Citation
Evaluation of Consumer Preferences Regarding Goat Meat in Florida

Material Information

Title:
Evaluation of Consumer Preferences Regarding Goat Meat in Florida
Creator:
Knight, Erica ( Dissertant )
House, Lisa ( Thesis advisor )
Degner, Robert ( Reviewer )
Nelson, Mack ( Reviewer )
Place of Publication:
Gainesville, Fla.
Publisher:
College of Agricultural and Life Sciences, University of Florida
Publication Date:
Copyright Date:
2005
Physical Description:
Thesis

Subjects

Subjects / Keywords:
Consumer prices ( jstor )
Consumer surveys ( jstor )
Demography ( jstor )
Fats ( jstor )
Food ( jstor )
Goats ( jstor )
Hispanics ( jstor )
Meat packing industry ( jstor )
Meats ( jstor )
Telephones ( jstor )

Notes

Abstract:
Florida is a part of the United States' goat production region and its goat inventory increased sixty-three percent between 1992 and 1997. Goat meat has always been a minor food item in the United States, but the importance of meat goats to farm income has increased in recent years. The rapid growth of ethnic populations has led to increased consumption since Hispanics, Muslims, and individuals with African ancestry are major consumers of goat meat products. Florida has a strong potential for a meat goat market; however, the lack of consumer information has hindered producers' , processors' , and marketers' ability to fully delineate the profitability and viability of the industry. The overall objective of this research project is to identify the factors that influence goat meat consumption among Hispanic and non-Hispanic consumers. This information can be used by goat meat industry officials to better understand who their consumer is and further develop their markets. The data for this study were obtained using telephone surveys that targeted Florida residents. The Institute for Behavior Research and the Survey Research Center at the University of Georgia in Athens, GA, collected the data using the random digit dialing probability technique to minimize bias. The questionnaire sought information on demographic as well as consumer preferences towards goat meat. According to the results from a probit analysis, consumption of other meats and various demographic and psychographic characteristics influenced the willingness to try goat meat. Respondents from households with 5 or more people were most likely to try goat meat, and as household size decreased the willingness to try tended to decline. The results also suggest individuals who had previously consumed lamb were more inclined to try goat meat.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Knight, Erica. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
5/1/2005
Resource Identifier:
71315329 ( OCLC )

Downloads

This item has the following downloads:

knight_e ( .pdf )

knight_e_Page_010.txt

knight_e_Page_048.txt

knight_e_pdf.txt

knight_e_Page_089.txt

knight_e_Page_028.txt

knight_e_Page_064.txt

knight_e_Page_092.txt

knight_e_Page_020.txt

knight_e_Page_004.txt

knight_e_Page_100.txt

knight_e_Page_034.txt

knight_e_Page_062.txt

knight_e_Page_023.txt

knight_e_Page_019.txt

knight_e_Page_054.txt

knight_e_Page_051.txt

knight_e_Page_078.txt

knight_e_Page_036.txt

knight_e_Page_069.txt

knight_e_Page_087.txt

knight_e_Page_060.txt

knight_e_Page_096.txt

knight_e_Page_022.txt

knight_e_Page_058.txt

knight_e_Page_005.txt

knight_e_Page_086.txt

knight_e_Page_011.txt

knight_e_Page_017.txt

knight_e_Page_065.txt

knight_e_Page_008.txt

knight_e_Page_095.txt

knight_e_Page_061.txt

knight_e_Page_029.txt

knight_e_Page_021.txt

knight_e_Page_040.txt

knight_e_Page_043.txt

knight_e_Page_090.txt

knight_e_Page_038.txt

knight_e_Page_027.txt

knight_e_Page_046.txt

knight_e_Page_015.txt

knight_e_Page_025.txt

knight_e_Page_003.txt

knight_e_Page_049.txt

knight_e_Page_001.txt

knight_e_Page_067.txt

knight_e_Page_002.txt

knight_e_Page_098.txt

knight_e_Page_055.txt

knight_e_Page_088.txt

knight_e_Page_093.txt

knight_e_Page_079.txt

knight_e_Page_074.txt

knight_e_Page_016.txt

knight_e_Page_080.txt

knight_e_Page_032.txt

knight_e_Page_066.txt

knight_e_Page_057.txt

knight_e_Page_006.txt

knight_e_Page_094.txt

knight_e_Page_035.txt

knight_e_Page_056.txt

knight_e_Page_068.txt

knight_e_Page_026.txt

knight_e_Page_009.txt

knight_e_Page_024.txt

knight_e_Page_084.txt

knight_e_Page_039.txt

knight_e_Page_077.txt

knight_e_Page_031.txt

knight_e_Page_037.txt

knight_e_Page_070.txt

knight_e_Page_071.txt

knight_e_Page_018.txt

knight_e_Page_042.txt

knight_e_Page_097.txt

knight_e_Page_030.txt

knight_e_Page_082.txt

knight_e_Page_053.txt

knight_e_Page_072.txt

knight_e_Page_013.txt

knight_e_Page_052.txt

knight_e_Page_099.txt

knight_e_Page_075.txt

knight_e_Page_081.txt

knight_e_Page_007.txt

knight_e_Page_047.txt

knight_e_Page_012.txt

knight_e_Page_083.txt

knight_e_Page_014.txt

knight_e_Page_059.txt

knight_e_Page_045.txt

knight_e_Page_091.txt

knight_e_Page_085.txt

knight_e_Page_073.txt

knight_e_Page_063.txt

knight_e_Page_076.txt

knight_e_Page_044.txt

knight_e_Page_033.txt

knight_e_Page_050.txt

knight_e_Page_041.txt


Full Text











EVALUATION OF CONSUMER PREFERENCES REGARDING GOAT MEAT IN
FLORIDA

















By

ERIKA KNIGHT


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2005

































This document is dedicated in loving memory of my grandmother, Mattie Hugley Dixon.





























Copyright 2005

by

Erika Knight















ACKNOWLEDGMENTS

First and foremost, I thank God for giving me the strength I needed to successfully

complete this master's program. I know that He is the reason I have made it this far in

life. I thank family and friends for the encouragement, support, and patience throughout

the past two years. These individuals are another reason I have made it to this point of

my life.

I express my deepest appreciation to my committee, Drs. Lisa House, Robert

Degner and Mack Nelson, for the interest and support given while preparing my thesis. I

would especially like to thank my committee chair, Dr. Lisa House, for her advisement,

guidance and the knowledge she shared while completing my master's program. I am

truly grateful to my committee for all of their assistance and I would like to express my

gratitude once again.

Finally, I would like to thank my fellow graduate students in the Food and

Resource Economics Department and members of the Black Graduate Student

Organization (BGSO) for their support and encouragement. I thank them for the

memories; each of them has made this graduate experience unforgettable.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES ......... .......................................... ....... .. ........ ........ vii

LIST OF FIGURE S ........ ........ .......................................... .............. viii

ABSTRACT .............. .......................................... ix

CHAPTER

1 IN TR O D U C T IO N ............................................................. .. ......... ...... .....

P re a m b le ................................................................ 1
Problem atic Situation.............................................. 1
R searchable Problem ..................... .... .......................... .. ........ .............. ...
O bje ctiv e s ..................................................................................................... . 8
Hypotheses............................ .. ..... ..... ......... ...... 9

2 LITERA TU RE REV IEW ............................................................ ............ 11

3 SUREVEY CON TEN TS .............................. ........ ........... ...... ............... 19

Su rv ey In strum ent....... ........................................................................ ...... ........ .. 19
Survey C ontents.................................................. 21

4 D A T A .................................................................................................................... 2 4

5 THEORETICAL MODEL AND MODEL SPECIFICATIONS..............................32

T h eo retical M o d el ............................. ............................................ .. ..................3 2
P ro b it M o d e l ............................................................................................................... 3 3
M odel Specification ........ ....... ............................................................... ... .... .... .. 36

6 EM PIRICAL RE SU LTS ................................................... ............................... 40

Entire Survey Sample Probit Estimates...... ............ ..................40
H isp an ic M o d el............................ ........................................... 4 5




v









7 CONCLUSIONS AND IMPLICATIONS ...................................... ...............49

S u m m a ry ............................................................................................................... 4 9
C o n c lu sio n s........................................................................................................... 5 2
Im p licatio n s ................................................................5 3

APPENDIX

A SOUTHERN REGION GOAT CONSUMPTION SURVEY............... ................ 56

B SUMMARY OF DEMOGRAPHIC INFORMATION .............................................81

L IST O F R E F E R E N C E S ........................................................................ .....................85
















LIST OF TABLES


Table p

1-1. Florida farms with goats, goat numbers and goats sold, goat production region,
and United States: 2002 and 1997 ................ .......................................... ... 3

1.2. United States meat goat imports and exports. ...................................................

4-2. Summary of demographic information. ...................................... ............... 28

4-3. Descriptive statistics on additional factors that influence the willingness to try. ....29

4-4. Summary of demographic information for Hispanic respondents .........................30

4-5. Descriptive statistics on additional factors that influence the Hispanic
respondents' willingness to try ....... ..... ......... ................... ... ..... .......... 31

5-1. Probit model variables and description. ...................................... ............... 39

6-1. Empirical results from the whole survey probit model.........................................41

6-2. Empirical results from the Hispanic probit model. .............................................46
















LIST OF FIGURES


Figure p

4-1. Comparison of Florida census data and survey respondents by age......................27

4-2. Comparison of Florida census data and Hispanic survey respondents by
educational attainm ent levels. .............................................................................28

6-1. Changes in the probability of trying goat meat with respect to household size.
(All household sizes are compared to the base, households containing five or
m ore in div idu als)............................................................................ ............... 4 3

B-1. Respondents' consumption preference for goat meat by ethnicity/race .................81

B-2. Respondents' consumption preference for goat meat by gender............................81

B-3. Respondents' consumption preference for goat meat by educational attainment
lev els .............................. ......... .................. .............. ................. 82

B-4. Hispanic respondents' consumption preference for goat meat by origin .................82

B-5. Hispanic respondents' consumption preference for goat meat by gender ..............83

B-6. Hispanic respondents' consumption preference for goat meat by educational
attain ent levels ................... ............ ......... ..........................83

B-7. Hispanic respondents' consumption preference for goat meat by generation in
U united States. ...................................................... ................. 84















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

EVALUATION OF CONSUMER PREFERENCES REGARDING GOAT MEAT IN
FLORIDA

By

Erika Knight

May 2005

Chair: Lisa House
Major Department: Food and Resource Economics

Florida is a part of the United States' goat production region and its goat

inventory increased sixty-three percent between 1992 and 1997. Goat meat has always

been a minor food item in the United States, but the importance of meat goats to farm

income has increased in recent years. The rapid growth of ethnic populations has led to

increased consumption since Hispanics, Muslims, and individuals with African ancestry

are major consumers of goat meat products. Florida has a strong potential for a meat goat

market; however, the lack of consumer information has hindered producers', processors',

and marketers' ability to fully delineate the profitability and viability of the industry.

The overall objective of this research project is to identify the factors that

influence goat meat consumption among Hispanic and non-Hispanic consumers. This

information can be used by goat meat industry officials to better understand who their

consumer is and further develop their markets. The data for this study were obtained

using telephone surveys that targeted Florida residents. The Institute for Behavior









Research and the Survey Research Center at the University of Georgia in Athens, GA,

collected the data using the random digit dialing probability technique to minimize bias.

The questionnaire sought information on demographic as well as consumer

preferences towards goat meat. According to the results from a probit analysis,

consumption of other meats and various demographic and psychographic characteristics

influenced the willingness to try goat meat. Respondents from households with 5 or more

people were most likely to try goat meat, and as household size decreased the willingness

to try tended to decline. The results also suggest individuals who had previously

consumed lamb were more inclined to try goat meat.














CHAPTER 1
INTRODUCTION

Preamble

Columbus introduced goats to North America in the area that is presently known as

Texas during the 1600's and goat production in the South has flourished ever since their

introduction (Walsh, 1995). In the past decade, meat goat production has dramatically

increased throughout the Southeastern region of the United States, Florida in particular.

The meat goat industry in Florida has evolved and producers who once viewed their

animals as a sideline operation now consider it a serious business (Simpson, 1995).

Consumption of goat meat and goat meat products has become more popular in the

United States and the strongest demand for goat meat is along the eastern coast, Southern

California, Florida, Detroit, and the northeast region stretching from Washington, D.C. to

Boston (McKenzie-Jakes, 2004). The lack of consumer information poses a barrier in

goat meat marketing and it has prevented producers, processors, and marketers from fully

ascertaining the profitability and viability of the industry. This thesis will use cross-

sectional survey data to assist in identifying and understanding the factors that influence

the consumption of goat meat in Florida.

Problematic Situation

A 2000 study by McLean-Meyinsse identified meat goat production as a new

enterprise that is believed to enhance farm incomes for small farmers. Goat meat is an

extremely popular food item in the Mediterranean, the Middle East, Africa, South

America, Central America, the West Indies, and Southeastern Asia. However, goat meat









consumption in the United States has been historically low. In recent years, U.S.

domestic demand for goat meat has increased with the largest demand along the eastern

coast of the United States, especially in Florida. Florida's, goat meat consumption

exceeds production (McKenzie-Jakes, 2004) creating a need to import live goats or goat

meat in order to satisfy demand (Gipson, 1999). Despite the product's increase in

popularity, the lack of availability of goat meat in Florida hampers the industry's

profitability.

Florida is a part of the eleven state southern goat production region stretching from

Texas to North Carolina (TX, LA, OK, AK, MS, AL, FL, GA, TN, SC, and NC). In

2002, this region accounted for 80 percent of meat-type goat production in the United

States, and Florida contributed three percent of the region's meat goat inventory (USDA-

NASS, 2002). During the past few decades, Florida's meat goat industry has transformed

from one in which goats were raised as a minor part of subsistence level farm system into

a more structured industry oriented approach that is viewed as a business (Simpson,

1995). According to the 2002 Census of Agriculture, 1,764 meat goat farms existed in

Florida, a six percent decrease from the total number of meat goat farms in 1997.

However, during the same time period, the state's meat goat inventory increased twenty-

five percent (Table 1-1). In addition between 1997 and 2002, the number of goats sold

increased 47. Meat goat farms accounted for over eighty-eight percent of the goat farms

within the state in 2002.

Growth in Florida's meat goat industry is attributed to a number of factors.

Historically, the majority of U.S. immigrants were from European countries, but changes

in immigration patterns have occurred.











Table 1-1. Florida farms with goats, goat numbers and goats sold, goat production region, and
United States: 2002 and 1997.
Percent Florida's
Geographical Area 2002 1997 Change Contribution
Florida
Number of Goat Farms 1,992 2,114 (5.8)
Number of Meat Goat Farms 1,764 1,931 (8.6)
Numbers of Meat Goats on Farms 36,020 28,737 25.3
Number of Meat Goats Sold 18,769 13,700 37.0
Percent of Meat Goat Farms 88.6 91.3
Percent of Meat Goat Sold 52.1 47.7
Goat Production Region
Number of Goat Farms 42,487 37,303 13.9 4.7
Number of Meat Goat Farms 1,729,158 1,727,978 0.1 4.6
Numbers of Meat Goats on Farms 1,433,589 975,931 46.9 2.5
Number of Meat Goats Sold 765,622 407,078 88.1 2.5
Percent of Meat Goat Farms 91.2 88.7
Percent of Meat Goat Sold 53.4 41.7
United States
Number of Goat Farms 91,462 76,543 19.5 2.2
Number of Meat Goat Farms 74,980 63,422 18.2 2.4
Numbers of Meat Goats on Farms 1,938,924 1,231,762 57.4 1.9
Number of Meat Goats Sold 1,109,619 532,792 108.3 1.7
Percent of Meat Goat Farms 82.0 82.9
Percent of Meat Goat Sold 57.2 43.3
Source: USDA-NASS, 2002.

The majority of current immigrants are people from Hispanics, Caribbean Islanders,

Muslims, and Asian populations, as opposed to the historical pattern of immigration from

Europe. These changes in immigration patterns from European countries to those from

regions of the world that are perceived to have a dietary preference for goat meat have

increased the demand for the product (McKenzie-Jakes, 2003; Pinkerton et al., 1994;

Sherman, 2002; Walsh, 1995). People from these populations tend to retain food

preferences and religious affiliations in an effort to maintain their ethnic identity when

merging with a dominant group (Harwell, 1995). Thus, the demand for goat meat among









target consumers is expected to be inelastic (Harwell, 1995; Pinkerton et al., 1994). It is

projected that by the year 2025, the U.S. population will increase 44 million due to the

increase of foreign populations, many of which consume goat meat. Therefore, the

demand for this product in the United States is expected to expand as population

increases and as minority purchasing power increases (Russell, 2000).

Health conscious consumers and members of the "yuppie community" who

consume the meat as a gourmet item create an additional demand for goat meat products

(Harwell, 1995; McKenzie-Jakes, 2004; McLean-Meyinsse, 2003; Pinkerton et al., 1994;

Sherman, 2002). Members of the health food sector consume goat meat because of the

product's nutritional attributes. Goat meat is a naturally lean meat that is low in

cholesterol, low in fat and high in proteins (Johnson, 1995).

Advancements in Florida's food distribution infrastructure allow for refrigerated

goat meat to be delivered from Texas in an inexpensive and quick manner to consumers.

Additionally, low ocean freight rates have permitted goat imports from Australia to occur

more frequently (Simpson, 1995). Previously, goat meat has previously been viewed as

an unusual food item in the United States; however it is now more mainstream.

Researchers have found it difficult to accurately estimate per capital consumption

levels of goat meat due to discrepancies in goat inventories, auction runs, and slaughter

numbers (Gipson, 1999; Harwell, 1995). Therefore, it is popular for researchers to use

indirect methods to approximate per capital consumption. Degner and Moss (1999)

surveyed wholesalers in Fort Lauderdale, Miami, and Tampa, Florida. These markets are

considered to consist of significant populations with preferences for goat meat. Using

estimated wholesale sales, the researchers quantified the annual per capital goat meat









consumption to be a mere 0.21 pounds, nearly 20 percent less that per capital consumption

levels in 1986. However, a 1999 study conducted by Putnam and Allshouse estimated

that the national per capital goat meat consumption significantly increased to about one

pound per capital in 1997. Gipson (1999) utilized National Statistical Service data and

Foreign Agricultural Service import/export data as two indicators to measure goat meat

demand. In 1998 almost 450,000 goats were slaughtered which was a 1,000 percent

increase over a 20-year period. It is worth mentioning that the estimates do not include

goats slaughtered in non-United States Department of Agricultural facilities or state

inspection facilities and farm slaughter or personal slaughter. It is commonly accepted

that demand for goat meat is underestimated because the bulk of goat meat consumed is

undocumented, i.e. the majority of the slaughtering occurs in non-UDSA inspected

facilities (Davis and Willard, 1996). According to the Foreign Agricultural Service data,

the United States has been a net importer of goat meat product since 1991 (Gipson,

1999). Table 1-2 illustrates the quantity of goat meat imported to the U.S. and the

quantity exported by the United States from 1989 through 1994. Based on the data, the

United States does not have an adequate supply of goat meat production to keep up with

demand. In 1998, imports rose to 4,500 metric tons, which is equivalent to over 600,000

pounds. The United States imports the majority of its goat meat from Australia and New

Zealand.

Available research suggests the demand for goat meat is influenced by consumers'

age, gender, race, household sized, and martial status (McLean-Meyinsse, 2003; Nelson

et al., 1999). Additionally, carcass weight and carcass size preferences differ among

target populations within the niche market (Pinkerton et al., 1994).













