CONCLUSION Overview This study address ed two gaps found in the literature. The first one being whether consumers beliefs about conventionally produced food products are affected by production labeling claims and whether this onpackage marketing can also affect intent to support an animal welfare ballot initiative The second gap was to test the predictions of loss aversion and regulatory focus theories using a qualitative means of presenting equivalent gains and nonlosses in the context of labeling claims on meat The purpose of this study wa s to compare the persuasive effects of gainand nonloss framed labeling claims With loss aversion and regulatory focus theory as the theoretical framework, the objectives of this study were to determine the effects of differently framed labeling claims on consumers attitudes toward the product with production claims the conventional product, and voting intention. Two types production labeling claims were constructed (gain and nonloss) to market a package of boneless, skinless chicken breasts as being good for animal welfare and the environment. Originally, health claims were also intended to be used, but extensive pretesting warranted eliminating them from the study. A convenience sample of college students were randomly assigned to receive one of thre e sets of labeling claims: nonloss framed claims on animal welfare and environmental impact, gain framed claims on animal welfare and environmental impact, or general /neutral claims related to the cut of meat. To determine the effects of the treatments, at titudes toward the product with the claims, attitudes toward the product without the claims, and the decision on a hypothetical voting scenario were measured.

PAGE 111

111 Chapter 4 discussed data analyses and results from 660 subjects. The mean age was 21 years old and the sample was primarily female. Most subjects considered themselves from suburban areas and did not have a livestock production background nor planned to in the near future. The subjects tended to be more Democrat and Liberal in their political viewpoi nt than Republican and Conservative or Independent and neither Liberal nor Conservative. Most subjects indicated that they consume meat on a daily to twice daily basis and are the primary grocery shopper for themselves or household. Less than half of the subjects pay attention to five types of production labeling claims and most do not purchase products with these claims or only every so often (less than once per month). They pay the most attention to hormone labeling and the least to claims about how the animal was raised. This chapter presents the key findings, discussion/ implications, limitations, recommendations, and conclusions. Key Findings Descriptive analysis of the data found that subjects were generally more promotion focused than prevention focus ed ; however, having a prevention focus did not influence attitudes toward the product with or the one without the claims and a promotion focus accounted for a negligible amount (<1%) of the variance in those attitudes. A total of four hypotheses were tes ted in the present study. Based on the theories of loss aversion, framing effects, and regulatory focus, t he first hypothesis predicted that w hen controlling for regulatory focus, subjects exposed to nonloss framed claims will have more positive attitudes toward the product with production claims than those exposed to gainframed labeling claims or control group claims This hypothesis was partially supported. Regulatory focus (prevention, promotion) did not influence attitudes towa rd the product with the claims so it was not included as a covariate in the analysis.

PAGE 112

112 The gain and nonloss framed production labeling claims did not lead to significantly different attitudes toward the product with the claims Subjects in both treatment conditions had positive attitudes toward the product with the claims that were, in fact, nearly the same whether they were exposed to nonloss or gain claims. Subjects exposed to gain or nonloss claims had more positive attitudes towards the product with the claims than those exposed to neutral general product claims, but this was more likely an effect of the treatment conditions use of production claims than the frames themselves. The second hypothesis predicted that w hen controlling for regulatory focus, subjects exposed to nonloss framed claims will have less positive attitudes toward the product without production claims than those exposed to gainframed labeling claims or control group claims This hypothesis was partially supported. Subjects exposed to gain claims did not differ from those exposed to nonloss claims in their attitudes toward the product without the claims Subjects exposed to gain or nonloss claims had less positive attitudes towards the product without the claims than those exposed to neutral general product claims. Again, this was likely because of the production claims subjects were exposed to in the treatment conditions rather than the framing. The third hypothesis stated that subjects exposed to a food product with production claims and a product without such claims will have less positive attitudes toward the product without the claims than those who do not see a food product with production claims. This hypothesis was supported. The analyses for H1 and H2 suggested this would be the case, but a direct analysis was conducted to confirm it. An independent samples t test revealed a significant effect of exposure to the production

PAGE 113

113 labeling claims on att itudes toward the product without the claims S ubjects exposed to production claims had less positive attitudes toward the product without the claims, whereas those who were not exposed to production claims had a more positive attitude toward the product w ithout the claims. The fourth and final hypothesis predicted that subjects exposed to a food product with production claims will be more likely to vote yes for an animal welfare ballot initiative than those who do not see a food product with production c laims This hypothesis was not supported. The majority of the subjects voted yes for the animal welfare ballot initiative. Implications The results of this study offer several theoreti cal implications for loss aversion, framing effects, and regulatory focus theories. The theoretical implications are followed by an explanation of the practical implications based on the effects production claim labeling had on attitudes toward the product without those cl aims and voting intentions on an animal welfare ballot initiative. Theoretical Researchers agree that t he way information is framed can influence consumer s judgment and decision about products (see Levin et al., 1998; Rabin, 1998; Boettecher, 2004). Previ ous loss aversion research consistently showed that people have stronger reactions to information presented as potential losses /nonlosses in comparison to equivalent potential gains /nongains ( Kahneman & Tversky, 1979; Tversky & Kahneman, 1981; Boettcher, 2004; McDermott, 2004). Conversely, a few other studi es suggested that gains garner a stronger reaction than nonlosses ( Idson et al., 2000; Liberman et al., 2005).

PAGE 114

114 Th e present study did not find loss/gain asymmetry in support of either prediction. Whether subjects were exposed to gainframed production labeling claims or nonloss claims did not matter; their attitudes toward the products were affected similarly. This could be because the application of the message/information was directly connected with an ordinary market good: food. Horowitz and McConnell (2002) found that the more a good is like an ordinary market good then the lower is the degree of gain/loss asymmetry The production claims themselves however, were less about the product itself and more about the implications for environmental impact and animal welfare as a result of producing that product The environment and animal welfare are non market goods and cannot be directly experienced by the consumer; that is the nature of credence attributes (Darbi & Karni, 1973) Perhaps the predictions of loss aversion would hold when testing the production labeling claims in the absence of the food product. While that would be a clearer test of the prediction, it is less representative of the reality of how these production claims are frequently encountered by consumers. Also, since attitudes toward the product (rather than the claims) were measured, the utilitarian value inherit in the chicken may be confounding the framing effect This is commonly the case in advertising and marketing research; the basis of the field of advertising is that the information provided with and/or about the product affects consumer judgment of the product (Young, 2008). Another reason could be that the information (the prod uction labeling claims) in this study w as presented in a qualitative manner rather than the typical quantitative manner used in many previous studies supporting loss aversion (Kahneman & Tversky, 1979; Tversky & Kahneman, 1981; Levin et al., 1998; Boettcher, 2004; McDermott,