Table 1.2. United States meat goat imports and exports.
Year U.S. Imports* U.S. Exports* Balance*
1989 86,067 122,056 35,989
1990 99,353 115,413 16,060
1991 122,932 53,246 -69,686
1992 172,280 60,444 -111,836
1993 136,364a 3,504b -132,860
1994 138,481a Noneb -138,481
*Values are expressed in metric tons.
a These figure probably reflect reduced Australians in exports of goat due to serve
drought conditions.
b The steep drop in exports as imports fell markedly in 1993 and 1994, thus
conserving domestic supplies.
Source: Ohio Cooperative Development Center, 2004

Hispanics tend to prefer young kids, weighing 15-25 pounds live weight and young goats

weighing about 50 pounds live weight. Hispanics are the fastest growing ethnic group

and the largest minority group in the United States. An emphasis on understanding the

factors that influence consumption should be placed on Hispanics because they are the

only group that has an expected year round demand for goat meat products (Spaugh,

1997). Muslims favor heavier goats than Hispanics and consume goats that are about 70

pounds live weight, male, and intact. Additionally, Jamaicans, Haitians, West Africans,

African Americans have a preference for mature goats (Pinkerton et al., 1994).

There is a relatively thin body of published literature regarding the marketing of

goats and meat goat, but researchers seem to agree on the barriers surrounding the

marketing of goat meat. Pinkerton et al. (1992) found that the marketing constraints that

hinder the demand for goat meat include an unorganized marketing infrastructure, lack of

quality grades, seasonal demand, inconsistent supply, negative consumer attitudes,

inconsistent quality, and insufficient research to identify new market and expand existing









markets. Pinkerton et al. (1994) suggest that a lack of marketing information is a result

of inadequate interest by many state departments of agriculture and the structure of the

industry. Pinkerton et al. (1994) also suggested that in niche markets, located primary in

urban centers, demand for goat exceeds domestic supply, with the shortages caused by

inefficiencies in markets. Studies by Degner and Moss (1999) and Degner and Locascio

(1988) identified the constraints commercial retailers suggest hamper ability to sell goat

in Florida as insufficient demand, supply problems, cheaper substitutes, and product

form.

Growth in Florida's meat goat inventory has been encouraged by numerous factors,

particularly in that Americans are becoming more ethnically diverse and health-

conscious, and more are revealing greater willingness to try new exotic food products

such as rabbit and goat (McLean-Meyinsse, 2000). Many consumers lack knowledge

and exposure to goat meat products, creating an additional barrier faced by the industry

(Zachery and Nelson, 1992). Studies suggest that once consumers are informed on the

product's nutritional attributes and preparation methods, consumer interest in consuming

the product increases (Miller, 1995; Rhee et al. 2000). For instance, Rhee et al. (2000)

found that after participants were educated on goat meat's nutritional values more than 60

percent of the respondents reversed their previous perception of the product. Existing

research also identifies the carcass size preferred by populations within the niche market,

but fails to identify the specific cuts of goat meat that are preferred by populations in the

niche market. However, little inductive research has been completed to identify the

characteristic of goat meat consumers or the characteristics of the product that influence

consumption. Therefore, information that describes the factors that influence









consumption is needed in order to fully understand the marketing and advertising

approaches that will assist in increasing the presence of goat meat in the U.S. food

industry as well as impacting the goat meat industry.

Researchable Problem

The meat goat industry has experienced substantial gains in recent years due to

the increase in demand for goat meat products. The demand for goat meat products in

Florida is expected to continue growing as target populations within the niche market

increase. Researchers believe Florida has the potential for a meat goat market that is

profitable. However, lack of consumer information has hindered producers', processors',

and marketers' ability to delineate fully the profitability and viability of Florida's meat

goat industry. Consumer information is a critical component to completely

understanding the possible economic impact of goat meat productions and marketing in

Florida.

A relatively thin body of empirical research is available that evaluates the impact of

demographic and socioeconomic characteristics on consumers' preferences and

perception for goat meat products. This research will assist in filling the voids in

research by assessing the factors that influence the willingness to try goat meat within the

Hispanic and general markets. This study will identify the barriers that restrict

consumption and make recommendations for opportunities to increasing consumption of

goat meat products within the state of Florida.

Objectives

The overall objective of this research project is to identify the factors that

influence and the barriers that reduce consumption, with a special focus on the Hispanic

population. We will collect information for existing and potential goat meat consumption









in Florida and study the effects of perception, consumer preferences, and demographic

and socioeconomic characteristics on willingness to try goat meat. The specific objectives

for this project are:

* To develop an understanding of the factors that influences the willingness to try
goat meat products for Hispanics and non-Hispanics.

* To disseminate the results from this study and make recommendations for
opportunities of increased consumption of goat meat products to producers and
potential producers.

Hypotheses

Hispanics are the primary consumers of goat meat products and compose the

majority of this developing niche market. There are numerous factors that impact

preferences, perceptions, and willingness to try goat meat. In previous red meat studies,

attributes other than price found to influence a consumers' decision to buy the product

included safety, availability, advertisement, and government labels (Hui et al., 1995).

Thus, these attributes may also be important in the consumer purchase decision of goat

meat. Given these considerations, several hypotheses will be tested. The hypotheses are:

* Increases in household size and educational attainment will positively affect the
consumers' willingness to try goat meat.

* Respondents that associate a positive product image with goat meat will be more
inclined to express a willingness to try the product.

* Consumers that rate safety, convenience and cholesterol and fat content as
important factors in purchasing meat will be more likely to try goat meat.

* Ethnicity, age, and gender will influence the probability of trying goat meat. For
instances Hispanics are more likely to try goat meat whereas females are less likely
to try goat meat.

* Consumers that rate price specials as important factors in purchasing meat will be
least likely to try goat meat products.

* Lamb consumption will have a positive influence on the willingness to try goat
meat.






10


* Hispanics from different geographic origins will have differing levels of
willingness to try goat meat.














CHAPTER 2
LITERATURE REVIEW

The body of literature regarding the marketing of goat meat is relatively thin.

Numerous sensory evaluation studies have been conducted revealing the overall

palatability of goat meat is acceptable among consumers, but their lack of knowledge

about the product limits the demand and decreases its profitability within the meat

industry (Degner, 1991; James and Berry, 1997; Rhee et al., 2003; Smith et al., 1974).

Few studies have focused on the factors that influence goat meat consumption, and none

of the available research appears to investigate the attributes that were important to goat

meat consumers. Understanding consumers' preferences for goat meat is essential so that

marketing strategies can be developed to increase the demand for the product.

Two previous studies, McLean-Meyinsse (2003) and Degner and Lin (1993)

examined consumption of goat meat in the South. McLean-Meyinsse (2003) identified

that demographic, socioeconomic, and geographic factors that influenced previous

consumption, willingness to try goat eat, and interest in purchasing various goat meat

products. The study's sample consisted of 1,421 respondents from Alabama, Arkansas,

Florida, Georgia, Kentucky, Louisiana, Mississippi, North Carolina, Oklahoma, South

Carolina, Tennessee, Texas, and Virginia. A binomial logit was used to estimate the

relationship between prior consumption and the selected explanatory variables.

Additionally, ordered probit models were used to estimate the probabilities of

nonconsumers' willingness to eat goat meat and the likelihood of this individual to make

future purchases for goat nuggets, patties, roast, or marinated ready-to-cook goat meat









products. The study found that goat meat consumption was the highest among older

respondents, households with more than three persons, among African-Americans, other

non-Caucasian races, men and Texas residents. According to the marginal probabilities,

individuals from other races were 15 percent more inclined to consume goat meat

products than their Caucasians counterparts and women were 14 percent less likely than

men to have previously consumed goat meat.

The results from the ordered probit suggested that amongst non-triers, women and

residents of Arkansas, Florida, Kentucky, Louisiana, North Carolina, Oklahoma, South

Carolina, and Virginia were more likely to consume goat meat within the next month.

Younger consumers, households containing less than three persons, African-Americans,

or other races were less likely to try goat meat within the next month. Women and

residents from Florida, Kentucky, North Carolina, Oklahoma, South Carolina, and

Virginia were more willing to try goat meat at restaurants. Respondents least likely to try

free goat meat samples in supermarkets were younger, lived in smaller households, and

were non-Caucasian.

In summary, age, race, household size, religion, gender, and state residency were

found to affect consumption. Consumers most likely to eat goat nuggets, patties, or

roasts were individuals from larger households, non-Caucasians, men, or Texas residents.

Additionally, respondents living in larger households, other races, men or Texas residents

were more likely to be willing to purchase marinated, ready to cook goat meat.

The results of McLean-Meyinsse (2003) research were consistent with the

outcomes of a study conducted by Degner and Lin (1993) that analyzed willingness to

consume goat meat at restaurants or home. Data for their study consisted of responses









from consumer surveys conducted in Jacksonville and Tampa, Florida. Three-hundred

interviews were conducted in each city and participants were 18 years of age or older. A

probit model was utilized to evaluate the impact of demographic, geographic, and

socioeconomic factors on the willingness to consume goat meat at home and in

restaurants. The study found that if the respondent possessed a positive attitude or

perception of the product, they were more likely to order goat meat at a restaurant or

purchase the meat for at home consumption. The study found that household income,

gender, and household size effected participants willingness to consume goat meat.

Respondents from households with annual incomes of less than $10,000 were more likely

to order goat at a restaurant than individuals from households with a yearly income

between $10,000 and $19,000. Also, consumers between the ages of 35-49 were more

inclined to order goat meat in a restaurant than any other age group, but age had no

significant affect on purchase intentions.

An empirical study conducted by Hui et al. (1995) rated the importance of 12

selected meat attributes among various demographic, geographic, or socioeconomic

characteristics. The data for the survey were obtained using a telephone survey that was

administered to 1002 randomly selected households in Louisiana and Texas. The

primary shopper of each household was interviewed. The Kruskal-Wallis test was used

to determine whether the level of importance for each attribute differed amongst the

respondents. The simultaneous multiple comparison model ranked the attributes in order

of importance. An ordered-probit model was used to estimate the impact of the

consumer's demographic, geographic, and socioeconomic characteristics on the

importance of each attribute. The dependent variables were the 12 selected attributes,









which included low fat content, low sodium content, low in cholesterol, lack of chemical

additives, taste, red meat, white meat, appearance, price, freshness, USDA labels, and

tenderness and their relative importance while the dependent variables consisted of the

demographic, geographic, or socioeconomic characteristics.

The results from the ordered-probit model suggest that older consumers were more

concerned with prices, a stated preference for red meat, appearance of meat, USDA labels

and tenderness of meat. Larger households considered low sodium levels and lack of

chemical additives as influential attributes. Low-income households indicated that low

sodium content and red meat as valuable characteristics that influence meat purchases.

Respondents from high-income households were more concerned about sodium levels

and USDA labels and were worried least about fat, cholesterol, and prices.

The results from the Hui's (1995) study suggest that freshness and taste were the

most important attributes to consumers, followed by appearance. The next tier of

attributes consumers deemed important by included USDA labels, tenderness, and lack of

chemical additives. The last group consisted of nutritional attributes, which included low

levels of fat, sodium, and cholesterol. The results indicated that females were more

concerned than males about attributes such as fat, sodium, cholesterol, chemical

additives, prices, appearance, freshness, tenderness, and USDA labels when making

buying decisions. Non-white respondents were more worried about fat, cholesterol, and

price than white respondents.

According to the results, retailers, wholesalers, and processors should develop a

marketing plan that emphasizes the tastiness, appearance, and freshness of the meat and

include recipes when promoting meats. In addition, the marketing channels should









minimize transportation and holding time to ensure freshness. This study provided

knowledge on the relationship between consumer characteristics and the significance of

various meat attributes that may assist in creating effective marketing opportunities for

farmers, processors, wholesalers, and retailers in the meat industry.

Melton et al. (1996) conducted experimental auctions to evaluate the significance

of attributes and how to develop effective marketing plans for pork. The willingness to

pay results suggests that the appearance of the meat is most important for first-time

buyers and repeat purchasers were interested in the pork chop's taste. Melton et al.

(1996) concluded that first-time buyers of fresh pork chops may be misled by relying on

appearance when making purchases, selecting chops that were less desirable when eaten.

As a result, theses consumers were unlikely to make repeat purchases, hampering the

product's long term market success.

Similar to the studies conducted by Hui et al. (1995) and Melton et al. (1996), Chen

et al. (2002) examined the relative importance of fresh pork attributes among Asian-

origin consumers in San Francisco, California. The results from the Kruskal-Wallis test

suggest that freshness is the most important attribute followed by attributes of the color of

meat, lowness in fat, and whiteness of fat. The price of fresh pork is also an attribute of

considerable importance. The empirical results for the ordered-probit model indicate that

particular demographic and socioeconomic characteristics influence the ranking of

attributes among consumers. For instance, fat content was more important to highly

educated males. They study also found differences within segments of Asian-origin

populations; for example, Chinese origin respondents were more price sensitive than

other Asian consumers. This final finding is important because one can infer that









Hispanic origin consumers were heterogeneous group and that the significance of various

product attributes may differ amongst each identifiable subgroup.

In the United States, goat meat is viewed as specialty food item. The study

conducted by Schupp et al. (1998), found that consumers expressed resistance to meats

that they believed came from exotic animals. Species that respondents in the study

considered exotic included but were not limited to deer, alligator, rabbit, goat, emu, and

wild duck. Therefore, meat attributes that were deemed significant to consumers may

differ for alternative meats.

McLean-Meyinsse (2000) conducted a study that utilized ordered probit models to

evaluate the impact of socioeconomic, demographic, and geographic characteristics on

the primary grocery shoppers' attitude toward rabbit; and previous consumption or intent

of consuming the product. The data for the study were obtained from a random telephone

survey that sought information regarding the meat purchasing and consumption

decisions. The sample of consisted of 1,002 primary food shoppers from Louisiana and

Texas. As previously mentioned, rabbit meat and goat meat both tend to be considered

exotic amongst consumers.

The independent variables in the study were various socioeconomic, demographic,

and geographic characteristics such as age, income gender, educational attainment, and

race, to name a few. McLean-Meyinsse (2000) used an ordered probit model to determine

whether the explanatory variables effect the probability of shoppers' attitudes towards

rabbit meat and the likelihood of consumption or interest in eating it in the future

consumption. The results from the attitudinal model suggest that gender, religion, and

employment status have a statistically significant effect on shoppers' attitude towards









rabbit meat. The marginal effects suggest that men, Catholics, and white-collar workers

are more positive about goat meat than their counterparts. Forty-eight percent of male

shoppers possessed a positive opinion about rabbit meat, while 54 percent of women held

a unfavorable opinion about the meat. The consumption model suggests that gender and

employment status impact the likelihood of consuming or interest in consuming rabbit

meat. For instance, if the grocery shopper was a male rather a female, the probability of

rabbit meat consumption increased by 12.33 percentage points. Females were more

willing to express a willingness to try rabbit meat. Also, if shoppers had a positive

opinion about rabbit meat, they were more likely to have consumed it.

McLean-Meyinsse (1999) also evaluated the marketing outlook for specialty meats

such as alligator, goat, and/or rabbit meat in southern states. The objectives of the study

were to determine the percentage of individuals shopping at outlets offering specialty

products, identify the demographic, socioeconomic, and geographic characteristics

influencing shoppers' likelihood of buying from these stores, and profile of the

consumers most likely to purchase specialty meats. The study analyzed various

explanatory variables (i.e. age, household size, education, gender, income, martial status,

religion and occupation) that were expected to affect likelihood of non-triers, late-triers,

and early triers of specialty foods to shop at specialty stores using an ordered probit

model. According to the results, a 42-year old grocery shopper or a shopper from a three

person household was more likely to shop at stores offering specialty meats. Shoppers

with less than a high school diploma were 18 percent more likely visit these outlets than

individuals with higher level of education. Shoppers indicated that price effected their

decisions to make purchases at stores offering specialty meats. In fact, there was a 22









percent difference between the early triers and nontriers on the relevance of prices to the

meat purchasing decision. It is worth mentioning that the price of specialty meats is

usually more expensive than traditional meats such as beef, pork, and chicken.

In conclusion, very few empirical studies that focus on goat meat consumption are

available in literature. Existing studies conducted by McLean-Meyinsse (2003) and

Degner and Lin (1993) identify target consumers for goat meat marketing and

promotional efforts, but the studies stop short of identifying the underlying factors that

encourage individuals to purchase or be willing to purchase the product. The meat goat

industry is still in its developing stages, and before this industry experiences economic

success, additional information explaining the demand characteristics for goat meat

consumers is needed. Understanding these factors, as well as consumers' preferences and

perceptions towards goat meat, is important when developing marketing strategies and

increasing the presence of goat meat in supermarket.














CHAPTER 3
SURVEY CONTENTS

Survey Instrument

Telephone surveying is a method of collecting data from respondents that is more

cost efficient than conducting personal interviews (Dillman, 1978). The primary

advantage of telephone surveying is that it provides the researcher the opportunity of

controlling and monitoring the data collection process to ensure that data gathered is of

high quality, thus providing accurate estimates. For instance, the researcher has the

ability to regulate sampling, respondent selection, and questionnaire contents (Lavarakas,

1993). Telephone surveying is more cost efficient when compared to personal

interviews because it allows a larger number of interviews in a short period of time.

Telephone surveys are usually more expensive than mail surveys but due to the fact that

telephone surveys have the potential of minimizing total survey error, this method is

generally preferred. Additional advantages associated with telephone surveys include:

(1) the scheduling of call-backs to contact hard to reach, but critical respondents, (2) the

ability to minimize biased responses, and (3) the capability of the interviewer to clarify

questions (Lavarakas, 1993).

Despite the advantages associated with telephone surveying, a few disadvantages

exist. A major drawback with the telephone surveying technique is that surveyors are

unable to reach the cell-phone-only population, making the sample statistically

unbalanced because it does not contain both cell phone and land line users. The

Associated Press (2005) found that the cell-phone-only population is growing rapidly and









it currently accounts for approximately 7 percent of the population. In fact, nearly one in

every five individuals between the ages of 15 and 24 had only cell phones. Other

limitations are related to the limitations with complexity of the questions asked and the

length of the survey. Respondents may grow tired if kept on the telephone for longer

than 20-30 minutes. However, respondents tend not to suffer from fatigue when

participating in personal interviews and this issue is not applicable to mail surveys

because the questionnaire is completed at the respondent's leisure (Lavarakas, 1993).

Similarly, complicated questions are impossible to ask via telephone. In spite of the

disadvantages associated with this surveying technique, the advancements in telephone

technology and infrastructure give the researcher accessibility to nearly any population

via telephone, making this surveying approach more attractive than other methods (Frey,

1989).

The Institute for Behavioral Research, Survey Research Center at the University of

Georgia has a history of successfully conducting telephone surveys. This center was used

to collect the data for this study. The trained research staff utilized the random digit

dialing technique ensuring that all adult Florida residents with landline telephone service

had an equal chance of selection for inclusion in the sample, regardless of whether the

number is unlisted, which reduces the sampling error (Salant and Dillman, 1994).

The telephone survey in this study was administered twice, first to the general

population and secondly to Hispanic households. An emphasis was placed on Hispanic

households because previous research suggested that these individuals provide a steady

year round demand for goat meat products (Spaugh, 1997). Two hundred thirty-seven

households participated in the general population survey. For the latter survey, a









telephone directory data base was obtained that consisted of households with Hispanic

surnames and it consisted of 198 observations. The surveyed Hispanic households were

randomly selected from the directory database. The original survey was translated to

Spanish to better serve the Hispanics consumers.

Once the potential participants were contacted, the objective of the study was

explained to the respondents and they were asked to participate in the study. The

surveyor spoke with the primary grocery shopper as they were expected to be primary

decision maker in purchasing goat meat. The respondent was asked to complete the

survey, which sought information on demographics and consumers shopping preferences

for goat meat products.

Survey Contents

The purpose of the telephone survey was to collect information on consumer

shopping preferences for goat meat. The survey sought three categories of information:

(1) demographics, (2) family linkages with an emphasis on the introduction to goat meat

products and the transmission of food habits from generation to generation including

consumption levels, and consumer perception and (3) identifications of factors

influencing consumption or the willingness to consume. All survey participants had to be

18 years old or older. If the respondent was unavailable to complete the interview, a

callback was scheduled. The questionnaire used in this study is included in the

Appendix.

The dependent variable in this study was the willingness to try goat meat. To

estimate this variable, respondents were placed in one of three categories: respondents

that had previously consumed goat meat, respondents willing to try, and respondents that

have not consumed and are unwilling to try goat meat. Respondents were first asked if









they had ever consumed goat meat. If the respondents had previously consumed goat,

they were then asked a series of questions regarding the size, age, cut, etc of the meat

consumed. These questions were asked to develop an understanding of the type of

product desired by different ethnic groups. Previous research suggested Hispanics,

Muslims, African-Americans, Haitians, and Jamaicans consume goat meat product, but

carcass preferences differs among each group. To estimate current consumption levels,

respondents that had previously consumed goat were asked to quantify how many pounds

their family consumed each year.

To identify potential consumption, respondents that had never consumed goat meat

were asked if they were willing to consume if it was available in stores. Research

suggests that availability is a major obstacle faced by the goat meat industry, thus,

restricting consumption. Locascio and Degner (1988) surveyed supermarket

representatives in Florida and found that 28 of 168 orl6.7 percent, of the stores run by six

chains sold goat meat. A more recent study conducted by Degner and Moss (1999) found

that 18 percent of meat wholesalers in Florida sold goat meat. As previously mentioned

the dependent variable for the study is the willingness to try goat meat; therefore,

respondents that had previously consumed or willing to try if available in grocery stores

are considered those willing to try goat meat and all other respondent are non-triers.

The next section of the questionnaire solicited information on the psychographic

factors that consumers' believed to have importance when making decisions to purchase

meats. Previous meat studies indicated freshness, convenience, and sodium and

cholesterol content influence meat consumption. However, because goat meat is a

specialty food item, it is unknown if these characteristics are equally important to goat









consumption. Psychographic factors are needed because they provide information

necessary to develop an understanding of how of consumers feel and think about the

product (Peter and Donnelly, 2003). Respondents were asked to rate the relative

importance of factors such as price specials, convenience, safety, cholesterol, and fat to

meat purchases. Participants were also asked how they viewed goat meat. The responses

could have ranged from very positive to very negative. It was expected that this portion

of the survey would reveal consumers knowledge about goat meat.

Information regarding respondents' consumption of other meats such as chicken,

beef, and seafood was also collected because according to demand theory, consumption is

expected to be affect by substitutable and complementary products. Participants were

asked the frequency and then quantity of the alternative meat consumed. First,

respondents were asked to specify if the selected food item was consumed everyday,

more than once a week, once a week, more than monthly, monthly, on special occasions,

or never. If the respondent indicated he/she had consumed the meat, respondent was then

asked to identify the quantity consumed per sitting. When consumers make a decision to

try or consume a product, they usually evaluate the alternative products available. This

study incorporated a variable that captured the consumption of other meats to evaluate

the relationships between other selected meats and the willingness to try goat meat.

The final section of the questionnaire solicited socio-economic and demographic

information, such as household income, household size, educational attainment, gender,

and race. Based on previous empirical studies, this information is expected to be

significant when analyzing the factors that influence goat meat consumption.














CHAPTER 4
DATA

The data for this study were attained through a telephone consumer survey in

conjunction with the SARE #LS502-138, An Investigation of the General Goat Meat

Demand and the Sustainability of Goat Production. Adult consumers within the state of

Florida were surveyed via telephone to establish consumer preferences for goat meat and

the level of consumption and potential levels of consumption. The Survey Research

Center at the University of Georgia collected the data during the Spring of 2004. Florida

was targeted because the strongest demand for meat goats is found along the East Coast,

especially within Florida (McKenzie-Jakes, 2004). In addition, Hispanics were targeted

because these consumers are expected to provide a more stable demand for goat meat

products throughout the year (Spaugh, 1997). Thus, understanding the factors that effect

their demand for goat meat is imperative.

Florida households were surveyed utilizing random digit dialing to assess the

factors influencing their consumption decisions of goat meat. As previously mentioned,

telephone surveying is an effective method of data collection that allows a large amount

of data to be collected in a short period of time. Additionally, random digit dialing

ensures that all residents with a landline telephone have an equal chance of being

included in the sample. Selection response is minimized, and inferences about Florida's

adult population can be made with greater assurance from the results obtained in the

survey. Four hundred thirty-five consumers participated in the survey, and data is

summarized in Appendix B. However, due to incomplete demographic and









socioeconomic information such as age, educational attainment, and gender, sample

observations were deleted. Thus, the information from 365 responses is summarized in

this analysis.

When the sample results are compared with US Census data for Florida (US

Census Bureau, 2000) Hispanics were over represented within the sample, accounting for

41.4 percent while making up only 16.1 percent Florida's population. However,

Caucasians and other races were underrepresented in the sample making up 50.1 percent

and 15.5 of the sample, respectively. An emphasis was placed on Hispanic consumers

because these individual are perceived to have a historical preferences towards goat meat;

therefore, the results were expected to be biased towards Hispanics. Among the Hispanic

respondents, few inconsistencies exist between their geographic origins for their

respective populations and sample, making the two reasonably comparable. For instance,

Cubans represent 31.1 percent of the Hispanic population in Florida and 25.8 percent of

the sample, Mexicans account for 13.6 percent of the population and 16.6 percent of the

sample, Puerto Ricans represent 18 percent of the population and 16.6 percent of the

sample, and all other Hispanics descents account for 37.4 percent of the population and

41.1 percent of the sample.

The majority of the survey respondents were female, 70.1 percent. The results are

biased towards females because the questionnaire targeted the primary grocery shopper

as a means of accurately estimating consumption history, the willingness to try, and

psychographic characteristics sought by consumers and potential consumers of goat meat.

According to literature, 70 percent of all females in the United States were considered the

primary grocery shopper (Progressive Grocer, 2002).









Figure 4-1 illustrates the percentage of Florida's population and survey

respondents in various age groups. The sample tolerance for this study is +/- 5.1 percent.

With this in mind, the sample and population are comparable despite the small

discrepancies that exist. The sample's median age fell in the 35-44 age category,

comparing to the median age in Florida, 38.6. The sample's average household consisted

of three people, which exceeds the state's average household size of 2.46 individuals.

Additionally, the average household size for Florida's Hispanic population was 3.12

persons, slightly less than that of the sample, 3.32 persons. Finally, survey respondents

were more educated than average. According to survey response, 47.7 percent of the

respondents possessed either a high school diploma or lower, 26.5 percent had some

college education, and nearly 25.8 percent had at least a college degree, which compares

to the populations 49.8 percent, 29.6% and 20.6 % for the respective categories.

Likewise, 52.3 percent of the Hispanic respondents had more than a high school diploma,

only 40.1 percent of the Hispanic population had surpassed this mark. Furthermore, 25.8

percent of the Hispanic sample and 15.6 percent of the population had received degrees

from a four year college or more advanced degree (Figure 4-2).

Among the 365 respondents, 43.6 percent expressed a willingness to try goat meat

if it were available in food stores (159 respondents), whereas, 56.4 percent were

uninterested in the product (206 respondents). Male respondents were more likely to try

goat than females, 52.3 percent and 39.8 percent, respectively. Hispanics were more

likely than Caucasians, but less likely than other races to express a willingness to try goat

meat. Over 60 percent of respondents that perceived goat meat in a positive manner








indicated that they would try goat meat. However, only 6.1 percent of respondent with a

negative view of goat meat were willing to try the product.

Other explanatory variables incorporated in the study include the frequency at

which goat meat substitutes are consumed, psychographic characteristics, and various

demographic and socioeconomic variables. Descriptive statistics for the survey sample

are found in Table 4-2 and Table 4-3. Since a separate analysis involving Hispanic

respondents will be conducted, reasoning is explained in the next chapter, descriptive

statistics for only the Hispanic respondents are shown in Tables 4-4 and 4-5.


Figure 4-1. Comparison of Florida census data and survey respondents by age.


25%

20%

15%

10%

5%

0%


O Florida
u Survey


18-24 25-34 35-44 45-54 55-64 65
and
older






















HS Diploma or Some College
LT


O Florida
* Survey


College Grad &
Beyond


Figure 4-2. Comparison of Florida census data and Hispanic survey respondents by
educational attainment levels.

Table 4-2. Summary of demographic information.
Non- Overall
Triers Triers Sample
Number of Observations 159 206 365
Ethnicity % % %
Hispanic 45.9 37.9 41.4
Caucasian 44.0 54.9 50.1
Other races/ethnicities 10.1 7.3 8.5
Gender
Percent Female 64.2 74.8 70.1
Education
HS Diploma or less 37.7 37.9 37.8
Some College 26.4 25.2 25.8
College Degree and above 35.9 36.9 36.4
Household Size
1 Only 9.4 11.2 10.4
2 People 34.0 33.0 33.4
3 People 18.2 17.0 17.5
4 People 17.0 25.2 21.6
5 and above 21.4 13.6 17.0
Age of Respondents
18-24 7.6 8.7 8.2
25-34 23.9 17.5 20.3
35-44 13.8 19.9 17.3
45-54 18.9 27.5 18.1
55-64 18.2 12.1 14.8
65 and older 17.6 24.3 22.2









Table 4-3. Descriptive statistics on additional factors that influence the willingness to
try.
Triers Non-Triers Overall Sample
Frequency of Chicken Consumed % % %
More than once a week 78.0 79.1 78.6
Once a week 17.6 13.6 15.3
Special Occasions 4.4 5.3 5
Never 0.0 1.9 1.1
Frequency of Beef Consumed
More than once a week 52.8 49.0 50.7
Once a week 24.5 30.1 27.7
Special Occasions 18.9 15.1 16.7
Never 3.8 5.8 4.9
Frequency of Pork Consumed
More than once a week 23.3 17.0 19.7
Once a week 23.3 24.8 24.1
Special Occasions 41.5 40.3 40.8
Never 12.0 18.0 15.3
Frequency of Fish/Seafood Consumed
More than once a week 43.4 36.4 39.5
Once a week 34.6 23.3 28.2
Special Occasions 18.9 31.6 26.0
Never 3.1 8.7 6.3
Frequency of Lamb Consumed
Previously Consumed 51.6 25.3 36.7
Never 48.4 74.8 63.3
Frequency of Turkey Consumed
Previously Consumed 88.7 85.0 86.6
Never 11.3 15.1 13.4
View of Goat Meat
Positive 60.4 16.0 35.3
Neutral 33.3 38.4 36.2
Negative 6.3 45.6 28.5
Importance of Cholesterol to Meat Purchasing
Important 83.6 78.9 80.2
Unimportant 16.4 21.2 19.8
Importance of Fat in Purchasing Meat
Important 86.6 87.4 86.8
Unimportant 13.8 12.6 13.2
Importance of Convenience in Purchasing Meat
Important 67.3 61.7 64.1
Unimportant 32.7 38.4 35.9
Importance of Price to Meat Purchasing
Important 83.0 65.1 72.9
Unimportant 17.0 35.0 27.1
Importance of Safety to Meat Purchasing
Important 95.6 90.8 92.9
Unimportant 4.4 9.2 7.12











Table 4-4. Summary of demographic information for Hispanic respondents.
Overall
Triers Nontriers Sample
Number of Observation 73 78 151
Descent % % %
Mexican 17.8 15.4 16.6
Cuban 28.9 23.1 25.8
Puerto Rican 15.1 18.0 16.6
Other Descents 38.7 43.6 8.5
Generation in U.S.
First generation 65.8 60.3 62.9
Other generation 34.2 40.7 37.1
Gender
Percent Female 67.1 74.4 70.9
Education
HS Diploma or less 49.3 46.2 47.7
Some College 24.7 28.2 26.5
College Degree and above 26.0 25.6 25.8
Household Size
1 Only 5.5 9.0 7.3
2 People 24.7 21.8 23.2
3 People 17.8 24.4 21.2
4 People 23.3 29.5 26.5
5 and above 28.8 15.4 21.9
Age of Respondents
18-24 11.0 5.1 8.0
25-34 28.8 28.2 28.5
35-44 16.4 20.5 18.5
45-54 20.6 16.7 18.5
55-64 11.0 11.5 11.3
65 and older 12.3 18.0 15.2








Table 4-5. Descriptive statistics on additional factors that influence the Hispanic
respondents' willingness to try.
Triers Non-Triers Overall Sample
Frequency of Chicken Consumed % % %
More than once a week 84.9 83.3 84.1
Once a week 11.0 7.7 9.3
Special Occasions 4.1 5.1 4.6
Never 0.0 3.9 2.0
Frequency of Beef Consumed
More than once a week 54.8 38.5 46.4
Once a week 27.4 34.7 31.1
Special Occasions 16.4 15.4 15.9
Never 1.4 11.5 6.6
Frequency of Pork Consumed
More than once a week 24.7 18.0 21.2
Once a week 26.0 18.0 21.9
Special Occasions 34.3 32.0 33.1
Never 15.0 32.0 23.8
Frequency of Fish/Seafood Consumed
More than once a week 42.5 35.9 39.1
Once a week 37.0 26.9 31.8
Special Occasions 15.0 25.6 20.5
Never 5.5 11.5 8.6
Frequency of Lamb Consumed
Previously Consumed 42.5 24.4 33.1
Never 57.5 75.6 66.9
Frequency of Turkey Consumed
Previously Consumed 79.5 69.2 74.2
Never 20.5 30.7 25.8
View of Goat Meat
Positive 72.6 29.5 50.3
Neutral 23.3 39.7 31.8
Negative 4.1 30.8 17.9
Importance of Cholesterol to Meat Purchasing
Important 93.2 88.5 90.7
Unimportant 6.8 11.5 9.3
Importance of Fat in Purchasing Meat
Important 89.0 88.5 88.7
Unimportant 11.0 11.5 11.3
Importance of Convenience in Purchasing Meat
Important 72.6 60.3 66.2
Unimportant 27.4 40.7 33.8
Importance of Price to Meat Purchasing
Important 87.7 62.8 74.8
Unimportant 12.3 37.2 25.2
Importance of Safety to Meat Purchasing
Important 98.6 92.3 95.4
Unimportant 1.4 7.7 4.6














CHAPTER 5
THEORETICAL MODEL AND MODEL SPECIFICATIONS

Theoretical Model

Neoclassical demand theory indicates that the determinants of demand may be

categorized under four headings: (1) population size and its distribution by various

demographic and geographical characteristics, (2) consumer incomes, (3) prices and

availability of complements and substitutes, and (4) consumer tastes and preferences.

The homogeneity condition provides a theoretical basis for consumer behavior. The

condition states that the sum of the own-price elasticity, cross-price elasticity and the

income elasticity for a given commodity equals zero (Tomek and Robinson, 1990):

Eii + Eil + Ei2+ ...+ Ey = 0,

where Eii is the own price elasticity, Eil ... Ei2 are the cross-price elasticities and Eiy is the

income elasticity. The condition implies that the substitution and the income effect on

own-price change must be consistent with the cross-price and income price elasticites for

a particular commodity.

Goat meat prices are very volatile and have been unavailable for many years;

therefore, this research does not attempt to estimate demand nor calculate the elasticities.

The prices of substitutes were not included in the study; however, information about the

relationship between goat meat and its substitutes may give insight on the cross-price

elasticites. The central focus of this study is to develop an understanding of factors

influencing goat consumption and the willingness to try goat meat. Sensitivity of

willingness to try goat meat with price and income is assessed in qualitative terms.









This study will examine factors influencing willingness to try goat meat among

Hispanic and non-Hispanic consumers using data obtained from a telephone survey

administered in Spring 2004. The survey instrument has been discussed in the previous

chapter. This study utilizes a probit analysis to estimate the factors that influence the

dependent variable, willingness to try goat meat. Since research suggested Hispanics

have a historic preference for goat meat, two models are used: (1) an estimation of factors

that influence consumption for the entire sample and (2) one that evaluates the factors

that influence the willingness to try goat meat among Hispanics.

Probit Model

Due to their popularity, linear regressions models may be one of the most misused

analytical techniques in the social sciences (Aldrich and Nelson, 1984). Linear

regression models assume that the dependent variable is continuous; therefore, when the

endogenous variable is qualitative, the estimates from the regression analysis may be

robust in errors, causing inaccurate statistical inferences (Aldrich and Nelson, 1984). The

endogenous variable in this study is a yes or no variable, the willingness to try goat meat.

When the regressand is discrete rather than continuous a different analytical technique is

needed.

Probit models estimate the probability of the binary dependent variable, y,

occurring given K observable, explanatory variables, k = 1,..., K. Each of the

observations on y, yi, y2,... ,yN, are statistically independent of each other, ruling out

serial correlation. Additionally, the model assumes that data are generated from a random

sample of size of N observations, with each sample point indicated by i, i = 1,..., N

(Aldrich and Nelson, 1984). Probit analysis requires that there is no exact linear









dependence among 4ik's. This implies the number of observations exceed the number of

explanatory variables, N>K, that there is variation among each explanatory variable

across the observations, and that no two or more 4ik's are perfectly correlated. The

expected outcomes of the dependent variable, yi, are considered to be mutually exclusive

and exhaustive (Gujarati, 2003).

Probit models assume the dependent variable depends on a latent variable, yi*,

which is and observed and determined by one or more independent variables.

yi* =xi' + Ei, si ~ NID(0,o2)

The larger yi*, the greater the probability of an event, y, occurs. An event is assumed to

occur if the utility differences exceed a certain threshold level. Probit analysis follows a

cumulative normal probability distribution with the same mean and variance, providing

information on the nature of the latent variable and its parameters. The dependent

variable, y, may take on values of zero or one and if the latent variable is defined as y*,

then the probit model is described as follows:

yi* = i'1 + Ei, si ~ NID(0,o2)

yi= 1 ify* > 0

= 0 ify* <0,

The point of interest relates to the probability of the event occurring, Y=l. Utilizing the

information above, we have:

P(yi = 1) = P(yi* > 0) = P(i'P1 + Ei > 0) = P(i < i' 3) = D(i'P1),

where D denotes the cumulative distribution of si (Verbeek, 2004).

Maximum likelihood estimation techniques are used to obtain the value of the

parameters, p, that maximize the probability of observing the outcome, y. The maximum









likelihood estimation model is nonlinear and asymptotic, producing better results as the

sample size increases. It produces estimates that are nonbiased (estimates are centered

around the true values on average), efficient (no other unbiased estimator has lower

sampling variance) and normal (we can know how to perform hypothesis testing and

draw other inferences) (Aldrich and Nelson, 1984). Maximizing and then taking the first

derivative of the log likelihood function produce the parameters for each explanatory

variable. The log likelihood function is as follows:

log L(P) = Y yi log F(Li'P) + E (1 yi) log yi log F(1 (i')),

where p is included in the probabilities to accentuate that the likelihood function is a

function of p. The parameters derived from the log likelihood function are known as

marginal effects or marginal probabilities. The marginal probabilities measure the change

in probabilities resulting from a unit change in one of the regressors while holding the

other regressors constant. Predicted marginal probabilities assist in understanding the

relationship between the dependent and independent variables and the signs of the

parameter estimates and their statistical significance indicate the direction of the

relationship (Gujarati, 2003; Verbeek, 2004).