PAGE 115

115 2004) and in those supporting regulatory focus theory (Idson et al., 2000; Liberman et al., 2005) These studies did not always use numbers, but some used examples that could be quantified (i.e., reducing cholesterol in Levin et al., 2001). Holistic environmental impact and animal welfare are difficult to quantify objectively ( Broom 1991; Stolze, Piorr, H ring, & Dabbert, 2000), or, at best, would be difficult for the average consumer to fully interpret (Bateman, Dent, Peters, Slovic, & Starmer 2007) Consumers rely on food production certif ication agencies (government and thirdparty) to make the interpretations and provide them a trustworthy generalization of the meanings of good animal welfare and environmental impact ( Caswell & Mojd uszka 1996; Golan et al. 2001 ; Auriol & Schilizzi 2003 ) Also, framing information as gains and nonlosses primarily affects the reference point people use to make judgments and decisions (Heath et al., 1999). Soman (2004) explained that values are coded as gains and losses relative to a reference point, meaning a decision is reference dependant Presenting information about a product in a qualitative manner might cause consumers to automatically adjust their reference point because numerical values are not available to encode the message as a gain or a nonloss. Therefore, qualitatively created frames (i.e., no negative environmental impacts vs. good for the environment) may not communicate the intended reference point strongly enough, but are, therefore, e qually persuasive on attitudes Bateman, Day, Jones, and Jude (2009) suggested that an individual is able to interpret that one numeric value is larger than another without necessarily understanding its meaning, thereby leading to the reliance on heuristic s and biases to form judgment This study attempted to frame nonlosses and gains equivalently, but qualitatively. The results suggest that in the absence of numbers or

PAGE 116

116 quantifiable information, the bias es of loss aversion and framing effects are minimized. The message may need to include terms that more strongly suggest a reference point, such as reduce environmental impact or improve environmental impact, to induce the biases. Regulatory focus (prevention, promotion) also did not explain much of the variance in attitudinal response. This could be for the same reasons as explained earlier with respect to loss aversion and framing effects; however, regulator y focus studies are more likely to use what researchers call prevention and promotion messages that do not involve quantifiable information and are usually not equivalent (see Aaker & Lee, 2001; Wang & Lee, 2006; Florack & Scarabis, 2006; Zhao & Pechmann, 2007) According to regulatory focus theory, subjects with a high promotion (prevention) focus score should have been more sensitive to the presence of gains (nonlosses), meaning they should have had a stronger attitudinal response to the labeling claim frame that fit their regulatory focus. Previous research has found that negative (loss) frames are more persuasive with preventionfocused individuals than positive (nonloss) f rames, and that positive (gain) frames are more persuasive than negative (nongain) frames with promotionfocused individuals (Zhao & Pechmann, 2007) This may explain why a very small correlation existed between promotionfocus and attitudes but not between prevention focus and attitudes Many regulatory focus researchers advise marketers to create and frame messages that are in line with audiences regulatory focus for effective persuasion (Aaker & Lee, 2001; Wang & Lee, 2006; Florack & Scarabis, 2006; Zhao & Pechmann, 2007). This study implies that a simplistic message, such as a few words i n a labeling claim may not be strong enough to elicit the regulatory fit effect It may take

PAGE 117

117 additional priming of a regulatory focus to fully induce it (Freitas & Higgins, 2002; Florack & Scarabis, 2006) or a more precise manipulation of message wording (e.g., avoid environmental damage, achieve positive environmental impacts) Practical The theoretical hypotheses testing showed that production claims framed as gains or nonlosses produced similar positive attitudinal effects on the credence attribute product with the production claims Gain framed claims produced slightly (but not statistically significant) more positive attitudes toward the product with the claims, but slightly less negative attitudes toward the product without the claims. M arketers of cr edence attribute food products could potentially encourage purchase by placing products with gainframed claims in their own section of the grocery store (away from the conventional products without the claims) and those with nonloss framed claims next to the conventional items. However, additional research adding price variation as an additional independent variable would need to be considered because this study held price consistent across both product types. When examining the attitudinal effects of production labeling claims, s ubjects exposed to those claims had less positive attitudes toward the product without the claims than those exposed to general product claims. The product without the claims was meant to represent the conventional commodity product to determine how the marketing of credence attribute products affects peoples attitudes toward the conventional product. The results did not show that exposure to production claims produces negative attitudes toward conventional products but it did produce markedly less positive attitudes. The results show that consumers view the conventional product inferior to the credence attribute product on the aspects of safety, healthiness, humane

PAGE 118

118 animal treatment, and environmental friendliness as well as on more general aspects Thus, the production claims are a source of information reducing consumers positive attitudes toward those aspects of conventional agriculture production and its food product s Previous work suggested this may be the case (Klonsky & Tourte, 1998), but this study shows that it is. The claims could serve as a prompt, causing consumers to recall negative information from the news media and mass media ( Craven & Johnson, 1999) It is unclear how much of consumers beliefs and subsequent attitudes can be accounted for by various communication channels (i.e., advertisements, labels, media, websites, social media, etc.) For example, would the majority of consumers know (or be concerned about) antibiotic and hormone use or confinement in livestock production if they werent inundated with more expensive products claiming to be absent of those inputs? This study shows that the production claims are a source of information that produces inferior attitudes toward conventional products without suc h claims. The exposure to production labeling claims and subsequent attitudes produced, however, did not translate into voting intention on an animal welfare ballot initiative. Subjects overwhelmingly supported this law. The reason no treatment effects wer e seen could be due to several reasons. First, political decision making information that affects decisions typically comes in forms of communication (i.e., TV ads, websites, news and editorials, etc.). This study intended to determine if food labels could be a source of communication affecting political decision making, but did not find that to be the case. Another reason could be that the measure was a oneitem, dichotomous measure of behavioral intent. Multiple item measurement with a wider scale would better capture the variance that naturally exists in complex decision making.

PAGE 119

119 Interestingly, the subjects indicated they pay the least amount of attention to animal welfare claims and purchase food with such claims the least in comparison to four other types of claims ( no hormones, no antibiotics, organic, and environmentally friendly). This food shopping characteristic is similar to studies surveying general adult consumer populations ( Verbeke & Viaene, 1999; Yiridoe et al., 2005 ; Hughner, McDonagh, Prothero, Shultz, & Stanton, 2007). The data shows, on the other hand, that they are willing to support legislation that would make it required of all livestock producers to provide their livestock more space in confinement, which is an animal welfare considerat ion. While subjects were willing to support a government policy, they are not willing to put their money where their mouth is. When evaluating a potential risk, decision making research has shown people must be paid more to accept a risk than they are wi lling to pay to avoid that risk (Thaler, 1980 ; Horowitz & McConnell, 2002) While this concept is not entirely related to the findings, it suggests that people are generally willing to pay less to avoid a risk, and supporting this legislation costs no mone y (at least immediately) to avoid the risk completely. These subjects were not provided any additional information about the cons equences of passing such a law. W hen this particular law was passed in California, 63.5 % of the voters supported it. This study shows that, without education and persuasive communication efforts, that number could be much higher with young adults. From the post hoc analyses that explored the relationships between some of the demographic variables and dependent variables, it i s interesting to note the relationship between community upbringing and effects on attitude toward the product without the cla ims. Those who indicated they grew up in a rural area (not a farm) had more positive