A goodness of fit measure is a summary statistic suggesting the accuracy with

which the model approximates the observed data (Verbeke et al., 2000). When the

dependent variable is qualitative the accuracy of the model is determined by comparing

the fit between the calculated probabilities and observed response frequencies or through

the model's ability to forecast observed responses. Goodness of fit measures are usually

based on a comparison between a model that contains only a constant as the independent

variable. Thepseudo R2 takes into account the two likelihood values, log L1 and log Lo,









where L1 represents the maximum log likelihood value of the model of interest and Lo

stands for the maximum value of the log likelihood function when the intercept is the

only parameter value that is not equal to zero. The difference between the log L1 and log

L2 serves as an indicator of the explained variation of the underlying latent variable

caused by the additional parameters (Laitila, 1993; Verbeek, 2004). In summary, the

pseudo R2 is a tool used to evaluate the explained variation in a model. It is important to

mention that this measure has two shortcomings: (1) the pseudo R2 usually decreases as

additional parameters are included in the model and (2) the measure does not adjust for

the degrees of freedom of the model (Laitila, 1993).

Model Specification

Existing empirical studies provide the basis for the variables selected in the

model. Earlier studies conducted by McLean-Meyinsse (2003) and Degner and Lin

(1993) used a probit analysis to evaluate the factors that influence consumers' willingness

to consume goat meat and goat meat products. Their studies indicated that race, age,

household size, geographical location, and gender affect the willingness to consume or

try goat meat. For example, the studies found non-Caucasians, men, and those living in

larger households were most likely consumers of goat. Studies conducted to assess the

factors that influence the consumption of specialty meats McLean-Meyinsse (1999) and

McLean-Meyinsse (2000) found ethnicity, education, household size, and gender

influenced consumers' attitude toward exotic animal food item and their willingness to

consume. Finally, studies by Hui et al. (1995) and Chen et al. (2002) used probit model

simulations evaluated the impact selected meat attributes had on meat consumption

among various demographic, socioeconomic, and geographical characteristics. The

results from the Hui et al. (1995) study suggested that female and non-white consumers









are more concerned with fat, cholesterol, and price. Chen et al. (2002) suggested

segments within Asian populations have specific taste and preferences and are not a

homogenous group.

In this study, probit models are used to estimate the willingness to try goat meat

for Hispanic and non-Hispanic populations with respect to several explanatory variables.

The entire survey sample model will included demographic and socioeconomic factors

(i.e. age, gender, income, household size, and educational attainment), perception of goat

meat, frequency of other meat consumption, and consumer characteristics. It is believed

that segments within the Hispanic population have different variables affecting their

willing to try goat meat. Additional independent variables that will be included only in

the Hispanic model are descent and generation in the U.S. Specification of the probit for

the entire survey sample is as follows:

Yki* = PklETH2 + Pk2 ETH3 + Pk3 GENDER + Pk4 AGE1 +
Pk5 AGE2 + Pk6AGE3 + Pk7 AGE4 +pk8 AGES + pk9 HSIZE1 +
PkloHSIZE2 + pkll HSIZE3 + pk12 HSIZE4 + Pk13 BEEF1 +
Pk14 BEEF2 + Pk15 CHICK 1 +pk16 CHICK2 +pkl7FISH1 +
Pkl8FISH2 + pk19 PORK1 + Pk20PORK2 + Pk21TURK +
Pk22LAMB + pk23VIEW1 +Pk24VIEW2 + Pk25FAT +
Pk26SAFETY + pk27CONVEN +Pk28PRICE
Specification of the Hispanic model is as follows:

Yji* = PjiMEXICAN + pj2 PTRICAN + Pj30THERD +
pj4 GENERA + Pj5 AGE1 + Pj6AGE2 + Pj7AGE3 + Pj8AGE4 +
Pj9AGE5 + PjloHSIZE1 +Pj3nHSIZE2 + Pj12HSIZE3 +
Pj13HSIZE4 + pj14BEEF1 +Pj15BEEF2 + Pj06CHICJ1 +
Pj17CHICJ2 +Pj3sFISH1 + Pj19FISH2 + Pj2oPORJ1 +
Pj21PORJ2 + Pj22TURJ +Pj23LAMB + Pj24VIEW1 +
pj25VIEW2 + Pj26FAT +pj27SAFETY + Pj28CONVEN +Pj29PRICE






38


Yi= J if respondent is willing to try goat meat
L0 if respondent is unwilling to try goat meat
The probit model estimates the influence the selected explanatory variables have

on consumers' preferences of goat meat. The analysis also predicts the probabilities of

the consumers' willingness to try goat meat under several variable levels. A description

of the variables used in this study can be seen in Table 5-1.








Table 5-1. Probit model variables and description.


Variant
Willingness to try goat
meat
Ethnicity
Hispanic Origin



Gender
Age
Education


Household Size




Perception of Goat
Meat


Consumer Attributes




Substitutes Consumed


Variable NameDescription


TRY
ETH
MEXICAN
CUBAN
PTRICAN
OTHER
GENDER
AGE1
EDU1
EDU2
EDU3
HSIZE1
HSIZE2
HSIZE3
HSIZE4
HSIZE5

VIEW
VIEW2
VIEW
FAT
SAFETY
CHOLES
CONVEN
PRICE
BEEF1
BEEF2
BEEF3
BEEF4
CHICK 1
CHICK2
CHICK3
CHICK4
LAMB1
PORK
PORK2
PORK3
PORK4
TURK1


1 if willing to try goat meat
1 if Hispanic, 0 otherwise
1 if of Mexican, 0 otherwise
1 if of Cuban, 0 otherwise
1 if of Puerto Rican, 0 otherwise
1 if of Mexican, 0 otherwise
1 if male,
1 if 35 or older, 0 otherwise
High School diploma or less
Some College
College 4 year degree and beyond
1 if one person
1 if two people
1 if three people
1 if four people
1 if five or more people

1 if positive view
1 if neutral view
1 if negative view
1 if fat is important to the consumer
1 if safety is important to the consumer
1 if convenience is important to the consumer
1 if convenience is important to the consumer
1 if price specials are important to consumer
1 if consumed more than once a week
1 if consumed once of week
1 if consumed during monthly or less frequently
1 if never consumed
1 if consumed more than once a week
1 if consumed once of week
1 if consumed during monthly or less frequently
1 if never consumed
1 if consumed, 0 otherwise
1 if consumed more than once a week
1 if consumed once of week
1 if consumed during monthly or less frequently
1 if never consumed
1 if consumed, 0 otherwise














CHAPTER 6
EMPIRICAL RESULTS

Using data collected from a consumer survey, the specification set forth in the

previous chapter and maximum likelihood procedures, two independent probit models

were estimated with the dependent variable representing the consumers' willingness to

try goat meat. Due to the nature of the dichotomous dependent variable, a probit analysis

was utilized to predict the likelihood of trying goat meat given various exogenous

variables. The probit model coefficients and marginal probabilities from the two models -

one for the survey population and one for only Hispanic respondents are shown in Tables

6-1 and 6-2, respectively. According to Greene (2003), marginal probabilities should be

used to draw inferences about the relationship between the dependent and independent

variables rather than coefficient estimates. The marginal probabilities measure the

change in probability of the willingness try goat meat from a unit change in one of the

explanatory variables, while holding the other regressors at their sample means. The

results from these models are discussed in this chapter, beginning with the whole

population model and followed by the Hispanic only model.

Entire Survey Sample Probit Estimates

The entire survey sample probit model (Table 6-1) correctly predicted 75.8

percent of consumers' responses (incorrectly predicting both a consumers' willingness to

try goat meat and non-willingness to try 12.1 percent of the time). This compares to a

naive, which resulted in correct prediction 56.4 percent of the time. The chi-squared









value is 137.0 is statistically significant to the .01 confidence level, which implies good

predictive power of the variables included in the model.

Table 6-1. Empirical results from the whole survey probit model.

Variable Estimated Coefficient Standard Error Marginal Effects
ETH2 0.0554 0.2020 0.0217
ETH3 0.0424 0.3031 0.0165
GENDER -0.2985 0.1783 -0.0830
AGE1 0.0404 0.3484 0.0159
AGE2 0.0862 0.2719 0.0317
AGE3 -0.0493 0.2898 -0.1922
AGE4 0.0780 0.2536 0.0307
AGE5 0.4215 0.2606 0.1668***
EDU2 0.0883 0.2006 0.0347
EDU3 0.0761 0.3031 -0.2973
HSIZE1 -0.6960** 0.3234 -0.2235**
HSIZE2 -0.4697*** 0.2451 -0.1790**
HSIZE3 -0.5380** 0.2606 -0.1984**
HSIZE4 -0.6253* 0.2508 -0.2295*
VIEW1 1.6244* 0.2286 0.5829*
VIEW2 0.7787* 0.2070 0.3019*
FAT -0.5495** 0.2818 -0.2165**
SAFETY -0.2617 0.3042 -0.1037
CHOLES 0.1972 0.2471 0.0760
CONVEN -0.0671 0.1691 -0.0263
PRICE 0.5373* 0.5376 0.2017**
BEEF1 -0.1805 0.2132 -0.0709
BEEF2 -0.2505 0.2305 -0.0967
CHICK1 -0.7486** 0.3215 -0.2918*
CHICK2 -0.7544** 0.3728 -0.2662**
PORK1 0.0635 0.2136 0.0249
PORK2 -0.1158 0.1955 -0.0451
FISH1 0.3839** 0.1951 0.1505**
FISH2 0.4458** 0.2103 0.1756**
TURK 0.0635 0.2502 0.0318
LAMB 0.4887** 0.1690 0.1915*
Log- Likelihood = -181.45
R2 (Psuedo) = 27.2 %
Chi-squared= 137.03
% of corrected predictions = 75.9%
*,****** indicate significance at .01, .05, and .10 levels, respectively









The results indicate that there is no significant difference between the willingness

to try amongst demographic characteristics such as ethnicity, gender, nor educational

attainment levels. Caucasians or consumers of other races levels of willingness to try

were compared to those of Hispanics consumers, the base. Unexpectedly, the willingness

to try goat meat amongst ethnic groups was not statistically different, even though

apparent differences existed between each groups responses. Raw statistics indicated that

24.5 percent of Hispanics currently ate goat meat, compared to 12.0 percent of

Caucasians. However, 31.6 percent of non-goat meat consuming, Hispanics were willing

to goat meat, compared to 29.8 percent of non-consuming Caucasians.

Statistically significant demographics variables include household size and age. As

hypothesized, if less than five individuals were present in a household the chances of

trying goat meat diminishes (Figure 6-1). For example, as the household changes from

five or more people to one person, the willingness of trying goat meat decreased by 22.3

percent. Additionally, as household size changes from the base, five or more individuals,

to a two, three, or four person household the likelihood of trying goat decreased 17.9,

19.8, and 23 percent, respectively. Respondents' age also had a significant effect on the

willingness to try goat meat, with respondents between the ages of 55 and 64 more 42.2

percent likely the participants 65 years and older to try goat meat.

The study failed to investigate the reasons respondents were willing or not willing

to try goat meat and goat meat attribute that influence consumption such as, but not

limited to freshness, price, presence of chemical additive, and various nutritional

attributes. However, the survey did focus on the psychographic characteristics that are

important to consumers when making all meat purchases.






43



0o0, 6' -. i_ ,i r ,


-5%/


-10%--

15% /- Household Size
-15?b"


-20%/


-25% /
One person Two people Three people Four people



Figure 6-1. Changes in the probability of trying goat meat with respect to household size.
(All household sizes are compared to the base, households containing five or
more individuals).

The psychographic factors were included in the model to provide insight on potential

goat meat consumers. Of convenience, price specials, fat content, cholesterol, and safety,

price specials and fat content were the only significant variables. The relationships

between price specials and fat content with willingness to try were the opposite of the

original hypotheses. Consumers that viewed price specials as important were 20 percent

more likely than consumers that rated price special as unimportant to be willing to try

goat meat. Additionally, consumers that perceived fat content as relevant were 21.6

percent less likely to be willing to try goat meat than those that view fat levels as

insignificant. These results reveal that many consumers are not knowledgeable of the on

the characteristics, of goat meat, including price. As mentioned in previous chapters,

goat meat is a lean meat and a high source of proteins. Therefore, consumers that believe

fat levels are important should be more inclined to try goat meat. One potential









explanation for the opposite result is that people believe goat meat in high in fat. Also,

goat meat prices are usually higher than traditional meats (McLean-Meyinsse (1999)

found that goat prices range from $1.79 to $2.79 per pound in Louisiana), suggesting

those that believe price specials are important should have been less willing to try goat

meat. The same potential explanation, lack of knowledge about goat meat, could explain

this result. It is not alarming that convenience was not significant. If respondents lack

information on preparation methods, they may be unaware if goat meat is easy to cook or

not.

The probit model indicated that consumers' perception of goat meat had a

statistically significant impact on the likelihood of trying goat meat. As the consumers'

perception changed from negative to positive, the probability of willingness to try goat

meat increased 58.3 percent. Likewise, if consumers possessed a neutral view of goat

meat rather than a negative view the likelihood of consuming goat meat increased 30.2

percent. Based on the results, one can assume that as the consumers' attitude towards

goat meat becomes more positive, their chances of trying goat meat increases at

significant rates.

Finally, the relationship between trying goat meat and the frequency of

consumption of other meat substitutes was examined. Chicken, fish, and lamb

consumption were found to significantly affect the willingness to try goat meat.

Respondents that had previous consumed lamb were more willing to try goat meat. In

fact, participants that had consumed lamb were 48.9 percent more likely to be willing try

goat meat than individuals that had not eaten lamb. The probit estimates also indicated

that fish and other seafood consumption positively effect the likelihood of trying goat









meat. Participants that consumed seafood more than once a week were 15 percent more

likely to express a willingness to try goat meat. Whereas, respondents that consumer fish

weekly were 17.6 percent more incline to be willing to try goat meat. Respondents that

consume chicken more than once a month are less likely to be willing to try goat meat. If

a respondent consumed chicken more than once a week, the probability of consumer goat

meat decreased by 29.1 percent. Likewise if the respondent consumed chicken at least

once a week, the chances of trying goat meat declined 26.6 percent. For the results, one

can infer that goat meat consumption will occur less often as the frequency of chicken

consumed increases. On the other hand, fish and lamb consumption may serve as an

indicator of potential goat consumption.

Hispanic Model

The probit model that focused on the Hispanic respondents only correctly

predicted consumers' willingness to try goat meat 76.2 percent of the time (incorrectly

predicting a consumers' willingness to try goat meat 12.5 percent of the time and non-

willingness to try 11.3 percent of the time). This is better than naive prediction, 51.7.

The chi-squared value is 64.3 and is statistically significant to the .01 confidence level,

which implies this model has good predictive ability of forecasting the willingness to try.

Although ethnicity was statistically insignificant in the general population model, a

separate probit analysis was conducted using Hispanic respondents to reveal if different

factors affected the willingness to consumers between Hispanics and the general

population.










Table 6-2. Empirical results from the Hispanic probit model.

Variable Estimated Coefficient Standard Error Marginal Effects
MEXICAN -0.4283 0.4764 -0.1673
PTRICAN -0.7114 0.4405 -0.2690***
OTHERD -0.2024 0.3570 -0.8056
GENERA -0.1514 0.2801 -0.0603
GENDER 0.1312 0.3171 0.0522
AGE1 2.0406* 0.7452 0.5417*
AGE2 0.1318 0.4631 0.0525
AGE3 0.1973 0.5136 0.0785
AGE4 0.3302 0.4392 0.1307
AGE5 0.3676 0.5432 0.1448
EDU2 0.0520 0.3108 0.0275
EDU3 0.1250 0.3545 0.4983
HSIZE1 -0.3877 0.5957 -0.1510
HSIZE2 -0.2943 0.3831 -0.1164
HSIZE3 -0.8757* 0.4041 -0.3267*
HSIZE4 -0.5672 0.3782 -0.2206
VIEW1 1.4971* 0.4403 0.5457*
VIEW2 0.8390 0.4239 0.1521
FAT -0.6843 0.5699 -0.2603
SAFETY -1.0338 0.7084 0.3605*
CHOLES 0.3370 0.6091 0.1321
CONVEN 0.1072 0.2887 0.0427
PRICE 1.2540* 0.3541 0.4469*
BEEF1 0.3021 0.3544 0.1201
BEEF2 0.3916 0.3902 0.1551
CHICK1 -1.5846* 0.6051 -0.5132*
CHICK2 -1.6177** 0.7845 -0.4825*
PORK1 0.5338 0.3672 0.2087
PORK2 0.3141 0.3477 0.1245
FISH1 0.7546** 0.3583 0.2936**
FISH2 0.3966 0.3689 0.1570
TURK 0.2715 0.3294 0.1076
LAMB 0.3545 0.299 0.1406
Log- Likelihood = -72.41
R2 (Psuedo) =15%
Chi-squared = 64.44
% of corrected predictions = 76.2%
*,**,*** indicate significance at .01, .05, and .10 levels, respectively









Additionally, the second probit analysis was conducted to examine if the levels of trying

goat meat vary amongst the consumers of various Hispanics origins and as the

generations the respondent had spent in the United States (and expected acculturation

level) increases. Consistent with the whole survey sample results, there is no significant

difference between the willingness to try amongst genders and educational attainment

levels. When willingness to try levels for individuals of Puerto Rican, Mexican, and

other descents were compared to those of Cuban descent a statistical significant

difference was found when the individual are of Puerto Rican descent. The results

revealed that if the respondent was of Puerto Rican rather than Cuban descent, the

respondents' willingness to try levels decreased 25.9 percent. According to probit results

there was not a significant difference in the willingness to try goat meat between first

generation Hispanics and Hispanics whose families have been in the United States for

more than one generation. This may suggest that as acculturation increases, consumption

patterns for goat meat remain unchanged, a positive indicator for the goat meat industry.

When only considering Hispanic respondents, the results of the demographic

variable did change. Household size remained significant, but now only the household

size of three was statistically different from the size of five, (32.7 percent less likely to be

willing to try), unlike the larger model, where all sizes were significantly different.

Again only one age variable was significant, but this time it was the youngest age group

(18 and 24). Respondents in this group were 51.4 percent more likely to indicate a

willingness to try goat.

Psychographic factors were again were slightly different in the Hispanic only

model. Safety and price specials were the only significant variables. Consumers that









rated safety as important were 36 percent less likely than individuals that feel safety is

unimportant to try goat meat. Similar to the entire sample, as the significance of price

specials changed from important to unimportant, the chances of trying goat meat

increased 44.7 percent.

Respondents' perception of goat meat again had a statistically significant impact on

the willingness of try goat meat. As the respondents' perception changed from negative

to positive, the likelihood of willingness try goat meat increased 54.6 percent. Unlike the

general population model, there is no significant difference between the levels of trying

goat meat as the consumers' opinion towards the products change from negative to

neutral.