PAGE 120

120 attitudes toward the conventional product tha n those from subdivisions or urban areas., While no other statistically significant relationships were found, examination of the means across the four groups shows that those from farm and rural communities generally have more positive attitudes towards both products. This relationship and pattern could be because th ose from a farm or rural community feel a stronger connection with agriculture and its products as a result of knowing farmers in their community, exposure to high school agriculture programs, and/or involvement in 4H and/or FFA. Rural communities are more likely to have farmers and youth agricultural programs in comparison to suburban and urban communities (Frick, Birckenholz, Gardener, & Matchtmes, 1995). The other relationship of interest f ound in the post hoc analyses was the small, negative correlation between attention to production claims and attitude toward the product without claims, and purchase frequency and attitude toward the product without claims. While paying attention to more c laims and purchasing these products more frequently does not correlate with having more positive attitudes toward products that carry those claims, these behaviors do reduce their attitudes toward products without such claims. It seems that what potentiall y encourages attention to and purchase of products with production claims is the devaluation or fear of conventional products. Limitations While the present study offers several useful theoretical and practical insights, there were some limitations that should be considered. The convenience sample of college students is one of the key limitations, primarily for the practical implications and recommendations. College students are still developi ng their consumer habits and civic engagement. These may change w ith further maturity, experience, and when starting a

PAGE 121

121 family. For example, c onsumers with children are more likely to learn about and purchase organic foods (Hughner et al., 2007) Readers should carefully consider the demographic information before applyi ng conclusions to other populations. Research consistently shows that consumers purchase and prefer organic and natural meat products for health reasons (AMI & FMI, 2008, Yiridoe et al., 2005; Hughner et al., 2007). While consumers often make connections between environmental impacts and animal welfare and health concerns (Hughner et al., 2007), those elements themselves are not as concerning as chemicals, hormones, and antibiotics, which are primarily healthrelated (Yiridoe et al., 2005). This study was unable to find a plausible health claim in equivalent gain and nonloss frames related to those elements to test the effects. Th e present study used a onetime only exposure to the production labeling claims Strong attitudes resistant to change require repeated exposure to persuasive messages (Perloff, 2008 ). R epeated exposure over longer periods of time could reveal a greater influence of the production labeling claims on attitudes. Though the claims used in this study are those that do exist in reality. Given about half of the subjects in this study at least pay attention to such claims, this was likely not their first exposure to them. Another limitation of the study was the less than ideal reliability of the prevention focus scale. Perhaps the use of a nother regulatory focus scale (e.g., Lockwood, Jordan, & Kunda, 2002) would prove more reliable.

PAGE 122

122 Recommendations For Future Research and Theory From a theoretical perspective, more research needs to be done examining the effects of gains versus nonlosses. This study attempted to further some of the previous research in that area (Idson et al., 2004; Liberman et al., 2005), but perhaps due to the qualitative nature of the frames and the nature of the application (food product), did not find asymmetry in the attitudinal reactions to gains versus nonlosses. Researchers in these theoretical areas should consider future studies that attempt to manipulate gains and nonlosses qualitatively to determine if biases are minimized as a result. The finding that regulatory focus did not influence attitudinal response to gainor nonloss framed claims should be supplemented with using regulatory focus priming to determine potential effects of this cognitive style in combination with the use of a more reliable measure for r egulatory focus (e.g., Lockwood et al., 2002). One of the goals of this research was to determine if and how production labeling claims affects attitudes toward the product, and therefore, only assumptions can be made regarding how it affects beliefs and attitudes about agricultural production. A branch of this study would be to determine how this type of labeling affects consumers belief s and attitudes about agricultural production and food safety directly Furthermore, a national survey of where consumers obtain their information about agriculture production, food, and farming life would be beneficial to agricultural communications researchers, especially as the U.S. population employed in agriculture continues to dwindle. This would help determine the role marketing and advertising, as well as other forms of communication, play in forming beliefs and attitudes.

PAGE 123

123 The present study held sev eral variables consistent to determine the effect of the differently framed production labeling claims on attitudes. Additional manipulations of variables such as product type, price brand, and other packaging characteristics would be beneficial to market ers and may produce different attitudinal effects. Also, as mentioned in the limitations, further research into production claim labeling should test health claims since that is the primary reason driving consumer perceptions and purchase of such products (AMI & FMI, 2008, Yiridoe et al., 2005; Hughner et al., 2007). A follow up study should also include other dependent measures that may be affected by food labeling claims. B ehavior, such as willingness to pay and purchase likelihood would offer additional insight into the effects of food labels In addition, w hile attitudes can be a useful measure of food label communication effects, it would be worthwhile to examine other effects such as risk perceptions. In peoples subjective evaluation of risk, nine general properties of activities or technologies emerge: (1) voluntariness of risk, (2) immediacy of effect, (3) knowledge about the risk by the person who are exposed to the potentially hazardous risk source, (4) knowledge about the risk in science, (5) con trol over the risk, (6) newness, i.e. are the risks new and novel or old and familiar ones, (7) chronic/ catastrophic, (8) common/dread, i.e. whether people have learned to live with and can think about the risk reasonably and calmly, or is it a risk that people have great dread for, and (9) severity of consequences (Fischoff, Slovic, Lichtenstein, Read, & Combs, 2000) Using those dimensions of risk, a measure of livestock production risk perceptions could be measured. As previously mentioned, a more compl ex measure of voting intention would also be useful in capturing greater variability in the potential effects of food labeling as communication affecting political

PAGE 124

124 decisions. The manipulations of nonloss and gain messages in these studies should include terms like reduce and improve to more strongly suggest a reference point that is moved toward or away from to determine if the biases of loss aversion and regulatory focus fit effect are subsequently induced. For Practitioners In this study, exposure to production labeling claims about animal welfare and environmental impact reduced positive attitudes toward the product without such claims. Specifically, the conventional product was viewed as inferior to the credence attribute p roduct on the aspects of safety, healthiness, humane animal treatment, and environmental friendliness, as well as on more general aspects. While this is likely viewed as a positive finding for those with a vested interest in alternative agriculture product ion and products, it is probably concerning to those who believe in the merits of conventional agriculture. The marketing of credence attribute products contributes to the devaluation of products that do not have such claims; however, many products, even t hose from conventional systems could qualify for many different types of production claims. It is recommended that those within the agricultural industry develop a system to explore the facets of farming operations that may qualify food products for produc tion and/or processing claims especially those related to health and food safety, animal welfare, and environmental impact The results of this study also imply that agricultural communicators working on behalf of conventional agriculture need to help reb uild attitudes toward that type of production system and help consumers understand the meanings and implications of various food labels They also need to assist in communication efforts with opinion leaders, policy makers, and voters on agricultural polic y issues. Beyond that, agricultural communicators should help their organizations

PAGE 125

125 and businesses understand and value these attitudes because the controversy over alternative agriculture and conventional agriculture is far from over (see Paarlberg, 2010 and Lapp 2010) Government food regulators must consider the effects of food labeling to ensure the policies, standards, and guidelines for such labels are balancing the market for agricultural products and n ot misleading consumers (Golan et al., 2001). If organic and other credence attribute labeled foods continue to be perceived as the safer and better food choice, the market for conventional foods could potentially suffer. Government regulators must meticulously consider these types labeling claims bef ore approving them and be responsible for communicating their meaning to consumers. Marketers and advertisers intending to apply the principles of loss aversion, framing effects, and regulatory focus need to carefully consider the limitations of these theories, especially when making general, qualitative claims. Careful development and testing of the framed messages needs to be conducted to ensure the messageframe audience combinations produce the intended effects. Conclusions Using loss aversion and regulatory focus as a theoretical framework, the objectives of this study were to determine the effects of differently framed labeling claims on consumers attitudes toward the product with production claims the conventional product, and voting intention. Thi s study attempted to frame nonlosses and gains equivalently, but qualitatively. The results suggest that in the absence of numbers or quantifiable information, the biases of loss aversion, framing effects, and regulatory focus fit effect are minimized. Adv ertisers and marketers should carefully consider the

PAGE 126

126 potential limitations of these theories and thoroughly test differently framed messages or claims before intending to leverage the power cognitive heuristics and biases. The mere exposure to the product ion labeling claims, no matter how framed, produced equally positive attitudes toward the product; however, their presence also decreased the positive attitudes held toward the product without such claims. These types of food labels are a source of information affecting consumers attitudes towards conventional agriculture products and perhaps even the production system. Agricultural communicators should not underestimate the effects that food marketing and advertising can have on consumers attitudes towar d conventional agriculture and its products, and consider these effects in addition to messages put forth by activist groups and mass media.

PAGE 127

127 APPENDIX A VERBAL PRENOTIFICATION PowerPoint slide used by instructors of courses which contained the sample for the study to announce the upcoming study.

PAGE 128

128 APPENDIX B FIRST CONTACT E MAIL SENT TO SUBJECTS Dear ${m://FirstName}, I am conducting a study about food products and would like to gather your opinions about them through an online survey. Your instructor has agreed to offer you extra credit in ${e://Field/course} for participating in the study. The survey will take approximately 15 minutes of your time. To take the survey, you will need to know your unique participant ID number, which is ${e://Field/ID}. Please take extra care to type in the correct ID number. Do NOT delete this email if you think you might want to take the survey later because it contains your unique participant ID number. The su b sequent reminder emails may not contain this required information. The survey is online. Click or copy paste the following link into your Internet browser: ${l://SurveyURL}. The link will only be active until 11:59pm, March 28. If you have questions or problems accessing the survey, please email me at kchodil@ufl.edu or call 352392 0502 ext. 238. Follow this link to the Survey: ${l://SurveyLink?d=Take the Survey} Or copy and paste the URL below into your internet browser: ${l://SurveyURL} Tha nk you, Katie Abrams Graduate Assistant University of Florida

PAGE 129

129 APPENDIX C FIRST AND SECOND REMINDER E MAIL SENT TO SUBJECTS Dear ${m://FirstName}, You are receiving this email because I do not have record of you completing the survey for extra credit in ${e://Field/course} for completing it. You certainly are not obligated to complete the survey; this is simply a reminder message. It will take approximately 15 minutes of your time. To take the survey, you will need to know your unique participant ID number, which is ${e://Field/ID}. Please take extra care to ensure you type in the correct participant ID number. Your instructor will have the list of names for extra credit by March 31, so please check with them at that time to ensure you got the points The survey is online. Click or copy paste the following link into your Internet browser: ${l://SurveyURL}. The link will only be active until 11:59pm, Sunday, March 28. If you have questions or problems accessing the survey, please email me at kchodil @ufl.edu or call 352392 0502 ext. 238. Follow this link to the Survey: ${l://SurveyLink?d=Take the Survey} Or copy and paste the URL below into your internet browser: ${l://SurveyURL} Thank you, Katie Abrams Graduate Assistant University of Florida

PAGE 130

130 APPENDIX D THIRD REMINDER E MAIL SENT TO SUBJECTS Dear ${m://FirstName}, Today is the final day to take the online survey for extra credit in ${e://Field/course}. The survey will take approximately 15 minutes of your time. To take the survey, you will need to know your unique participant ID number, which is ${e://Field/ID}. Please take extra care to ensure you type in the correct participant ID number. Your instructor will have the list of names for extra credit by March 31, so please check with them at that time to ensure you got the points. The survey is online. Click or copy paste the following link into your Internet browser: ${l://SurveyURL}. The link will only be active until 11:59pm tonight. If you have questions or problems accessing the surv ey, please email me at kchodil@ufl.edu or call 352392 0502 ext. 238. Follow this link to the Survey: ${l://SurveyLink?d=Take the Survey} Or copy and paste the URL below into your internet browser: ${l://SurveyURL} Thank you, Katie Abrams Graduate Assistant University of Florida

PAGE 131

131 APPENDIX E INSTRUMENT

PAGE 132

132

PAGE 133

133

PAGE 134

134

PAGE 135

135

PAGE 136

136

PAGE 137

137 Gain condition

PAGE 138

138 Gain condition

PAGE 139

139 Other Treatment Groups Nonloss Condition Photos Control Condition Photos

PAGE 140

140

PAGE 141

141

PAGE 142

142

PAGE 143

143

PAGE 144

144

PAGE 145

145

PAGE 146

146

PAGE 147

147 Gain condition

PAGE 148

148 Gain condition

PAGE 149

149 Control condition

PAGE 150

150 LIST OF REFERENCES Abrams, K. M. & Meyers, C. A. (2009). A comparison of persuasive message factors and frames in animal agriculture communication campaigns on the Web. Paper presented at the Association for Communication Excellence Conference in Agriculture, Natural Resources, and Life Sciences. Des Moines, IA. Abrams, K. M., Meyers, C. A., & Irani, T. A. ( 2009). Naturally confused: Consumers perceptions of organic and allnatural pork products. Journal of Agriculture and Human Values [in press] doi: 10.1007/s1046000992345 American Meat Institute & Food Marketing Institute. (2008). The power of meat: An indepth look at meat through the shoppers eyes. Paper present ed at the Annual Meat Conference, Nashville, TN. Retrieved August 17, 2009, from http://www.meatconference.com/ht/a/GetDocumentAction/id/38142 Ary, D., Jacobs, L. C., & Razavieh, A (2002). Introduction to research in education (6th ed). Belmont, CA: Wadsworth/Thomson Learning. Auriol, E. & Schilizzi, S.G.M. (2003). Quality signaling through certification: Theory and an application to agricultural seed markets Working paper no 165, IDEI University of Toulouse. Bateman, I. J., Day, B. H., Jones, A. P., & Jude, S. (2009). Reducing gain loss asymmetry: A virtual reality choice experiment valuing land use change. Journal of Environmental Economics and Management, 58(1), 106 118. Bateman I. J. Dent, S., Peters, E., Slovic P. & Starmer, C. (2007) The affect heuristic and the attractiveness of simple gambles. Journal of Behavioral Decision Making 20(4), 365 380 Batra, R., & Ahtola, O. T. (1991). Measuring the hedonic and utilitarian s ources of consumer attitudes. Marketing Letters 2 (2), 159170. Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D. (2001). Bad is stronger than good. Review of General Psychology, 5(4), 323 370. Bech Larsen, T., Grunert, K. G., & Poulsen, J. B. (2001). The acceptance of functional foods in Denmark, Finland and the United States: A study of consumers' conjoint evaluations of the qualities of functional foods and perceptions of general health factors and cultural values. MAPP working paper no 73. Aarhus, Germany: MAPP Center. Becker, T., Benner, E., & Glitsch, K. (2000). Consumer perception of fresh meat quality in Germany. British Food Journal, 102(3), 246 266.