Finally, the analysis examined the relationship between trying goat meat and the

frequency of consumption of other meat substitutes. Chicken and fish and seafood

consumption were found to significantly affect the willingness to try goat meat. Similar

to the entire survey sample model, respondents that consume chicken at least once a week

were less likely to exemplify a willingness to consume goat meat. Respondents that ate

chicken once a week were 48.2 percent less likely to be willing to try goat meat than

those that consume chicken on special occasions or less frequently. When the frequency

of consuming chicken changed from special occasions to more than once a week the

chances of being willing to try goat meat decreased 51.3 percent. Fish and other seafood

consumption positively effected goat meat only when they were consumed more than

once a week. The willingness to try goat meat increased 29.4 percent as the frequency of

fish and seafood consumption varied from more than once a week to on special occasion

and less frequently.














CHAPTER 7
CONCLUSIONS AND IMPLICATIONS

Summary

This thesis focuses on developing an understanding of factors that influence

willingness to consume goat meat in Florida. Researchers believe Florida has the

potential for a meat goat market that is profitable and the demand for goat meat products

in Florida is expected to continue growing as target populations within the niche market

increase. However, the lack of consumer information has hindered producers',

processors', and marketers' ability to fully delineate the profitability and viability of

Florida's meat goat industry. Consumer information is a critical component to totally

understand the possible economic impact of goat meat production and marketing in

Florida. Thus, the primary objective of this research is to identify factors that influence

and barriers that reduce consumption of goat meat.

A probit analysis of willingness to try goat meat indicted that factors influencing

willingness try goat meat differed between the whole survey model and the Hispanics

model. Hispanic consumers were unaware of the safety standards, which may be a result

of the lack of grades and standards in the meat goat industry. The general population

sample participants were uninformed of goat meats nutritional attributes (i.e. low levels

of fat and cholesterol). Results from a partial nutrient analysis (Johnson, 1995),

suggested that goat meat was comparable to chicken in total grams of fat, percent calories

from fat and cholesterol. Also, the nutrient profile indicated that goat meat was similar

beef in iron content. According to probit model results, safety and fat content had a









negative influence on the willingness to try. Indicating that overall, consumer awareness

of goat meat attributes in low, and more information should be made available to

consumers. Both groups of shoppers indicated that price specials were important when

purchasing meat to the grocery store. Price specials increased that likelihood of trying

goat meat; however, specialty meat price are usually more expensive than traditional

meat prices. Therefore, marketers may want implement various pricing strategies,

pricing the meat lower than other meats to increase sales. This study did not identify

reasons for not consuming goat meat. Further research should be conducted to identify

these factors so that the industry can address and attempt to rectify these issues that

restrict consumption.

Perception of goat meat influenced respondents' willingness to try goat meat. As

the respondents' view of goat meat became more positive, the likelihood of trying goat

meat increased at least 30 percent. Unlike the results the entire survey model, the was not

a significant relationship between the willing to try goat meat and the perception as the

respondent view changed from negative to neutral.

Results show that the frequency of chicken and fish consumption effect the

willingness to try goat meat in both models. In addition to those meats, respondents that

consumed lamb were also more likely to be willing try goat meat in the model including

the entire sample. Prices of substitutes were not included in the model, as price for goat

meat were not available either. However the relationship between the willingness to try

goat meat and chicken, fish, and lamb indicated that goat meat has substitutes and

complements. Therefore, members of the goat meat industry could develop strategies

that differentiate goat meat from its competitors in an effort to increase market share. The









likelihood of trying goat meat increased when participants consumed fish on a weekly

basis and if consumers had previously consumed lamb; thus, meat marketing strategies

should be aimed at these individuals.

Income was not included in the model due to a high refusal rate in answering the

question; therefore income elasticities were not calculated. However, education was used

as a proxy for income and was found to be insignificant.

The most unanticipated result from this study was that ethnicity or race was

insignificant, which implies that the willingness to try goat meat is the same for

Hispanics, Caucasians, and other races is statistically the same. This notion means that

marketers should develop marketing strategies that target all consumers, and not focus on

one particular group per se Hispanics. Furthermore, our results indicate that there is

opportunity for growth in the goat meat industry. Gender was not significant, however,

females are usually the primary grocery shoppers; goat meat should be promoted in a

manner than accentuates the characteristics that have been identified as important to

female shoppers.

Other demographic factors that influenced the willingness to try goat meat were

household size and age. Consistent with previous research, larger households, those

containing five or more individuals were most likely to try goat for the entire sample.

Larger households may be more willing to try goat meat because consumers perceive it to

be inexpensive; thus, making goat meat an affordable meat alternative when feeding a

large family. The consumers in the youngest age group, 18-24, were most likely to

express a willingness to try goat meat among Hispanic consumers, where as there was

consumes between the age of 55-64 years were most likely for the entire sample. This is









an important finding for the goat meat industry, because if an individual develops a

fondness for goat meat at an earlier age, more than likely these individuals will become

life long consumers of the product, which may result an overall in goat meat

consumption.

Conclusions

This research suggests opportunities for expanding the goat meat industry exists in

Florida, with 43.6 percent of the sample indicating a willingness to try goat meat if it was

available in supermarkets. Demographic characteristics such ethnicity, gender, nor

educational attainment levels did not affect the dependent variable, but other

demographic characteristics did affect the willingness to try. The results revealed that as

the descent of the respondent changed from Cuban to Puerto Rican, the willingness to try

decreased 25.9 percent. Psychographic factors that effect the willing to try included price

specials, fat content, safety, and the consumers' perception of the product.

Consumers indicated fat content and safety were important to their purchase

decisions were less likely to be willing to goat meat, while perception of goat meat and

price specials has a positive relationship with the willingness to try. As goat meat is low

in fat and often more expensive than other meats, these results seem counter-intuitive.

However, this may all be an indicator that knowledge about the attributes of goat meat

are low and in industry may benefit from educational efforts.

Finally, the analysis describes relationship between the willing to trying goat meat

and the frequency of consumption of other meats. Chicken, fish, and lamb consumption

were found to significantly effect the willingness to try goat meat. Respondents that had

previously consumed lamb were more willing to try goat meat. If the participants

consumed fish and seafood a minimum of once a week, the probability of trying goat









meat increases. Consumers that ate chicken more than once a month were less likely to

try goat meat. Pork and beef consumption were found to have no significant effect on

trying goat meat. One can conclude that goat meat consumption will occur less often as

the frequency of chicken consumed increases. On the other hand, fish and lamb

consumption may serve as an indicator of potential goat consumption.

Implications

In recent years the demand for goat meat has increased in the United States and it

appears that opportunities for expansion exist for Florida's goat meat industry. The

findings from this study can be used by the industry to develop marketing strategies that

will provide assistance in increasing the demand for the product. The results also suggest

potential consumers that should be targeted by the industry.

This study identified new consumers for goat meat. In order for the goat meat

industry to expand, market development strategies, which involve targeting new

consumers with a present product, can be used to enhance the demand for goat meat. It is

easier to target consumers that have an interest in trying goat meat than to market the

product to those that are unwilling to try it. If the goat meat industry wishes to increase

the demand for goat meat it is necessary to expand its target market beyond ethnic

populations and promote the products or all individuals. This study suggests that this

opportunity exists because there was no difference in willingness to try among the ethnic/

racial groups suggesting that the industry should tap into the Caucasian market. Research

suggests that the industry should direct its marketing efforts to individuals living in larger

households and those between the ages of 18-24. It is imperative for the industry to

attract long term consumers in order to have long run success and by targeting younger

consumers this is possible. If younger consumers develop a preference for goat meat









products at an earlier age, they are more than likely to consumer the product as they grow

older. Additionally, since ethnicity/race was an insignificant factor in this study, the

industry should also target Caucasian females. Other potential consumers are those that

live in larger households and lamb consumers.

It seems a major barrier that hinders the prosperity of the goat meat industry is

that consumers lack knowledge on goat meat. Consumers are becoming more health

conscious and they are consuming products that possess nutritional qualities such as

chicken and fish more frequently. Goat meat is very healthy, low in fat and cholesterol

and high in proteins, however consumers are unaware of these qualities; therefore,

educational information that increases consumers' awareness of goat meat. If this

information was known by consumers the industry may be able to repositioning the

product in consumers' mind. It seems that consumers perceive the product as cheap and

containing high levels of fat, but the in untrue. In store demonstrates, educational

advertising, recipes are promotional strategies that would increase consumers' awareness

of goat meat, resulting in increased consumption.

As a result of this study it is evident that further research that involves knowledge

testing is needed. This type of research would reveal the familiarity levels consumers

have for goat meat. These results would inform industry official on the subject matter

that needs to be discussed in the educational advertisements. Additionally, since

ethnicity/race was non-significant and more balance sample, on the does not place an

emphasis on Hispanics, should be conducted to investigate the reasons for consuming and

not consuming goat meat is needed. The Florida's goat meat industry has the opportunity






55


to flourish; however, effective marketing strategies are needed to increase consumers'

awareness and the availability of goat meat.















APPENDIX A
SOUTHERN REGION GOAT CONSUMPTION SURVEY

March 15, 2004


Hello, this is (NAME) calling from the University of Georgia in Athens. The Survey
Research Center is conducting a study this evening in conjunction with Mack C. Nelson,
a professor from Fort Valley State University (GA) concerning the use of goat meat and
we'd like to talk to the primary food shopper of your household. Do you have a few
minutes right now to complete an interview?

1. Yes (CONTINUE)
2. No (GET SR'S NAME, ARRANGE CALLBACK; APPLY PERSUADERS.
EVEN IF RESPONDENT DOESN'T EAT GOAT MEAT, WE WANT THEM TO
COMPLETE THE SURVEY-IT WILL BE SHORTER. )

[INTERVIEWER NOTE: IF PRIMARY FOOD SHOPPER DID NOT HEAR INITIAL
INTRODUCTION BUT DOES COME TO THE PHONE, REPEAT INTRO]

S1 Are you 18 years or older?

1. Yes (CONTINUE)
2. No (ASK TO SPEAK TO ADULT 18 YEARS OR OLDER. RETURN TO
INTRO. IF NECESSARY, GET SR'S NAME AND SET CALLBACK.)

Great! Before we begin, I need to let you know that the interview is completely
voluntary. All of the information you provide will be kept strictly confidential and you
don't have to answer any questions you don't want to. Also, my supervisor may listen to
part of the interview to be sure that I'm not making any mistakes.

Q1. Have you or any member of your immediate family ever eaten goat meat?

1. Yes [SKIP TO Q4]
2. No
3. Don't know
9. Ref/NA




Q2. DELETED






57



Q2.1 If goat meat was available in your area food stores, do you think you would try it?

1. Yes
2. No
9. Ref/DK/NA

Q2.2 If there were a cooperative that sold the meat of animals grown organically,
would you be willing to join?

1. Willing
2. Not willing
9. Ref/DK/NA

[ALL ANSWERS SKIP TO Q14]

Q3. DELETED

Q4. What's your preference in goat meat? Would it be the kid, small male, small
female, wether or something else?

1. Kid
2. Small male
3. Small female
4. Wether (castrated male)
5. Other [Specify]
9. Ref/DK/NA

Q5. What live weight do you prefer?

1. Less than 30 pounds
2. 30 50 pounds
3. 51 69 pounds
4. 70 pounds or more
9. Ref/DK/NA

Q6. Do you prefer a certain cut of meat?

1. Yes
2. No [SKIP TO Q7]
9. Ref/DK/NA [SKIP TO Q7]






58


Q6.1 How much do you prefer the shoulder? Do you prefer the shoulder very much,
somewhat, not much or not at all?

1. Very much
2. Somewhat
3. Neutral (Doesn't matter)
4. Not much
5. Not at all
9. Ref/DK/NA

Q6.2 How much do you prefer the ribs? Would it be very much, somewhat, not much
or not at all?

1. Very much
2. Somewhat
3. Neutral (Doesn't matter)
4. Not much
5. Not at all
9. Ref/DK/NA

Q6.3 How much do you prefer the hind leg?

1. Very much
2. Somewhat
3. Neutral (Doesn't matter)
4. Not much
5. Not at all
9. Ref/DK/NA

Q6.4 How much do you prefer loin chops?

1. Very much
2. Somewhat
3. Neutral (Doesn't matter)
4. Not much
5. Not at all
9. Ref/DK/NA









Q6.5 How much do you prefer loin cubes?

1. Very much
2. Somewhat
3. Neutral (Doesn't matter)
4. Not much
5. Not at all
9. Ref/DK/NA

Q6.6. Are there any other cuts of meat that you prefer to eat?

1. Name cuts of meat
9. No, Ref/DK/NA

Q7. Do you normally buy a whole goat?

1. Yes
2. No
3. Don't know
9. Ref/NA

Q8. Are there certain seasons of the year that you eat more goat meat?

1. Yes
2. No [SKIP TO Q9]
9. Ref/DK/NA [SKIP TO Q9]

Q8.1 What seasons are those?

[INTERVIEWER NOTE: CHOOSE ALL THAT APPLY]

[PROGRAMMER NOTE: YES/NO TOGGLE]

1. Winter
2. Spring
3. Summer
4. Fall
5. Ref/DK/NA
6. Exit









Q9 Do you eat goat meat on special occasions?

1. Yes
2. No [SKIP TO Q10]
9. Ref/DK/NA [SKIP TO Q10]

Q9.1 Which special occasions are those?

[INTERVIEWER NOTE: CHOOSE ALL THAT APPLY]

[PROGRAMMER NOTE: YES/NO TOGGLE]

1. Christmas
2. 4th of July
3. Family re-unions
4. Marriages
5. Ramadan
6. Cinco de Mayo
7. Other [Specify]
8. Ref/DK/NA
9. Exit

Q10. About how many pounds of goat meat do you think your family eats each year?

[INTERVIEWER NOTE: USE PERSUADERS IF NECESSARY: "I JUST
NEED A BALLPARK FIGURE."]

pounds
998 998 or more
999 Ref/DK/NA
[RANGE: 1 999]

Qll. Would your family eat more goat meat if it was available in your local grocery
stores?

1. Yes
2. No [SKIP TO Q12]
3. Don't know [SKIP TO Q12]
9. Ref/NA [SKIP TO Q12]






61


Q11.1 Which meat product would you eat less of if you increased the family
consumption of goat meat?

[INTERVIEWER NOTE: IF NECESSARY, READ: BEEF, PORK, SEAFOOD,
LAMB, CHICKEN OR TURKEY?]

1. Enter response
9. Ref/DK/NA

Q12. Please tell me how important the following attributes are in your decision to
purchase goat meat products. Would you say fresh, never frozen product is very
important, important, not very important or not at all important?

1. Very important
2. Important
3. Neutral (Doesn't matter)
4. Not very important
5. Not at all important
9. Ref/DK/NA

Q12.1 How important is the color of the meat-very important, important, not very
important or not at all important?

1. Very important
2. Important
3. Neutral (Doesn't matter)
4. Not very important
5. Not at all important
9. Ref/DK/NA

Q12.2 How important is the government inspection label?

1. Very important
2. Important
3. Neutral (Doesn't matter)
4. Not very important
5. Not at all important
9. Ref/DK/NA






62


Q12.3 How important is it that the goat meat is organically grown?

1. Very important
2. Important
3. Neutral (Doesn't matter)
4. Not very important
5. Not at all important
9. Ref/DK/NA

Q12.4 How important is it that there are a variety of cuts available?

1. Very important
2. Important
3. Neutral (Doesn't matter)
4. Not very important
5. Not at all important
9. Ref/DK/NA

Q12.5 How important is it that there are prepackaged cuts?

1. Very important
2. Important
3. Neutral (Doesn't matter)
4. Not very important
5. Not at all important
9. Ref/DK/NA

Q12.6 How important are cooking instructions?

1. Very important
2. Important
3. Neutral (Doesn't matter)
4. Not very important
5. Not at all important
9. Ref/DK/NA









Q12.7 How important are marinade cuts?


Very important
Important
Neutral (Doesn't matter)
Not very important
Not at all important
Ref/DK/NA


Q12.8 How important are convenience foods, such as sausage?

1. Very important
2. Important
3. Neutral (Doesn't matter)
4. Not very important
5. Not at all important
9. Ref/DK/NA


Q12.9 How important is the price in your decision to purchase goat meat products?


1. Very important
2. Important
3. Neutral (Doesn't matter)
4. Not very important
5. Not at all important
9. Ref/DK/NA


Q13. When you cook your goat meat, how often do you make soup? Would you say
frequently, some of the time, not very often or never?

1. Frequently
2. Some of the time
3. Not very often
4. Never
9. Ref/DK/NA






64


Q13.1 When you cook your goat meat, how often do you make meat sauce? Would you
say frequently, some of the time, not very often or never?


Frequently
Some of the time
Not very often
Never
Ref/DK/NA


Q13.2 How often do you make chili when you cook goat meat?

1. Frequently
2. Some of the time
3. Not very often
4. Never
9. Ref/DK/NA

Q13.3 How often do you make meat loaf when you cook goat meat?


Frequently
Some of the time
Not very often
Never
Ref/DK/NA


Q13.4 When you cook your goat meat, how often do you broil it? Would you say ...

1. Frequently
2. Some of the time
3. Not very often
4. Never
9. Ref/DK/NA


Q13.5 When you cook goat meat, how often do you oven roast it?


Frequently
Some of the time
Not very often
Never
Ref/DK/NA









Q13.6 When you cook goat meat, how often do you have sausage made?

1. Frequently
2. Some of the time
3. Not very often
4. Never
9. Ref/DK/NA

Q13.7 When you cook your goat meat, how often do you barbeque the meat?

1. Frequently
2. Some of the time
3. Not very often
4. Never

9. Ref/DK/NA

Q13.8 Are there any other ways that you cook your goat meat?

1. Enter response

9. Ref/DK/NA


Q14. This section is about which meats you eat most often. Would you say that you eat
beef every day, more than once a week, once a week, more than once a month,
once a month, on special occasions or never?

1. Every day
2. More than once per week
3. Once per week
4. More than once per month
5. Once per month
6. Special occasions
7. Never [SKIP TO Q15]


9. Ref/DK/NA [SKIP TO Q15]






66


Q14.1 When you eat beef how much do you usually eat? Would you say less than /4
pound, 14 pound, 12 pound, 1 pound or more than one pound?


Less than one-fourth pound
One-fourth pound
One-half pound
One pound
More than one pound
Ref/DK/NA


Q15 And how often would you say that you eat chicken? Would it be ...

1. Every day
2. More than once per week
3. Once per week
4. More than once per month
5. Once per month
6. Special occasions
7. Never [SKIP TO Q16]
9. Ref/DK/NA [SKIP TO Q16]

Q15.1 When you eat chicken how much do you usually eat? Would you say less than 14
pound, 14 pound, 12 pound, 1 pound or more than one pound?


Less than one-fourth pound
One-fourth pound
One-half pound
One pound
More than one pound
Ref/DK/NA


Q16 What about turkey, how often do you eat turkey?


Every day
More than once per week
Once per week
More than once per month
Once per month
Special occasions
Never [SKIP TO Q17]
Ref/DK/NA [SKIP TO Q17]






67


Q16.1 When you eat turkey how much do you usually eat? Would you say less than /4
pound, 14 pound, 12 pound, 1 pound or more than one pound?

1. Less than one-fourth pound
2. One-fourth pound
3. One-half pound
4. One pound
5. More than one pound
9.Ref/DK/NA

Q17 And how often do you eat lamb?

1. Every day
2. More than once per week
3. Once per week
4. More than once per month
5. Once per month
6. Special occasions
7. Never [SKIP TO Q18]
9. Ref/DK/NA [SKIP TO Q18]

Q17.1 When you eat lamb how much do you usually eat? Would you say less than 14
pound, 14 pound, 12 pound, 1 pound or more than one pound?