PAGE 151

151 Blandford, D., Bureau, J. C., Fulponi, L., & Henson, S. (2002). Potential implcations of animal welfare concerns and public policies in industrialized countries for international trade. In B. Krissof, M. Bohman, & J. Caswell, Global Food Trade and Demand for Quality (pp. 77 100). New York: Kluwar Ac ademic/Plenum Publishers. Bostrm M., & Klintman, M. (2003) Framing, debating, and standardising natural food in two different political contexts: Sweden and the U.S. Score Working Paper no. 2003:3. Stockholm, Sweden: Stockholm Center for Organizational Research, Stockholm School of Economics Brenner, L., Rottenstreich, Y., Sood, S., & Bilgin, B. (2007). On the psychology of loss aversion: Possession, valence, and reversals of the endowment effect. Journal of Consumer Research, 34 369376. Bruns K., Fjord, T., & Grunert, K (2002). Consumers food choice and quality perception. MAPP working paper no 77. Aarhus, Germany: MAPP Center. California Institute for Rural Studies. (2005). Regulating organic: Impacts of the national organic standards on consumer awareness and organic consumption patterns Retrieved August 20, 2009, from http://www.cirsinc.org/Documents/Pub1205.2.PDF Caswell, J A. (1998). How labeling of safety and process attributes affects markets for food. Agricultural and Resource Economics Review, 27(2), 151 158. Ca swell J. A & Mojduszka, E. M. ( 1996). Using informational labeling to influence the market for food quality American Journal of Agricultural Economics 78( 5 ), 12481253. Cesario, J., Higgins, E. T., & Scholer, A. A. (2008). Regulatory fit and persuasion: Basic principles and remaining questions. Social and Personality Psychology Compass 2 (1), 444 463. Craven, B. & Johnson, C. (1999). Politics, policy, poisoning and food scares. In Morris, J., & Bate, R. (Eds.). Fearing Food: Risk, Health and Environment (pp. 141 169). Oxford: Butterworth Heinemann. Crowell, S. (2009, April 2). The three Rs of the HSUS agenda. Farm and Dairy Retrieved August 19, 2009, from http://www.farmanddairy.com/columns/thethreers of the hsus agenda/11606.html Crowley, A. E., Spangenberg, E. R., & Hughes, K. R. (1992). Measuring the hedonic and utilitarian dimensions of attitudes toward product categories. Marketing Letters, 3(3), 239249. Dangour, A. D., Dodhia, S. K., Hayter, A., Allen, E., Lock, K., & Uauy, R (2009). Nutritional quality of organic foods: A systematic review. American Journal of Clinical Nutrition [in press]. doi:10.3945/ajcn.2009.28041

PAGE 152

152 Darbi, M. R., & Karni, E. (1973). Free competition and the optimal amount of fraud. Journal of Law and Economics, 16 (1), 67 88. DeGregori, T. R. (2003). Origins of the organic agriculture debate. Ames, IA: Iowa State Press. DeVellis, R. F. (2003). Scale development: Theory and applications (2nd ed). Thousand Oaks, CA: Sage. Downing, B. (2009, July 19). Humane Society, farmers prepare for war: Battle lines are forming over proposal to change Ohio rules on methods of confining livestock. Akron Beacon Journal Retrieved August 19, 2009, from http://www.ohio.com/ne ws/51120387.html Dubisch, J. (2004). You are what you eat: Religious aspects of the health food movement. In C. L. Delaney (Ed.). Investigating culture: An experiential introduction to anthropology (pp. 311 319). Malden, MA: Blackwell Publishing. Dunlap, R ., & Mertig, A. (Eds.). (1992). American environmentalism: The US environmental movement, 19701990. London: Taylor and Francis. Druckman, J. N. (2004). Political preference formation: Competition, deliberation, and the (ir)relevance of framing effects. Am erican Political Science Review, 98 671 86. Druckman, J. N. (2001). On the limits of framing effects: Who can frame? The Journal of Politics, 63 (4), 10411066. Eicher, A. (2003). Organic agriculture: A glossary of terms for farmers and gardeners Eurka, CA: University of California Cooperative Extension. Retrieved May 7, 2010 from http://ucce.ucdavis.edu/files/filelibrary/1068/8286.pdf Entman, R. (1993). Framing: T oward clarifi cation in a fractured paradigm. Journal of Communication, 43(4), 51 58 Evans, L. M., & Petty, R. E. (2003). Self guide framing and persuasion: Responsibility increasing message processing to ideal levels. Personality and Social Psychology Bulletin, 29(3), 313 324. Factory. (1989). Oxford English Dictionary Retrieved May 14, 2010, from http://dictionary.oed.com/cgi/entry/50081547 ? Fischoff, B., Slovic, P., Lichtenstein, S., Read S., & Combs, B. (2000). How safe is safe enough? A psychometric study of attitudes toward technological risks and benefits. In: P. Slovic (Ed.), The perception of risk (pp. 80104). London: Earthscan. Fitzgerald, D. (2003). Every farm a factory: The industrial ideal in American agriculture. New Haven, CT: Yale University Press.

PAGE 153

153 Florack, A. & Scarabis, M. (2006). How advertising claims affect brand pre ferences and category brand associations: The role of regulatory fit. Psychology and Marketing, 23(9), 741755. Florack, A., Scarabis, M., & Gosejohann, S. (2005). Regulatory focus and consumer information processing. In Kardes, F., Herr, P., & Nantel, J. (Eds.), Applying social cognition to consumer focused strategy (pp. 235 263). Mahwah, NJ: Lawrence Erlbaum Associates. Freitas, A. L., & Higgins, E. T. (2002). Enjoying goal directed action: The role of regulatory fit. Psychological Science, 13, 1 6. Frick, M., Birckenholz, R., Gardener, H., & Matchtmes, K. ( 1995). Rural and urban inner city high school student knowledge and perception of agriculture. Journal of Agricultural Education, 36(4), 1 9. Gabbett, R. J. (2008, Dec. 10). Meat industry faces emboldened animal rights lobby next year. Meatingplace Industry News Retrieved August 19, 2009, from http://www.meatingplace.com/MembersOnly/webNews/details.aspx?item=10700 Garner, R. A. (1993). Animals, politics & morality Manchester University Press: Manchester. Gilovich, T. (1991). How we know what isnt so: The fallibility of human reason in everyday life. New York: Free Press. Gilovich, T., & Griffin, D. W. (200 2). Heuristics and biases: Then and now. In Gilovich, T., Griffin, D. W., & Kahneman, D. ( Eds. ) Heuristics and biases: The psychology of intuitive judgment (pp. 1 18). Cambridge: Cambridge University Press. Gilovich, T., Griffin, D. W., & Kahneman, D. (20 02). Heuristics and biases: The psychology of intuitive judgment. Cambridge: Cambridge University Press. Golan, E., Kuchler, F. & Mitchell, L. (2001). Economics of food labeling. Journal of Consumer Policy, 24 117 184. Gold, M. V. (2010, March). Should I purchase organic foods? USDA Alternative Farming Systems Information Center Retrieved May 24, 2010, from http://www.nal.usda.gov/afsic/pubs/faq /BuyOrganicFoodsD.shtml Goodwin, J., & Rhoades, E. (2009). Agricultural legislation: The presence of California Proposition 2 on YouTube. Paper presented at the National American Association for Agricultural Education Conference, Louisville, KY. Retrieved August 19, 2009, from http://www.aaaeonline.org/files/national_09/ papers/2.pdf Gottlieb, R. (2005). Forcing the spring: The transformation of the American environmental movement Washington DC: Island Press.

PAGE 154

154 Grant, J ( 2008). The green marketing manifesto. John Wiley & Sons Available from http://common.books24x7.com/book/id_23397/book.asp Greene, C., Dimitri, C., Lin, B., McBride, W., Oberholtzer, L., & Smith, T. (2009). Emerging issues in the U.S. organic industry (Economic Information Bulletin No. 55). W ashington DC: United States Department of Agriculture Economic Research Service. Retrieved August 19, 2009, from http://www.ers.usda.gov/Publications/EIB55/EIB55.pdf Guthman, J. (2004). The trouble with organic lite in California: A rejoinder to the conventionalism debate. Sociologia Ruralis, 44(3), 301 316. Hale, T. (2009, June 12). Most households read food labels. Nielsen Wire Retrieved August 24, 2009, from http://blog.nielsen.com/nielsenwire/consumer/most households read food labels Harris poll results show who is buying organic foods, how frequently. (2007, October). Nutrition Business Journal, 12(10), 21. Heath, C., Larrick, R. P., & Wu, W. (1999). Goals as reference points. Cognitive Psychology, 38 79 109. Higgins, E. T. (1998). Promotion and prevention: Regulatory focus as a motivational principle. In Zanna, M. P. (Ed.), Advances in Experimental Social Psychology (pp. 1 46) New York: Academic Press. Higgins, E. T. (2002). How self regulation creates distinct values: The case of promotion and prevention decision making. Journal of Consumer Psychology, 12(3), 177 191. Higgins, E. T., Friedman, R. S., Harlow, R. E., Idson, L. C., Ayduk, O. N., & Taylor, A. (2001). Achievement orientations from subjective histories of success: Promotion pride versus prevention pride. European Journal of Social Psychology, 31, 3 23. Hig gins, E. T, & Silberman, I. (1998). Development of regulatory focus: Promotion and prevention as ways of living. In Heckhausen, J., & Dweck, C. (Eds.), Motivation and self regulation across the life span (pp. 78113). New York: Cambridge University Press Hooker, R. J. Food and drink in America. Indianapolis: The Bobs Merrill Company, Inc. Horowitz J. K. & McConnell, K. E. (2002) A review of WTA/WTP studies Journal of Environmental Economics and Management 44, 426 447 Hu, W., Woods, T., & Bastin, S. (2009). Consumer acceptance and willingness to pay for blueberry products with non conventional attributes. Journal of Agricultural and Applied Economics, 41 (1), 47 60.

PAGE 155

155 Hughner, R., McDonagh, P., Prothero, A., Shultz II, C., & Stanton, J. (2007). Who are organic food consumers? A compilation and review of why people purchase organic food. Journal of Consumer Behaviour 6 (2/3), 94110. doi:10.1002/cb.210. Humane Farm Animal Care. (n.d.). What is certified humane raised and handled? Retrieved Sept ember 2, 2009, from http://www.certifiedhumane.org/about/whatis.html Hurt, R. D. (2002). American agriculture: A brief history West Lafayette, IN: Purdue University Press. Hwang, Y., Roe, B., & Teisl, M. F. (2005). An empirical analysis of United States consumers' concerns about eight food production and processing technologies. AgBioForum, 8(1), 40 49. Idson, L. C., Liberman, N., & Higgins, E. T. (2004). Imagining how good or bad youd feel: A motivational experience beyond outcomes. Personality and Social Psychology Bulletin 30 926 937 Jacob, F., & Ehret, M. (2006). Self protection vs opportunity seeking in business buying behavior: A n experimental study Journal of Business & Industrial Marketing, 21(2), 106 117. Jauregui, C., & Ward, R. W. (2006, July). Do consumers really use food labels? Paper presented at the American Agricultural Economics Association (New Name 2008: Agricultural and Applied Economics Association) 2006 Annual meeting, July 2006, Long Beach, CA. Retrieved August 24, 2009, from http://purl.umn.edu/21142 Jones, S., JohnsonYale, C. Millermaier, S & Seoane Perez, F. (25 September 2009) Everyday life, online: U.S. college stud ents use of the Internet. First Monday [Online], 14 (10). Retrieved May 14, 2010 from http://www.uic.edu/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/2649/2301 Ka hneman, D., Diener, E., & Schwartz, N. (Eds.) (2003). Wellbeing: The foundations of hedonic psychology. New York: Russell Sage Foundation Publications Kahneman, D., Knetsch, J., & Thaler, R. (1990). Experimental test of the endowment effect and the coase theorem. Journal of Political Economy 98(6), 13251348. Kahneman, D., Ritov, I., & Schkade, D. (1999). Economic preferences or attitude expressions? An analysis of dollar responses to public issues. Journal of Risk and Uncertainty 19 220 242. Kahneman, D., & Tversky, A. (1979). Prospect theory : An analysis of decision under risk. Econometrica 47 263291.