Less than one-fourth pound
One-fourth pound
One-half pound
One pound
More than one pound
Ref/DK/NA


Q18 How often do you eat goat meat or chevon?


Every day
More than once per week
Once per week
More than once per month
Once per month
Special occasions
Never [SKIP TO Q19]
Ref/DK/NA [SKIP TO Q19]






68


Q18.1 When you eat goat/chevon how much do you usually eat? Would you say less
than 14 pound, 14 pound, 12 pound, 1 pound or more than one pound?


Less than one-fourth pound
One-fourth pound
One-half pound
One pound
More than one pound
Ref/DK/NA


Q19 How often do you eat fish or seafood?

1. Every day
2. More than once per week
3. Once per week
4. More than once per month
5. Once per month
6. Special occasions
7. Never [SKIP TO Q20]
9. Ref/DK/NA [SKIP TO Q20]


Q19.1 When you eat fish or seafood, how much do you usually eat? Would you say less
than 14 pound, 14 pound, 12 pound, 1 pound or more than one pound?


Less than one-fourth pound
One-fourth pound
One-half pound
One pound
More than one pound
Ref/DK/NA


Q20 And finally, how often do you eat pork?


Every day
More than once per week
Once per week
More than once per month
Once per month
Special occasions
Never [SKIP TO Q23]
Ref/DK/NA [SKIP TO Q23]









Q20.1 And when you eat pork how much do you usually eat? Would you say less than /4
pound, 14 pound, 12 pound, 1 pound or more than one pound?

1. Less than one-fourth pound
2. One-fourth pound
3. One-half pound
4. One pound
5. More than one pound
9. Ref/DK/NA

Q21 Question deleted (duplicate of Q 11)

Q22 DELETED

Q23. In your household, has your spouse consumed goat meat products?

1. Yes
2. No
3. Do not have spouse
8. Ref/DK/NA

[PROGRAMMER NOTE: IF Q23 = 3, SKIP Q36]

Q23.1 Have your children consumed goat meat products?

1. Yes
2. No
3. Do not have children [SKIP TO Q26.1]

9. Ref/DK/NA [SKIP TO Q26.1]

[PROGRAMMER NOTE: IF Q23.1 = 3, SKIP Q37]

Q24. Now I'd like to ask you the gender of your children. Remember, this is
confidential
and the answers will not be connected with your phone number in any way.

Q24.1 DELETED

Q24.2 Gender of first child under 18?

1. Male
2. Female
9. Ref/DK/NA [SKIP TO Q25]









Q24.3 DELETED

Q24.4 Gender of second child under 18?

1. Male
2. Female
3. No more children [SKIP TO Q25]
9. Ref/DK/NA [SKIP TO Q25]

Q24.5 DELETED

Q24.6 Gender of third child under 18?

1. Male
2. Female
3. No more children
9. Ref/DK/NA

Q25. If you and/or your spouse eat goat meat but your children don't, what are the
reasons your children don't consume the product?

[INTERVIEWER NOTE: CHOOSE ALL THAT APPLY]

[PROGRAMMER NOTE: YES/NO TOGGLE]

1. They don't like it
2. They were not reared where it was consumed regularly
3. It wasn't available
4. Their friends don't eat it, so they don't
5. Others (please list)
6. Ref/DK/NA
7. Exit

Q26. DELETED

Q26.1 To your knowledge, have any of the following family members eaten goat meat
products?

[INTERVIEWER NOTE: READ RESPONSES. CHOOSE ALL THAT APPLY]






71


[PROGRAMMER NOTE: YES/NO TOGGLE]

1. Mother
2. Father
3. In-Laws
4. Others [Please specify ]
5. Ref/DK/NA
6. Exit

Q27 I would like to know how you see goat meat products. Would you say your view
is:

1. Very positive
2. Positive
3. Somewhat positive
4. Neutral
5. Somewhat negative
6. Negative
7. Very negative
9. Ref/DK/NA


For the next few items, I'd like you to rate the importance of several factors in your
decision to buy or not to buy a goat meat product. Even if you do not currently consume
goat meat, please tell me how important the following items are in your decision to buy
goat meat and other kinds of meat.

Q28. How important are food page advertisements in your decision to buy goat meat?

1. Very important
2. Important
3. Neutral
4. Not very important
5. Not at all important
10. Ref/DK/NA

Q28.1 How important are food page advertisements in your decision to buy other meats?

1. Very important
2. Important
3. Neutral
4. Not very important
5. Not at all important
9. Ref/DK/NA






72


Q29. How important are store displays in your decision to buy goat meat?

1. Very important
2. Important
3. Neutral
4. Not very important
5. Not at all important
9. Ref/DK/NA

Q29.1 How important are store displays in your decision to buy other meats?

1. Very important
2. Important
3. Neutral
4. Not very important
5. Not at all important
9. Ref/DK/NA

Q30. How important are price specials in your decision to buy goat meat?

1. Very important
2. Important
3. Neutral
4. Unimportant
5. Not important at all
9. Ref/DK/NA

Q30.1 How important are price specials in your decision to buy other meats?

1. Very important
2. Important
3. Neutral
4. Unimportant
5. Not important at all
9. Ref/DK/NA

Q31. How important are in-supermarket taste tests in your decision to buy goat meat?

1. Very important
2. Important
3. Neutral
4. Unimportant
5. Not important at all
9. Ref/DK/NA






73


Q31.1 How important are in-supermarket taste tests in your decision to buy other meats?

1. Very important
2. Important
3. Neutral
4. Unimportant
5. Not important at all
9. Ref/DK/NA

Q32. How important are safety assurances such as USDA inspections of goat meat?

1. Very important
2. Important
3. Neutral
4. Unimportant
5. Not important at all
9. Ref/DK/NA

Q32.1 How important are safety assurances such as USDA inspections of other meats?

1. Very important
2. Important
3. Neutral
4. Unimportant
5. Not important at all
9. Ref/DK/NA

Q33. How important are convenience products such as sausage or ground meat from
goat meat?

1. Very important
2. Important
3. Neutral
4. Somewhat unimportant
5. Not important at all
9. Ref/DK/NA

Q33.1 How important are convenience products such as sausage or ground meat from
other meats?

1. Very important
2. Important
3. Neutral
4. Somewhat unimportant
5. Not important at all
9. Ref/DK/NA






74


Q34. How important is the fat content in goat meat?

1. Very important
2. Important
3. Neutral
4. Somewhat unimportant
5. Not important at all
9. Ref/DK/NA

Q34.1 How important is the fat content in other meats?

1. Very important
2. Important
3. Neutral
4. Somewhat unimportant
5. Not important at all
9. Ref/DK/NA

Q35. How important is the cholesterol content of goat meat?

1. Very important
2. Important
3. Neutral
4. Somewhat unimportant
5. Not important at all
9. Ref/DK/NA

Q35.1 How important is the cholesterol content of other meats?

1. Very important
2. Important
3. Neutral
4. Somewhat unimportant
5. Not important at all
9. Ref/DK/NA









For each of the following that I read to you, please rate how important each would be in
your decision to purchase goat meat products, just as you make decisions to buy beef,
poultry, or pork. Please use a scale of 1 to 5 where "1" is very important and "5" is not
important at all.

Q36. First, how important is your spouse's opinion on your decision to purchase or not
purchase goat meat products?

1. Very important
2. Important
3. Neutral
4. Somewhat unimportant
5. Not important at all
9. Ref/DK/NA

[PROGRAMMER NOTE: IF Q23 = 3, THIS QUESTION IS SKIPPED (Q36)]

Q37. How important are your children's opinions on your decision to purchase or not
purchase goat meat products?

1. Very important
2. Important
3. Neutral
4. Somewhat unimportant
5. Not important at all
9. Ref/DK/NA

[PROGRAMMER NOTE: IF Q23.1 = 3, THIS QUESTION IS SKIPPED (Q37)

Q38. How important is your own opinion on your decision to purchase or not purchase
goat meat products?

1. Very important
2. Important
3. Neutral
4. Somewhat unimportant
5. Not important at all
9. Ref/DK/NA

[PROGRAMMER NOTE: IF Q1 > 1, SKIP TO Q47]

Q39. DELETED


Q40. DELETED






76


Q41. How often does your family eat goat meat?

1. Weekly 5. Quarterly
2. Bi-Weekly 6. Special
occasions
3. Monthly 7. Refused
4. Less than quarterly 8. Don't know

Q42. Do you get your goat meat from .. .?

[INTERVIEWER NOTE: READ LIST, CHOOSE ALL THAT APPLY]

[PROGRAMMER NOTE: YES/NO TOGGLE

1. Supermarket
2. Farmer
3. A friend not a farmer
4. Farmer's Market
5. Grow our own
6. Restaurant
7. Other place (SPECIFY)
8. Ref/DK/NA
9. Exit

Q43. Would you be willing or not willing to join a consumer-farmer cooperative if
prices for goat meat were lower?

1. Willing
2. Not willing

9. Ref/DK/NA

Q44. Would you be willing or not willing to join a co-op if it had a more dependable
source of goat meat?

1. Willing
2. Not willing

9. Ref/DK/NA

Q45. Would you be willing or not willing to join a cooperative that grew its animals
organically?

1. Willing
2. Not willing






77


9. Ref/DK/NA

46. Would you be willing or not willing to join a cooperative that sold its members
fresh, never frozen goat meat?


Willing
Not willing
Ref/DK/NA


Q47. How many people live in your home?

number of people

99 Ref/DK/NA

[RANGE: 1 99]

Q48 How many of the people are less than 18 years old?

less than 18 years old

99 Ref/DK/NA

[RANGE: 0 99]

Now I just need to ask a few questions about you personally so that we can compare your
answers with different types of people.

Q49. What do you consider your race to be?


White [SKIP TO Q54]
African-American/Black [SKIP TO Q54]
Black Non African American [SKIP TO Q54]
Hispanic
Asian [SKIP TO Q54]
Multi-Racial (SPECIFY)( )[SKIP TO Q54]
Ref/DK/NA [SKIP TO Q54]









Q50. Are you of...?

[INTERVIEWER NOTE: READ RESPONSES. CHOOSE ONLY ONE]

1. Mexican descent
2. Cuban descent
3. Puerto Rican descent
4. Spaniard
5. Or Other (Please list)

7. Ref/DK/NA

Q51. Are you the 1st generation, 2nd generation or another generation to live in the
contiguous U. S. States?

1. 1st generation
2. 2nd generation
3. Other

9. Ref/DK/NA

Q52. DELETED

Q53. Which of your family members are Hispanic?

[INTERVIEWER NOTE: CHOOSE ALL THAT APPLY]

[PROGRAMMER NOTE: YES/NO TOGGLE]

1. Mother
2. Father
3. Spouse
4. Aunts or uncles
5. Cousins
6. Other
7. Ref/DK/NA
8. Exit









Q54. What is your age range?


Less than 20 years
20 24 years
25 34 years
35 44 years
45 54 years
55 59 years


60 64 years
65 74 years
75 84 years
85 years plus
Refused


Q55. What is the highest grade of school or year of college you completed?


High School Diploma/GED
Associate/Technical Degree
Some College
College Graduate
Post Graduate/Professional
Ref/DK/NA


Q56. INTERVIEWER: (If necessary: I know the answer to this question, but I am
required to ask. Are you:)


1. Male
2. Female
9. N/A


Q57. We're almost finished and I would like to ask what your total gross household
income for 2003 was I don't need an exact figure, just an approximate category.
So, from the list I am about to read to you, could you tell me if your total
household income was:


<$10,000
$10,000- $14,999
$15,000 $19,999
$20,000 $24,999
$25,000 $34,999
$35,000 $49,999
$50,000 $74,999
$75,000 $99,999
$100,000 or more


Don't Know
Refused


Q58. And finally, what state do you live in?

1. AL
2. AR
3. FL









4. GA
5. MS
6. NC
7. OK
8. LA
9. SC
10. TN
11. TX
99 Ref/DK/NA

This completes the survey and I want to thank you for taking the time to answer these
questions. Have a nice evening. Good bye.

IMPORT FIPS

IMPORT MSA/NON-MSA


QUOTA: 250 FOR EACH STATE















APPENDIX B
SUMMARY OF DEMOGRAPHIC INFORMATION


* Hispanics
E Cacausians
o Others


Had Consumed Willing to Try Had not Consumed
and Unwilling Try


Note: Chi-squared probability < .10.


Figure B-1. Respondents' consumption preference for goat meat by ethnicity/race.


* Male
O Female


Had Consumed Willing to Try Had not Consumed
and Unwilling Try


Note: Chi-squared probability < .05.


Figure B-2. Respondents' consumption preference for goat meat by gender.











60

50

40
40* HS Diploma or less
30 0 Some College

20-X 0 College Degree and above

10


Had Willing to Try Had not
Consumed Consumed and
Unwilling Try




Figure B-3. Respondents' consumption preference for goat meat by educational
attainment levels.


Had Consumed Willing to Try


Had not
Consumed and
Unwilling Try


* Mexican
O Puerto Rican
o Cuban
* Other


Figure B-4. Hispanic respondents' consumption preference for goat meat by origin.































Figure B-5. Hispanic respondents' consumption preference for goat meat by gender.


Figure B-6. Hispanic respondents' consumption preference for goat meat by educational
attainment levels.


40
Male
30
0z- El Female
20

10


Had Consumed Willing to Try Had not
Consumed and
Unwilling Try


8 30-
S* HS Diploma or less
20 O Some College
10-/ U I [O College Degree and above

0
Had Consumed Willing to Try Had not
Consumed and
Unwilling Try



























Figure B-7. Hispanic respondents' consumption preference for goat meat by generation in
United States.















LIST OF REFERENCES


Aldrich, John H., and Forrest D. Nelson. Linear Probability, Logit, and Probit Models.
Edited by Michael S. Lewis-Beck. Vol. 45, Quantitative Applications in the
Social Sciences. Newbury Park, CA: Sage Publications, Inc., 1984.

Associated Press. 2005." Polling the 'Cell Phone Only' Crowd: Pollisters, Researchers
Tackle Hard-to-Track Population." In, Cable News Network,
http://www.cnn.com/2005/TECH/02/25/polling.cell.users.ap/index.html.
(accessed February 25, 2005).

Chen, Kevin, Murad Ali, Michele Veeman, Jim Unterschultz, and Theresa Le. "Relative
Importance Rankings for Pork Attributes by Asian-Origin Consumers in
California: Applying an Ordered Probit Model to a Choice-Based Sample."
Journal ofAgricultural and Applied Economics 34 (2002): 67-79.

Davis, Ernie and Zane Willard. 1996. "Goat Meat the "New" Lean Meat." In, Texas
A&M University, Texas A&M Agriculture News,
http://agnews.tamu.edu/dailynews/stories/AGEC/Dec2096a.htm. (accessed May
22, 2003).

Degner, L. Robert. "Should You Market Chevon, Cabrito, or Goat Meat." Paper
presented at the Regional Small Farm and Trade Show Conference, Tallahassee,
FL, November 7, 1991.

Degner, L. Robert and J. David Locascio. "Distribution of Goat Meat in Selected
Metropolitan Florida Markets." Gainesville, FL: University of Florida, Food
Resource Economics Department, Florida Agricultural Market Research Center,
1988.

Degner, L. Robert, and C. T. Jordan Lin. "Marketing Goat Meat: An Evaluation of
Consumer Perceptions and Preferences." University of Florida, Food Resource
Economics Department, Florida Agricultural Market Research Center, 1993.

Degner Robert, L., and Susan D. Moss. "The Florida Market for Goat Meat: 1999 Survey
of Florida Meat Wholesalers." Gainesville, FL: University of Florida, Food
Resource Economics Department, Florida Agricultural Market Research Center,
1999.

Dillman, Don. Mail and Telephone Surveys: The Total Design Method. New York:
Wiley-Interscience Publication, 1978.










Frey, James, A. Survey Research by Telephone. 2nd ed. Vol. 150. Newbury Park, CA:
Sage Publications, 1989.

Gipson, Terry A. 1999. Demand for Goat Meat: Implications for the Future of the
Industry. In Demand for Goat Meat: Implicationsfor the Future of the Industry,
Langston University, E(Kika)de la Garza Institute for Goat Research,
http://www2.1uresext.edu/goats/library/field/goatmeatdemand99.htm. (accessed
January 20, 2004).

Greene, William H. Econometric Analysis. 5th ed. Upper Saddle River, NJ: Pearson
Education, Inc., 2003.

Gujarati, Damodor. Basic Econometrics. International Edition 2003 ed. New York, NY:
McGraw Hill Irwin, 2003.

Harwell, Lynn. "Goat Marketing Strategies Along the East Coast." In Florida's Goat
Meat Industry, edited by James R. Simpson, 15-21: Florida Cooperative
Extension Service, Institute of Food and Agricultural Sciences, University of
Florida, 1995.

Hui, Jianguo, Patricia E. McLean, and Dewitt Jones. "An Empirical Investigation of
Importance Rating of Meat Attributes by Louisiana and Texas Consumers."
Journal ofAgricultural and Applied Economics 27
(1995): 636-43.

James, N. A. and B. W. Berry. "Use of Chevon in the Development of Low-Fat Meat
Products." Journal ofAnimal Science 75 (1997): 571-77.

Johnson, Dwain. "Composition and Quality of Goat Meat Produced." edited by James R.
Simpson, 39-41: Florida Cooperative Extension Service, Institute of Food and
Agricultural Sciences, University of Florida, 1995.

Latalia, Thomas. "A Pseudo R2 Measure for Limited and Qualitative Dependent Variable
Models." Journal ofEconometrics 56 (1993): 341-56.

Lavrakas, Paul J. Telephone Survey Methods: Sampling, Selection, and Supervision.
Edited by Leonard Bickman. Vol. 7, Applied Social Research Method Series.
Newbury Park, CA: Sage Publications, Inc., 1987.


McKenzie-Jakes, Angela. 2004. "Markets for Goat Meat." In, Florida A&M University,
http://smallfarms.ifas.ufl.edu/Liverstock/goats.htm. (accessed July 18, 2003).









McLean-Meyinsse, Patricia E. "Assessing Factors Affecting Consumers' Decisions to
Shop at Stores Offering Specialty Meat." Journal ofFood Distribution Research
78 (1999): 134-39.

- "Assessing the Market Outlook for Rabbit Meat in Louisiana and Texas."
Journal ofFoodDistribution Research 31 (2000): 139-44.

Factors Influencing Consumption or Willingness to Consume a Variety of
Goat-Meat Products." Journal of Food Distribution Research 34 (2003): 72-79.

Melton, Bryan, E., Wallace E. Huffman, Jason F. Shogren, and John A. Fox. "Consumer
Preferences for Fresh Food Items with Multiple Quality Attributes: Evidence from
an Experimental Auction of Pork Chops." American Journal ofAgricultural
Economics 78 (1996): 916-23.

Miller, Pat. "Marketing Strategies for Goat Meat." In Florida's Meat Goat Industry,
edited by James A. Simpson, 22-26: Florida Cooperative Extension Service,
Institute of Food and Agricultural Sciences, University of Florida, 1995.

Nelson, Mack C., N.B. Brown, Jr., S. Mobini, S. Gelaye, and S. Leak. "Production and
Marketing Challenges for Goat Producers: Implication for Supply and Demand."
Paper presented at the Professional Ag Workers Conference (PAWC), December
1999.

Ohio Cooperative Development Center. "Market Analysis of Meat Goat in Ohio".
Ohio Cooperative Development Center. Available at
http://ocdc.osu.edu/pdf/anal_proposal.pdf. Accessed on. (accessed February 19,
2004).