PAGE 156

156 Kam, C. D., Wilking, J. R., & Zechmeister, E. J. (2007). Beyond the n arrow d ata b ase: Another convenience s ample for e xperimental r esearch. Political Behavior, 29(4). doi: 10.1007/s11109007 90376 Kempton, W. Boster, J. & Hartley, J. (1995). Environmental values in American culture. Cambridge: MIT Press. Klonsky, K., & Tourte, L. (1998). Organic agricultural production in the United States: Debates and directions. American Journal of Agricultural Economics, 80(5), 11191124. Knorr D., & Watkins, T. R. ( 1984) Alterations in food production. New York : Van Nostrand Reinh old Lapp A. (2010, April 29). Dont panic, go organic Foreign Policy. Retrieved May 7, 2010, from http://www.foreignpolicy.com/articles/2010/04/29/ dont_panic_go_organic Levin, I. P., & Gaeth, G. J. (1988). Framing of attribute information before and after consuming the product. Journal of Consumer Research, 15, 374 378. Levin, I., Schneider, S., & Gaeth, G. (1998). All frames are not created equal: A typology and critical analysis of framing effects. Organizational behavior and human decision processes, 76(2), 149 188. Levenstein, H. (2003). The paradox of plenty: A social history of eating in modern America. Berkeley and Los Angeles, CA: University of California Press. Liberman, N., Idson, L. C., & Higgins, E. T. (2005). Predicting the intensity of losses vs. nongains and nonlosses vs. gains in judging fairness and value: A test of the loss aversion explanation. Journal of Experimental Social Psychology, 41(5), 527534. Lindenberg, S., & Steg, L. (2007). Normative, gain and hedonic goal frames guiding environmental behavior. Journal of Social Issues, 63(1), 117137. Lobao, L., & Meyer, K. (2001). The g reat a gricultural t ransition: Crisis, change, and social consequences of t wentieth c entury US f arming. Annual Review of Sociology, 27(1), 103 124. Lockwood, P., Jordan, C., & Kunda, Z (2002). Motivation by positive or negative role models: Regulatory focus determines who will best inspire us. Journal of Personality and Social Psychology 83 (4), 854 864. doi: 10.1037/00223514.83.4.854 Loureiro, M. L., McCluskey, J. J., & Mittelhammer, R. C. (2005). Assessing consumer preferences for organic, ecolabeled, and regular apples. Journal of Agricultural and Resource Economics, 26 (2), 404416.

PAGE 157

157 Magkos, F., Arvaniti, F., & Zampelas, A. (2006). Organic food: Buying more safety or just peace of mind? A critical review of the literature. Critical Reviews in Food Science and Nutrition, 46(1), 23 56. Maheswaran, D. & M eyers Levy, J. (1990). The influence of message framing and issue involvement. Journal of Marketing Research, 27, 361 67. Maule, J., & Villejoubert, G. (2007). What lies beneath: Reframing framing effects. In Lagnado, D. A., & Read, D. (Eds.), Judgment and choice: Perspectives on the work of Daniel Kahneman. Thinking and Reasoning, 13, 2544. McDermott, R. (2004). Prospect theory in political science: Gaines and losses from the first decade. Political Psychology, 25(2), 289312. Miller, M. G., & Miller, K. U. (2000). Promoting safe driving behaviors: The influence of message framing and issue involvement. Journal of Applied Social Psychology, 30(4), 853 866. Morrison P C., Nehring, R., Banker, D ., & Somwaru, A. (2004). Scale economies and efficiency in U.S a griculture: Are traditional farms history? Journal of Productivity Analysis 22 185205. Nestle, M. (2007). Food politics: How the food industry influences nutrition and health. Berkeley, CA: University of California Press Ness, M., Matthew G., & Kuznesof, S. (2002). The student food shopper: Segmentation on the basis of attitudes to store features and shopping behavior. British Food Journal, 104(7), 506 526. Nisbet, M. C., Scheufele, D. A., Shanahan, J., Moy, P., Brossard, D., & Lewenstein, B. V. (2002). Knowledge, reservations, or promise? A media effects model for public perceptions of science and technology. Communication Research, 29(5), 584 608. Obach, B. K. (2007). Theoretical interpretations of the growth in organic agriculture: Agriculture modernization or an organic treadmill? Society & Natural Resources An International Journal, 23(3), 229 244. Oberholtzer, L., Greene, C., & Lopez, E. (2006). Organic poultry and eggs capture high price premiums and growing share of specialty market (LDP M 15001). Retrieved November 17, 2009, from the USDA Economic Research Service website: http://www.ers.usda.gov/Publications/LDP/2006/12Dec/LDPM15001/ldpm15001.p df O rganic Trade Association (2008). Health of the planet and its inhabitants Retrieved August 19 2009 from http://www.ota.com/organic/benefits/health.html

PAGE 158

158 Onyango, B. M., Hallman, W. K., & Bellow s, A. C. (2007). Purchasing organic food in US food systems: A study of attitudes and practice. British Food Journal, 109(5), 399411. Ostrom, T. M., Bond, Jr., C. F., Krosnick, J. A., & Sedikides, C. (1994). Attitude scales: How we measure the unmeasurabl e. In S. Shavitt & T.C. Brock (Eds.) Persuasion: Psychological insights and perspectives (pp. 1542). Needham Heights, MA: Allyn and Bacon. Padel, S., & Foster, C. (2005). Exploring the gap between attitudes and behavior: Understanding why consumers do or do not buy organic food. British Food Journal, 8, 606625. Paarlberg, R. (2010, May/June). Attention Whole Foods shoppers. Foreign Policy Retrieved May 7, 2010, from http://www.foreignpolicy.com/articles/2010/04/26/ attention_whole_foods_shoppers Perloff, R. M. (2008). The dynamics of persuasion: Communication and attitudes in the 21st century (3rd ed.). New York: Lawrence Erlbaum Associates. Peterson, R. A. (2001). On the u se of college students in social science r esearch: Insights from a s econd o rder m eta analysis Journal of Consumer Research, 28. doi : 10.1086/323732 Petty, R. E., & Cacioppo, J. T. (1986). Communication and persuasion: Central and peripheral routes to attitude change. New York: Springer Verlag. Pollan, M. (2004). The cheapest calories make you the fattest: A foodchain journalist looks for stories in our meals Retrieved August 18, 20009, from http://www.michaelpollan.com/press.php?id=7 Prop 2: Standards for confining farm animals (2008). Retrieved November 19, 2009 from California Voter Information Guide: http://voterguide.sos.ca.gov/past/2008/general/titlesum/prop2 title sum.htm Rabin, M. ( 1998). Psychology and economics. Journal of Economic Literature, 36(1), 11 46. Reips, UD. (2000). The web experiment method: Advantages, disadvantages, and solutions. In M. H. Birnbaum (Ed.), Psychological e xperiments on the web (pp. 89117). London: Academic Press. Rollin, B. E. (1990). Animal welfare, animal rights, and agriculture. Journal of Animal Science, 68(10), 3456 3461. Rollin, B. E. (2003). Farm a nimal w elfare: Social, b ioethical, and r esearch i ssues. Ames, IA: Iowa State Press.