Peter, J. Paul, and James H. Donnelly, Jr. A Preface to Marketing Management. 9th ed.
New York, NY: McGraw-Hill Irwin, 2003.

Pinkerton, Frank, Lynn Harwell, William Drinkwater, Nelson Escobar. "Marketing Meat
Goats: Channels, Supply, and Demand." Langston University, OK: Oklahoma
Cooperative Extension Service, 1994.

Pinkerton, Frank, David Scarfe, and Bruce Pinkerton. "Meat Goat Productions and
Marketing." Langston University, E(Kika)de la Garza Institute for Goat Research,
1992.

Putnam, Judith Jones, and Jane E. Allshouse. "Food Consumption, Prices, and
Expenditures, 1970-97." edited by Economic Research Service (USDA-ERS)
U.S. Department of Agriculture, 196, April 1999.

Progressive Grocer. Progressive Grocer, 69th Annual Report of the Grocery Industry."
April 2002.










Rhee, K. S., C.E. Meyers, and D.F. Waldron. "Consumer Sensory Evaluation of Plain and
Seasoned Goat Meat and Beef Products." Meat Science 65 (2003): 785-89.

Rhee, K. S., M. Oltman, and J. Han. "A Consumer Survey of Goat Meat: Perception,
Knowledge, and Use." Sheep and Goat Research Journal 16 (2000): 111-16.

Russell, Jeremy. "Targeting Hispanic Consumers." National Provisioner 216 (2002): 44-
53.

Salant, Priscilla, and Don A. Dillman. How to Conduct Your Own Survey. New York:
Wiley, 1994.

Schupp, Alvin, Jeffery Gillespie, and Debra Reed. "Consumer Choice among Alternative
Red Meats." Journal ofFood Distribution Research 29 (1998): 35-43.

Sherman, David. 2002. Current Tends in Goat Production in the United States. In,
http://www.neahi.org.nejlah/articles/articles_goatmeat.html. (accessed May 22,
2003).

Simpson, James. "Introduction to Florida's Meat Goat Industry." In Florida's Goat Meat
Industry, edited by James R. Simpson, 1-4: Florida Cooperative Extension
Service, Institute of Food and Agricultural Sciences, University of Florida, 1995.

Smith, C. G., M. I. Pike, and Z. L. Carpenter. "Comparison of the Palatability of Goat
Meat and Meat from Four Other Animal Species." Journal ofFood Science 39
(1974): 1145-46.

Spaugh, Beth C. 1997. "Goat Carcass Characteristics." In, Cornell Cooperative
Extension, http://www.cce.cornell.edu/clinton/ag/goat-carcass.html. (accessed
August 20, 2004).

Tomek, William G., and Kenneth L. Robinson. Agricultural Product Prices. 3rd ed.
Ithaca, New York: Cornell University Press, 1990.

U.S. Census Bureau. 2000. U.S. Census 2000. In,
http://factfinder.census.gov/home/saff/main.html?_lang=en. (accessed August 20,
2004).

U.S. Department of Agriculture, National Agricultural Statistical Service (USDA-NASS).
2002. Census of Agriculture 2002. In, USDA-NASS, http://www.nass.usda.gov.
(accessed December 17, 2004).

Verbeek, Marno. A Guide to Modern Econometrics. Southern Gate, Chichester, West
Sussex, England: John Wiley & Sons, 2004.






89



Verbeke, Wim, Ronald W. Ward, and Jacques Viaene. "Probit Analysis of Fresh Meat
Consumption in Belgium: Exploring BSE and Television Communication
Impact." Agribusiness 16 (2000): 215-34.

Walsh, Robb. "Getting Your Goat." Natural History 104 (1995): 48-49.

Zachery, Nikki, and Mack C. Nelson. "Consumers Knowledge and Use of Goat Products:
An Atlanta Case Study." Paper presented at the Professional Ag Workers
Conference (PAWC), Tuskegee University, December 1992.














BIOGRAPHICAL SKETCH

Erika Knight was in Warner Robins, GA. After graduating from Warner Robins

High School, she attended Fort Valley State University, Fort Valley, GA, and earned a

Bachelor of Science degree in agricultural economics. In August 2004, Erika began the

Food and Resource Economics Master of Science program and specialized in marketing.




Full Text

PAGE 1

EVALUATION OF CONSUMER PREFERENCES REGARDING GOAT MEAT IN FLORIDA By ERIKA KNIGHT A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2005

PAGE 2

This document is dedicated in loving memory of my grandmother, Mattie Hugley Dixon.

PAGE 3

Copyright 2005 by Erika Knight

PAGE 4

ACKNOWLEDGMENTS First and foremost, I thank God for giving me the strength I needed to successfully complete this masters program. I know that He is the reason I have made it this far in life. I thank family and friends for the encouragement, support, and patience throughout the past two years. These individuals are another reason I have made it to this point of my life. I express my deepest appreciation to my committee, Drs. Lisa House, Robert Degner and Mack Nelson, for the interest and support given while preparing my thesis. I would especially like to thank my committee chair, Dr. Lisa House, for her advisement, guidance and the knowledge she shared while completing my masters program. I am truly grateful to my committee for all of their assistance and I would like to express my gratitude once again. Finally, I would like to thank my fellow graduate students in the Food and Resource Economics Department and members of the Black Graduate Student Organization (BGSO) for their support and encouragement. I thank them for the memories; each of them has made this graduate experience unforgettable. iv

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES............................................................................................................vii LIST OF FIGURES.........................................................................................................viii ABSTRACT....................................................................................................................... ix CHAPTER 1 INTRODUCTION........................................................................................................1 Preamble....................................................................................................................... 1 Problematic Situation....................................................................................................1 Researchable Problem..................................................................................................8 Objectives..................................................................................................................... 8 Hypotheses....................................................................................................................9 2 LITERATURE REVIEW...........................................................................................11 3 SUREVEY CONTENTS............................................................................................19 Survey Instrument.......................................................................................................19 Survey Contents..........................................................................................................21 4 DATA......................................................................................................................... 24 5 THEORETICAL MODEL AND MODEL SPECIFICATIONS................................32 Theoretical Model.......................................................................................................32 Probit Model...............................................................................................................33 Model Specification....................................................................................................36 6 EMPIRICAL RESULTS............................................................................................40 Entire Survey Sample Probit Estimates......................................................................40 Hispanic Model...........................................................................................................45

PAGE 6

vi 7 CONCLUSIONS AND IMPLICATIONS.................................................................49 Summary.....................................................................................................................49 Conclusions.................................................................................................................52 Implications................................................................................................................53 APPENDIX A SOUTHERN REGION GOAT CONSUMPTION SURVEY....................................56 B SUMMARY OF DEMOGRAPHIC INFORMATION..............................................81 LIST OF REFERENCES...................................................................................................85

PAGE 7

LIST OF TABLES Table page 1-1. Florida farms with goats, goat numbers and goats sold, goat production region, and United States: 2002 and 1997..............................................................................3 1.2. United States meat goat imports and exports.............................................................6 4-2. Summary of demographic information....................................................................28 4-3. Descriptive statistics on additional factors that influence the willingness to try.....29 4-4. Summary of demographic information for Hispanic respondents...........................30 4-5. Descriptive statistics on additional factors that influence the Hispanic respondents willingness to try.................................................................................31 5-1. Probit model variables and description....................................................................39 6-1. Empirical results from the whole survey probit model............................................41 6-2. Empirical results from the Hispanic probit model...................................................46 vii

PAGE 8

LIST OF FIGURES Figure page 4-1. Comparison of Florida census data and survey respondents by age........................27 4-2. Comparison of Florida census data and Hispanic survey respondents by educational attainment levels...................................................................................28 6-1. Changes in the probability of trying goat meat with respect to household size. (All household sizes are compared to the base, households containing five or more individuals)......................................................................................................43 B-1. Respondents consumption preference for goat meat by ethnicity/race...................81 B-2. Respondents consumption preference for goat meat by gender..............................81 B-3. Respondents consumption preference for goat meat by educational attainment levels.........................................................................................................................82 B-4. Hispanic respondents consumption preference for goat meat by origin.................82 B-5. Hispanic respondents consumption preference for goat meat by gender................83 B-6. Hispanic respondents consumption preference for goat meat by educational attainment levels.......................................................................................................83 B-7. Hispanic respondents consumption preference for goat meat by generation in United States............................................................................................................84 viii

PAGE 9

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science EVALUATION OF CONSUMER PREFERENCES REGARDING GOAT MEAT IN FLORIDA By Erika Knight May 2005 Chair: Lisa House Major Department: Food and Resource Economics Florida is a part of the United States goat production region and its goat inventory increased sixty-three percent between 1992 and 1997. Goat meat has always been a minor food item in the United States, but the importance of meat goats to farm income has increased in recent years. The rapid growth of ethnic populations has led to increased consumption since Hispanics, Muslims, and individuals with African ancestry are major consumers of goat meat products. Florida has a strong potential for a meat goat market; however, the lack of consumer information has hindered producers, processors, and marketers ability to fully delineate the profitability and viability of the industry. The overall objective of this research project is to identify the factors that influence goat meat consumption among Hispanic and non-Hispanic consumers. This information can be used by goat meat industry officials to better understand who their consumer is and further develop their markets. The data for this study were obtained using telephone surveys that targeted Florida residents. The Institute for Behavior ix

PAGE 10

Research and the Survey Research Center at the University of Georgia in Athens, GA, collected the data using the random digit dialing probability technique to minimize bias. The questionnaire sought information on demographic as well as consumer preferences towards goat meat. According to the results from a probit analysis, consumption of other meats and various demographic and psychographic characteristics influenced the willingness to try goat meat. Respondents from households with 5 or more people were most likely to try goat meat, and as household size decreased the willingness to try tended to decline. The results also suggest individuals who had previously consumed lamb were more inclined to try goat meat. x

PAGE 11

CHAPTER 1 INTRODUCTION Preamble Columbus introduced goats to North America in the area that is presently known as Texas during the 1600s and goat production in the South has flourished ever since their introduction (Walsh, 1995). In the past decade, meat goat production has dramatically increased throughout the Southeastern region of the United States, Florida in particular. The meat goat industry in Florida has evolved and producers who once viewed their animals as a sideline operation now consider it a serious business (Simpson, 1995). Consumption of goat meat and goat meat products has become more popular in the United States and the strongest demand for goat meat is along the eastern coast, Southern California, Florida, Detroit, and the northeast region stretching from Washington, D.C. to Boston (McKenzie-Jakes, 2004). The lack of consumer information poses a barrier in goat meat marketing and it has prevented producers, processors, and marketers from fully ascertaining the profitability and viability of the industry. This thesis will use cross-sectional survey data to assist in identifying and understanding the factors that influence the consumption of goat meat in Florida. Problematic Situation A 2000 study by McLean-Meyinsse identified meat goat production as a new enterprise that is believed to enhance farm incomes for small farmers. Goat meat is an extremely popular food item in the Mediterranean, the Middle East, Africa, South America, Central America, the West Indies, and Southeastern Asia. However, goat meat 1

PAGE 12

2 consumption in the United States has been historically low. In recent years, U.S. domestic demand for goat meat has increased with the largest demand along the eastern coast of the United States, especially in Florida. Floridas, goat meat consumption exceeds production (McKenzie-Jakes, 2004) creating a need to import live goats or goat meat in order to satisfy demand (Gipson, 1999). Despite the products increase in popularity, the lack of availability of goat meat in Florida hampers the industrys profitability. Florida is a part of the eleven state southern goat production region stretching from Texas to North Carolina (TX, LA, OK, AK, MS, AL, FL, GA, TN, SC, and NC). In 2002, this region accounted for 80 percent of meat-type goat production in the United States, and Florida contributed three percent of the regions meat goat inventory (USDA-NASS, 2002). During the past few decades, Floridas meat goat industry has transformed from one in which goats were raised as a minor part of subsistence level farm system into a more structured industry oriented approach that is viewed as a business (Simpson, 1995). According to the 2002 Census of Agriculture, 1,764 meat goat farms existed in Florida, a six percent decrease from the total number of meat goat farms in 1997. However, during the same time period, the states meat goat inventory increased twenty-five percent (Table 1-1). In addition between 1997 and 2002, the number of goats sold increased 47. Meat goat farms accounted for over eighty-eight percent of the goat farms within the state in 2002. Growth in Floridas meat goat industry is attributed to a number of factors. Historically, the majority of U.S. immigrants were from European countries, but changes in immigration patterns have occurred.

PAGE 13

3 Table 1-1. Florida farms with goats, goat numbers and goats sold, goat production region, and United States: 2002 and 1997. Geographical Area 2002 1997 Percent Change Florida's Contribution Florida Number of Goat Farms 1,992 2,114 (5.8) Number of Meat Goat Farms 1,764 1,931 (8.6) Numbers of Meat Goats on Farms 36,020 28,737 25.3 Number of Meat Goats Sold 18,769 13,700 37.0 Percent of Meat Goat Farms 88.6 91.3 Percent of Meat Goat Sold 52.1 47.7 Goat Production Region Number of Goat Farms 42,487 37,303 13.9 4.7 Number of Meat Goat Farms 1,729,158 1,727,978 0.1 4.6 Numbers of Meat Goats on Farms 1,433,589 975,931 46.9 2.5 Number of Meat Goats Sold 765,622 407,078 88.1 2.5 Percent of Meat Goat Farms 91.2 88.7 Percent of Meat Goat Sold 53.4 41.7 United States Number of Goat Farms 91,462 76,543 19.5 2.2 Number of Meat Goat Farms 74,980 63,422 18.2 2.4 Numbers of Meat Goats on Farms 1,938,924 1,231,762 57.4 1.9 Number of Meat Goats Sold 1,109,619 532,792 108.3 1.7 Percent of Meat Goat Farms 82.0 82.9 Percent of Meat Goat Sold 57.2 43.3 Source: USDA-NASS, 2002. The majority of current immigrants are people from Hispanics, Caribbean Islanders, Muslims, and Asian populations, as opposed to the historical pattern of immigration from Europe. These changes in immigration patterns from European countries to those from regions of the world that are perceived to have a dietary preference for goat meat have increased the demand for the product (McKenzie-Jakes, 2003; Pinkerton et al., 1994; Sherman, 2002; Walsh, 1995). People from these populations tend to retain food preferences and religious affiliations in an effort to maintain their ethnic identity when merging with a dominant group (Harwell, 1995). Thus, the demand for goat meat among

PAGE 14

4 target consumers is expected to be inelastic (Harwell, 1995; Pinkerton et al., 1994). It is projected that by the year 2025, the U.S. population will increase 44 million due to the increase of foreign populations, many of which consume goat meat. Therefore, the demand for this product in the United States is expected to expand as population increases and as minority purchasing power increases (Russell, 2000). Health conscious consumers and members of the yuppie community who consume the meat as a gourmet item create an additional demand for goat meat products (Harwell, 1995; McKenzie-Jakes, 2004; McLean-Meyinsse, 2003; Pinkerton et al., 1994; Sherman, 2002). Members of the health food sector consume goat meat because of the products nutritional attributes. Goat meat is a naturally lean meat that is low in cholesterol, low in fat and high in proteins (Johnson, 1995). Advancements in Floridas food distribution infrastructure allow for refrigerated goat meat to be delivered from Texas in an inexpensive and quick manner to consumers. Additionally, low ocean freight rates have permitted goat imports from Australia to occur more frequently (Simpson, 1995). Previously, goat meat has previously been viewed as an unusual food item in the United States; however it is now more mainstream. Researchers have found it difficult to accurately estimate per capita consumption levels of goat meat due to discrepancies in goat inventories, auction runs, and slaughter numbers (Gipson, 1999; Harwell, 1995). Therefore, it is popular for researchers to use indirect methods to approximate per capita consumption. Degner and Moss (1999) surveyed wholesalers in Fort Lauderdale, Miami, and Tampa, Florida. These markets are considered to consist of significant populations with preferences for goat meat. Using estimated wholesale sales, the researchers quantified the annual per capita goat meat

PAGE 15

5 consumption to be a mere 0.21 pounds, nearly 20 percent less that per capita consumption levels in 1986. However, a 1999 study conducted by Putnam and Allshouse estimated that the national per capita goat meat consumption significantly increased to about one pound per capita in 1997. Gipson (1999) utilized National Statistical Service data and Foreign Agricultural Service import/export data as two indicators to measure goat meat demand. In 1998 almost 450,000 goats were slaughtered which was a 1,000 percent increase over a 20-year period. It is worth mentioning that the estimates do not include goats slaughtered in non-United States Department of Agricultural facilities or state inspection facilities and farm slaughter or personal slaughter. It is commonly accepted that demand for goat meat is underestimated because the bulk of goat meat consumed is undocumented, i.e. the majority of the slaughtering occurs in non-UDSA inspected facilities (Davis and Willard, 1996). According to the Foreign Agricultural Service data, the United States has been a net importer of goat meat product since 1991 (Gipson, 1999). Table 1-2 illustrates the quantity of goat meat imported to the U.S. and the quantity exported by the United States from 1989 through 1994. Based on the data, the United States does not have an adequate supply of goat meat production to keep up with demand. In 1998, imports rose to 4,500 metric tons, which is equivalent to over 600,000 pounds. The United States imports the majority of its goat meat from Australia and New Zealand. Available research suggests the demand for goat meat is influenced by consumers age, gender, race, household sized, and martial status (McLean-Meyinsse, 2003; Nelson et al., 1999). Additionally, carcass weight and carcass size preferences differ among target populations within the niche market (Pinkerton et al., 1994).