PAGE 159

159 Rozin, P., Spranca, M., Krieger, Z., Neuhaus, R., Surillo, D., Swerdlin, A. et al. (2004). Preference for natural: Instrumental and ideational/moral motivations, and the contrast between foods and medicines. Appetite, 43 (2), 147 154. Sassenrath et al. (200 8). Technology, complexity and change in agricultural production systems. Renewable Agriculture and Food Systems 23, 285 295. Shadish, W. R., Cook, T. D., & Campbell D. T. (2002). Experimental and quasi experimental designs for generalized causal inference Boston: Houghton Mifflin Company Shamaskin, A. (2009). Getting the message across: Examining information presentation and healthcare decision making among older adults (Unpublished thesis, Cornell University, 2009). Retrieved September 13, 2009, from Cornell University Library website http://hdl.handle.net/1813/12656 Shiv, B. Carmon, Z., & Ariely, D. (2005). Placebo effects of marketing actions: Consumers may get what they pay for. Journal of Marketing R esearch, 42(4), 383 393. Simon, H. A. (1955). A behavioral model of rational choice. The Quarterly Journal of Economics, 69(1), 99 118. Smith, R. (2009, March 5). Prop 2 opening door to promote veganism Feedstuffs Foodlink. Retrieved August 19, 2009, from http://www.feedstuffsfoodlink.com/ME2/ dirmod.asp?sid=F4A490F89845425D8362C0250A1FE984&nm=&type=news&mo d=News&mid=9A02E3B96F2A415ABC72CB5F516B4C10&tier=3&nid=B0D5904C A7FE4934ABEF9024D30444A8 Soman, D. (2004). Framing, loss aversion, and mental accounting. In Koehler, D. J., & Harvey, N. (Eds.), Blackwell handbook of judgment and decision making, (pp. 379 398) Malden, MA: Blackwell Publishing. Stolze, M., Piorr, A. Hring, A.M. & Dabbert, S. (2000) Environmental impacts of organic farming in Europe. Organic Farming in Europe: Economics and Policy 6 Universitt Hohenheim, Stuttgart Hohenheim. Retrieved from http://orgprints.org/8400 Storck, A. (2008). Organic meat and poultry will keep growing in 2009: trade group. Meat Place Industry News Retrieved A ugust 17, 2009, from http://www.meatingplace.com/MembersOnly/webNews/details.aspx?item=10801 Sustainable Agriculture Research & Education. (n.d.). Exploring sustainabilit y in agriculture. Retrieved September 20, 2009, from http://www.sare.org/publications/explore/explore.pdf Sustainable Table. (n.d.). What is sustainable agriculture? Retrieved September 20 2009, from http://www.sustainabletable.org/intro/whatis/

PAGE 160

160 Thaler, R. (1980). Toward a positive theory of consumer choice. Journal of Economic Behavior and Organization, 1, 39 60. The Humane Society of the United States. (2009, Oct. 12). Mich. Gov. Granholm signs historic farm animal welfare measure. Retrieved November 18, 2009, from http://www.humanesociety.org/news/press_releases/2009/10/mich_gov_granholm _signs.html The Humane Soc iety of the United States. (2008). Research Retrieved August 19, 2009, from http://www.hsus.org/farm/resources/research/ The NPD Group, Inc. (2009). Better for you foods to grow significantly ov er the next decade. The NPD Group, Inc. Press Release Retrieved August 17, 2009, from http://www.npd.com/press/releases/press_090707a.html Todorov, A., Chaiken, S., & Henderson, M. D. (2002). The Heuristic Systematic Model of social information processing. In Dillard, J., & Pfau, M. (Eds.), The persuasion handbook: Developments in theory and practice (pp. 195211). Thousand Oaks, CA: Sage Publications, Inc. Tversky, A. & Kahneman, D. (1981). The framing of decisions and the psychology of choice. Science, 211, 453 458. Traub, R. E. (1994). Reliability for the social sciences: Theory and applications Thousand Oaks, CA: Sage. United States Department of Agriculture Agric ultural Marketing Service. (2008). LS ISO Guide 65 Program. Retrieved September 2, 2009, from http://www.ams.usda.gov/ AMSv1.0/ams.fetchTemplateData.do?templa te=TemplateD&navID=GradingCertifi cationandVerfication&leftNav=GradingCertificationandVerfication&page=LSISO65 Program USDA Food and Safety Inspection Service. (2003). Food standards and labeling policy book Retrieved November 17, 2009, from http://www.fsis.usda.gov/oppde/larc / Policies/PolicyBook.pdf United States Department of Agriculture Food Safety and Inspection Service. (n.d.). Animal production claims Retrieved September 2, 2009, from http://www.fsis.usda.gov/OPPDE/larc/Claims/RaisingClaims.pdf University of Florida Office of Institutional Planning and Research. (2009). Enrollment: Final headcount enrollment by class level, gender and ethnicity (19972009). Retrieved from http://www.ir.ufl.edu/factbook/enroll.htm Verbeke, W., & Viaene, J. (1999). B eliefs, attitude and behavior towards fresh meat consumption in Belgium: Empirical evidence from a consumer survey. Food Quality and Preference, 10, 437 445.

PAGE 161

161 Vermeir, I., & Verbeke, W. (2006). Sustainable food consumption: Exploring the consumer "attitudebehavioral intention" gap. Journal of Agricultural and Environmental Ethics 19 169194. Wakker, P., K bberling, V., & Schwieren, C. (2007). Prospect theory s diminishing sensitivity versus economics intrinsic utility of money: How can the introduction of the Euro be used to disentangle the two empirically. Theory and Decision, 63, 205 231. Wang, J., & Lee, A. Y. (2006). The role of regulatory focus in preference construction. American Marketing Association, 43 28 38. Williams, D. L., & Wise, K. L. (1997). Perceptions of Iowa secondary school agricultural education teachers and students regarding sustainable agriculture. Journal of Agricultural Education, 38 (2), 15 20. Yale Center of Environmental Law and Policy's Environmental Attitudes and Behavior Project. (2007). Yale environmental poll. Retrieved September 20, 2009, from http://envirocenter.research.yale.edu/uploads/epoll/YaleEnvironmentalPoll2007Ke yfindings.pdf Yiridoe, E. K., Bonti Ankomah, S., & Martin, R. C. (2005). Comparison of consumer perceptions and preference toward organic versus conventionally produced foods: A review and update of the literature. Renewab le Agriculture and Food Systems, 20(4), 193 205 Young, C. (2008). The advertising research handbook (2nd ed). Seattle: Ideas in Flight. Zeithaml, V. A. (1988). Consumers perceptions of price, quality, and value: A means end model and synthesis of evidence. Journal of Marketing, 52(3), 2 22. Zhao, G., & Pechmann, C. ( 2007). The impact of regulatory focus on adolescents response to antismoking advertising campaigns. Journal of Marketing Research, 44(4), 671 687.

PAGE 162

162 BIOGRAPHICAL SKETCH Katherine (Katie) Abrams was born and raised for 18 years in Shorewood, IL. She received her B S from Purdue University and her M .S. from the University of Florida, both in agricultural communication s. Her overarching research goal is to gain an understanding of how people make sense of and participate in debates about agricultural and environmental issues. She often researches in the context of organic and natural foods because it provides an intersection for examining food safety, animal welfare, and environmental attitudes and thought processes. Her communication skills and teaching expertise are in visual communications, Web design and usability, social media, and public relations. In August 2010, Katie will begin her career as a visiting assistant professor at the University of Illinois, ChampaignUrbana in the department of advertising teaching agricultural communications.