PAGE 16

6 Table 1.2. United States meat goat imports and exports. Year U.S. Imports* U.S. Exports* Balance* 1989 86,067 122,056 35,989 1990 99,353 115,413 16,060 1991 122,932 53,246 -69,686 1992 172,280 60,444 -111,836 1993 136,364 a 3,504 b -132,860 1994 138,481 a None b -138,481 *Values are expressed in metric tons. a These figure probably reflect reduced Australians in exports of goat due to serve drought conditions. b The steep drop in exports as imports fell markedly in 1993 and 1994, thus conserving domestic supplies. Source: Ohio Cooperative Development Center, 2004 Hispanics tend to prefer young kids, weighing 15-25 pounds live weight and young goats weighing about 50 pounds live weight. Hispanics are the fastest growing ethnic group and the largest minority group in the United States. An emphasis on understanding the factors that influence consumption should be placed on Hispanics because they are the only group that has an expected year round demand for goat meat products (Spaugh, 1997). Muslims favor heavier goats than Hispanics and consume goats that are about 70 pounds live weight, male, and intact. Additionally, Jamaicans, Haitians, West Africans, African Americans have a preference for mature goats (Pinkerton et al., 1994). There is a relatively thin body of published literature regarding the marketing of goats and meat goat, but researchers seem to agree on the barriers surrounding the marketing of goat meat. Pinkerton et al. (1992) found that the marketing constraints that hinder the demand for goat meat include an unorganized marketing infrastructure, lack of quality grades, seasonal demand, inconsistent supply, negative consumer attitudes, inconsistent quality, and insufficient research to identify new market and expand existing

PAGE 17

7 markets. Pinkerton et al. (1994) suggest that a lack of marketing information is a result of inadequate interest by many state departments of agriculture and the structure of the industry. Pinkerton et al. (1994) also suggested that in niche markets, located primary in urban centers, demand for goat exceeds domestic supply, with the shortages caused by inefficiencies in markets. Studies by Degner and Moss (1999) and Degner and Locascio (1988) identified the constraints commercial retailers suggest hamper ability to sell goat in Florida as insufficient demand, supply problems, cheaper substitutes, and product form. Growth in Floridas meat goat inventory has been encouraged by numerous factors, particularly in that Americans are becoming more ethnically diverse and health-conscious, and more are revealing greater willingness to try new exotic food products such as rabbit and goat (McLean-Meyinsse, 2000). Many consumers lack knowledge and exposure to goat meat products, creating an additional barrier faced by the industry (Zachery and Nelson, 1992). Studies suggest that once consumers are informed on the products nutritional attributes and preparation methods, consumer interest in consuming the product increases (Miller, 1995; Rhee et al. 2000). For instance, Rhee et al. (2000) found that after participants were educated on goat meats nutritional values more than 60 percent of the respondents reversed their previous perception of the product. Existing research also identifies the carcass size preferred by populations within the niche market, but fails to identify the specific cuts of goat meat that are preferred by populations in the niche market. However, little inductive research has been completed to identify the characteristic of goat meat consumers or the characteristics of the product that influence consumption. Therefore, information that describes the factors that influence

PAGE 18

8 consumption is needed in order to fully understand the marketing and advertising approaches that will assist in increasing the presence of goat meat in the U.S. food industry as well as impacting the goat meat industry. Researchable Problem The meat goat industry has experienced substantial gains in recent years due to the increase in demand for goat meat products. The demand for goat meat products in Florida is expected to continue growing as target populations within the niche market increase. Researchers believe Florida has the potential for a meat goat market that is profitable. However, lack of consumer information has hindered producers, processors, and marketers ability to delineate fully the profitability and viability of Floridas meat goat industry. Consumer information is a critical component to completely understanding the possible economic impact of goat meat productions and marketing in Florida. A relatively thin body of empirical research is available that evaluates the impact of demographic and socioeconomic characteristics on consumers preferences and perception for goat meat products. This research will assist in filling the voids in research by assessing the factors that influence the willingness to try goat meat within the Hispanic and general markets. This study will identify the barriers that restrict consumption and make recommendations for opportunities to increasing consumption of goat meat products within the state of Florida. Objectives The overall objective of this research project is to identify the factors that influence and the barriers that reduce consumption, with a special focus on the Hispanic population. We will collect information for existing and potential goat meat consumption

PAGE 19

9 in Florida and study the effects of perception, consumer preferences, and demographic and socioeconomic characteristics on willingness to try goat meat. The specific objectives for this project are: To develop an understanding of the factors that influences the willingness to try goat meat products for Hispanics and non-Hispanics. To disseminate the results from this study and make recommendations for opportunities of increased consumption of goat meat products to producers and potential producers. Hypotheses Hispanics are the primary consumers of goat meat products and compose the majority of this developing niche market. There are numerous factors that impact preferences, perceptions, and willingness to try goat meat. In previous red meat studies, attributes other than price found to influence a consumers decision to buy the product included safety, availability, advertisement, and government labels (Hui et al., 1995). Thus, these attributes may also be important in the consumer purchase decision of goat meat. Given these considerations, several hypotheses will be tested. The hypotheses are: Increases in household size and educational attainment will positively affect the consumers willingness to try goat meat. Respondents that associate a positive product image with goat meat will be more inclined to express a willingness to try the product. Consumers that rate safety, convenience and cholesterol and fat content as important factors in purchasing meat will be more likely to try goat meat. Ethnicity, age, and gender will influence the probability of trying goat meat. For instances Hispanics are more likely to try goat meat whereas females are less likely to try goat meat. Consumers that rate price specials as important factors in purchasing meat will be least likely to try goat meat products. Lamb consumption will have a positive influence on the willingness to try goat meat.

PAGE 20

10 Hispanics from different geographic origins will have differing levels of willingness to try goat meat.

PAGE 21

CHAPTER 2 LITERATURE REVIEW The body of literature regarding the marketing of goat meat is relatively thin. Numerous sensory evaluation studies have been conducted revealing the overall palatability of goat meat is acceptable among consumers, but their lack of knowledge about the product limits the demand and decreases its profitability within the meat industry (Degner, 1991; James and Berry, 1997; Rhee et al., 2003; Smith et al., 1974). Few studies have focused on the factors that influence goat meat consumption, and none of the available research appears to investigate the attributes that were important to goat meat consumers. Understanding consumers preferences for goat meat is essential so that marketing strategies can be developed to increase the demand for the product. Two previous studies, McLean-Meyinsse (2003) and Degner and Lin (1993) examined consumption of goat meat in the South. McLean-Meyinsse (2003) identified that demographic, socioeconomic, and geographic factors that influenced previous consumption, willingness to try goat eat, and interest in purchasing various goat meat products. The studys sample consisted of 1,421 respondents from Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, and Virginia. A binomial logit was used to estimate the relationship between prior consumption and the selected explanatory variables. Additionally, ordered probit models were used to estimate the probabilities of nonconsumers willingness to eat goat meat and the likelihood of this individual to make future purchases for goat nuggets, patties, roast, or marinated ready-to-cook goat meat 11

PAGE 22

12 products. The study found that goat meat consumption was the highest among older respondents, households with more than three persons, among African-Americans, other non-Caucasian races, men and Texas residents. According to the marginal probabilities, individuals from other races were 15 percent more inclined to consume goat meat products than their Caucasians counterparts and women were 14 percent less likely than men to have previously consumed goat meat. The results from the ordered probit suggested that amongst non-triers, women and residents of Arkansas, Florida, Kentucky, Louisiana, North Carolina, Oklahoma, South Carolina, and Virginia were more likely to consume goat meat within the next month. Younger consumers, households containing less than three persons, African-Americans, or other races were less likely to try goat meat within the next month. Women and residents from Florida, Kentucky, North Carolina, Oklahoma, South Carolina, and Virginia were more willing to try goat meat at restaurants. Respondents least likely to try free goat meat samples in supermarkets were younger, lived in smaller households, and were non-Caucasian. In summary, age, race, household size, religion, gender, and state residency were found to affect consumption. Consumers most likely to eat goat nuggets, patties, or roasts were individuals from larger households, non-Caucasians, men, or Texas residents. Additionally, respondents living in larger households, other races, men or Texas residents were more likely to be willing to purchase marinated, ready to cook goat meat. The results of McLean-Meyinsse (2003) research were consistent with the outcomes of a study conducted by Degner and Lin (1993) that analyzed willingness to consume goat meat at restaurants or home. Data for their study consisted of responses

PAGE 23

13 from consumer surveys conducted in Jacksonville and Tampa, Florida. Three-hundred interviews were conducted in each city and participants were 18 years of age or older. A probit model was utilized to evaluate the impact of demographic, geographic, and socioeconomic factors on the willingness to consume goat meat at home and in restaurants. The study found that if the respondent possessed a positive attitude or perception of the product, they were more likely to order goat meat at a restaurant or purchase the meat for at home consumption. The study found that household income, gender, and household size effected participants willingness to consume goat meat. Respondents from households with annual incomes of less than $10,000 were more likely to order goat at a restaurant than individuals from households with a yearly income between $10,000 and $19,000. Also, consumers between the ages of 35-49 were more inclined to order goat meat in a restaurant than any other age group, but age had no significant affect on purchase intentions. An empirical study conducted by Hui et al. (1995) rated the importance of 12 selected meat attributes among various demographic, geographic, or socioeconomic characteristics. The data for the survey were obtained using a telephone survey that was administered to 1002 randomly selected households in Louisiana and Texas. The primary shopper of each household was interviewed. The Kruskal-Wallis test was used to determine whether the level of importance for each attribute differed amongst the respondents. The simultaneous multiple comparison model ranked the attributes in order of importance. An ordered-probit model was used to estimate the impact of the consumers demographic, geographic, and socioeconomic characteristics on the importance of each attribute. The dependent variables were the 12 selected attributes,

PAGE 24

14 which included low fat content, low sodium content, low in cholesterol, lack of chemical additives, taste, red meat, white meat, appearance, price, freshness, USDA labels, and tenderness and their relative importance while the dependent variables consisted of the demographic, geographic, or socioeconomic characteristics. The results from the ordered-probit model suggest that older consumers were more concerned with prices, a stated preference for red meat, appearance of meat, USDA labels and tenderness of meat. Larger households considered low sodium levels and lack of chemical additives as influential attributes. Low-income households indicated that low sodium content and red meat as valuable characteristics that influence meat purchases. Respondents from high-income households were more concerned about sodium levels and USDA labels and were worried least about fat, cholesterol, and prices. The results from the Huis (1995) study suggest that freshness and taste were the most important attributes to consumers, followed by appearance. The next tier of attributes consumers deemed important by included USDA labels, tenderness, and lack of chemical additives. The last group consisted of nutritional attributes, which included low levels of fat, sodium, and cholesterol. The results indicated that females were more concerned than males about attributes such as fat, sodium, cholesterol, chemical additives, prices, appearance, freshness, tenderness, and USDA labels when making buying decisions. Non-white respondents were more worried about fat, cholesterol, and price than white respondents. According to the results, retailers, wholesalers, and processors should develop a marketing plan that emphasizes the tastiness, appearance, and freshness of the meat and include recipes when promoting meats. In addition, the marketing channels should

PAGE 25

15 minimize transportation and holding time to ensure freshness. This study provided knowledge on the relationship between consumer characteristics and the significance of various meat attributes that may assist in creating effective marketing opportunities for farmers, processors, wholesalers, and retailers in the meat industry. Melton et al. (1996) conducted experimental auctions to evaluate the significance of attributes and how to develop effective marketing plans for pork. The willingness to pay results suggests that the appearance of the meat is most important for first-time buyers and repeat purchasers were interested in the pork chops taste. Melton et al. (1996) concluded that first-time buyers of fresh pork chops may be misled by relying on appearance when making purchases, selecting chops that were less desirable when eaten. As a result, theses consumers were unlikely to make repeat purchases, hampering the products long term market success. Similar to the studies conducted by Hui et al. (1995) and Melton et al. (1996), Chen et al. (2002) examined the relative importance of fresh pork attributes among Asian-origin consumers in San Francisco, California. The results from the Kruskal-Wallis test suggest that freshness is the most important attribute followed by attributes of the color of meat, lowness in fat, and whiteness of fat. The price of fresh pork is also an attribute of considerable importance. The empirical results for the ordered-probit model indicate that particular demographic and socioeconomic characteristics influence the ranking of attributes among consumers. For instance, fat content was more important to highly educated males. They study also found differences within segments of Asian-origin populations; for example, Chinese origin respondents were more price sensitive than other Asian consumers. This final finding is important because one can infer that

PAGE 26

16 Hispanic origin consumers were heterogeneous group and that the significance of various product attributes may differ amongst each identifiable subgroup. In the United States, goat meat is viewed as specialty food item. The study conducted by Schupp et al. (1998), found that consumers expressed resistance to meats that they believed came from exotic animals. Species that respondents in the study considered exotic included but were not limited to deer, alligator, rabbit, goat, emu, and wild duck. Therefore, meat attributes that were deemed significant to consumers may differ for alternative meats. McLean-Meyinsse (2000) conducted a study that utilized ordered probit models to evaluate the impact of socioeconomic, demographic, and geographic characteristics on the primary grocery shoppers attitude toward rabbit; and previous consumption or intent of consuming the product. The data for the study were obtained from a random telephone survey that sought information regarding the meat purchasing and consumption decisions. The sample of consisted of 1,002 primary food shoppers from Louisiana and Texas. As previously mentioned, rabbit meat and goat meat both tend to be considered exotic amongst consumers. The independent variables in the study were various socioeconomic, demographic, and geographic characteristics such as age, income gender, educational attainment, and race, to name a few. McLean-Meyinsse (2000) used an ordered probit model to determine whether the explanatory variables effect the probability of shoppers attitudes towards rabbit meat and the likelihood of consumption or interest in eating it in the future consumption. The results from the attitudinal model suggest that gender, religion, and employment status have a statistically significant effect on shoppers attitude towards

PAGE 27

17 rabbit meat. The marginal effects suggest that men, Catholics, and white-collar workers are more positive about goat meat than their counterparts. Forty-eight percent of male shoppers possessed a positive opinion about rabbit meat, while 54 percent of women held a unfavorable opinion about the meat. The consumption model suggests that gender and employment status impact the likelihood of consuming or interest in consuming rabbit meat. For instance, if the grocery shopper was a male rather a female, the probability of rabbit meat consumption increased by 12.33 percentage points. Females were more willing to express a willingness to try rabbit meat. Also, if shoppers had a positive opinion about rabbit meat, they were more likely to have consumed it. McLean-Meyinsse (1999) also evaluated the marketing outlook for specialty meats such as alligator, goat, and/or rabbit meat in southern states. The objectives of the study were to determine the percentage of individuals shopping at outlets offering specialty products, identify the demographic, socioeconomic, and geographic characteristics influencing shoppers likelihood of buying from these stores, and profile of the consumers most likely to purchase specialty meats. The study analyzed various explanatory variables (i.e. age, household size, education, gender, income, martial status, religion and occupation) that were expected to affect likelihood of non-triers, late-triers, and early triers of specialty foods to shop at specialty stores using an ordered probit model. According to the results, a 42-year old grocery shopper or a shopper from a three person household was more likely to shop at stores offering specialty meats. Shoppers with less than a high school diploma were 18 percent more likely visit these outlets than individuals with higher level of education. Shoppers indicated that price effected their decisions to make purchases at stores offering specialty meats. In fact, there was a 22

PAGE 28

18 percent difference between the early triers and nontriers on the relevance of prices to the meat purchasing decision. It is worth mentioning that the price of specialty meats is usually more expensive than traditional meats such as beef, pork, and chicken. In conclusion, very few empirical studies that focus on goat meat consumption are available in literature. Existing studies conducted by McLean-Meyinsse (2003) and Degner and Lin (1993) identify target consumers for goat meat marketing and promotional efforts, but the studies stop short of identifying the underlying factors that encourage individuals to purchase or be willing to purchase the product. The meat goat industry is still in its developing stages, and before this industry experiences economic success, additional information explaining the demand characteristics for goat meat consumers is needed. Understanding these factors, as well as consumers preferences and perceptions towards goat meat, is important when developing marketing strategies and increasing the presence of goat meat in supermarket.

PAGE 29

CHAPTER 3 SUREVEY CONTENTS Survey Instrument Telephone surveying is a method of collecting data from respondents that is more cost efficient than conducting personal interviews (Dillman, 1978). The primary advantage of telephone surveying is that it provides the researcher the opportunity of controlling and monitoring the data collection process to ensure that data gathered is of high quality, thus providing accurate estimates. For instance, the researcher has the ability to regulate sampling, respondent selection, and questionnaire contents (Lavarakas, 1993). Telephone surveying is more cost efficient when compared to personal interviews because it allows a larger number of interviews in a short period of time. Telephone surveys are usually more expensive than mail surveys but due to the fact that telephone surveys have the potential of minimizing total survey error, this method is generally preferred. Additional advantages associated with telephone surveys include: (1) the scheduling of call-backs to contact hard to reach, but critical respondents, (2) the ability to minimize biased responses, and (3) the capability of the interviewer to clarify questions (Lavarakas, 1993). Despite the advantages associated with telephone surveying, a few disadvantages exist. A major drawback with the telephone surveying technique is that surveyors are unable to reach the cell-phone-only population, making the sample statistically unbalanced because it does not contain both cell phone and land line users. The Associated Press (2005) found that the cell-phone-only population is growing rapidly and 19

PAGE 30

20 it currently accounts for approximately 7 percent of the population. In fact, nearly one in every five individuals between the ages of 15 and 24 had only cell phones. Other limitations are related to the limitations with complexity of the questions asked and the length of the survey. Respondents may grow tired if kept on the telephone for longer than 20-30 minutes. However, respondents tend not to suffer from fatigue when participating in personal interviews and this issue is not applicable to mail surveys because the questionnaire is completed at the respondents leisure (Lavarakas, 1993). Similarly, complicated questions are impossible to ask via telephone. In spite of the disadvantages associated with this surveying technique, the advancements in telephone technology and infrastructure give the researcher accessibility to nearly any population via telephone, making this surveying approach more attractive than other methods (Frey, 1989). The Institute for Behavioral Research, Survey Research Center at the University of Georgia has a history of successfully conducting telephone surveys. This center was used to collect the data for this study. The trained research staff utilized the random digit dialing technique ensuring that all adult Florida residents with landline telephone service had an equal chance of selection for inclusion in the sample, regardless of whether the number is unlisted, which reduces the sampling error (Salant and Dillman, 1994). The telephone survey in this study was administered twice, first to the general population and secondly to Hispanic households. An emphasis was placed on Hispanic households because previous research suggested that these individuals provide a steady year round demand for goat meat products (Spaugh, 1997). Two hundred thirty-seven households participated in the general population survey. For the latter survey, a

PAGE 31

21 telephone directory data base was obtained that consisted of households with Hispanic surnames and it consisted of 198 observations. The surveyed Hispanic households were randomly selected from the directory database. The original survey was translated to Spanish to better serve the Hispanics consumers. Once the potential participants were contacted, the objective of the study was explained to the respondents and they were asked to participate in the study. The surveyor spoke with the primary grocery shopper as they were expected to be primary decision maker in purchasing goat meat. The respondent was asked to complete the survey, which sought information on demographics and consumers shopping preferences for goat meat products. Survey Contents The purpose of the telephone survey was to collect information on consumer shopping preferences for goat meat. The survey sought three categories of information: (1) demographics, (2) family linkages with an emphasis on the introduction to goat meat products and the transmission of food habits from generation to generation including consumption levels, and consumer perception and (3) identifications of factors influencing consumption or the willingness to consume. All survey participants had to be 18 years old or older. If the respondent was unavailable to complete the interview, a callback was scheduled. The questionnaire used in this study is included in the Appendix. The dependent variable in this study was the willingness to try goat meat. To estimate this variable, respondents were placed in one of three categories: respondents that had previously consumed goat meat, respondents willing to try, and respondents that have not consumed and are unwilling to try goat meat. Respondents were first asked if

PAGE 32

22 they had ever consumed goat meat. If the respondents had previously consumed goat, they were then asked a series of questions regarding the size, age, cut, etc of the meat consumed. These questions were asked to develop an understanding of the type of product desired by different ethnic groups. Previous research suggested Hispanics, Muslims, African-Americans, Haitians, and Jamaicans consume goat meat product, but carcass preferences differs among each group. To estimate current consumption levels, respondents that had previously consumed goat were asked to quantify how many pounds their family consumed each year. To identify potential consumption, respondents that had never consumed goat meat were asked if they were willing to consume if it was available in stores. Research suggests that availability is a major obstacle faced by the goat meat industry, thus, restricting consumption. Locascio and Degner (1988) surveyed supermarket representatives in Florida and found that 28 of 168 or16.7 percent, of the stores run by six chains sold goat meat. A more recent study conducted by Degner and Moss (1999) found that 18 percent of meat wholesalers in Florida sold goat meat. As previously mentioned the dependent variable for the study is the willingness to try goat meat; therefore, respondents that had previously consumed or willing to try if available in grocery stores are considered those willing to try goat meat and all other respondent are non-triers. The next section of the questionnaire solicited information on the psychographic factors that consumers believed to have importance when making decisions to purchase meats. Previous meat studies indicated freshness, convenience, and sodium and cholesterol content influence meat consumption. However, because goat meat is a specialty food item, it is unknown if these characteristics are equally important to goat

PAGE 33

23 consumption. Psychographic factors are needed because they provide information necessary to develop an understanding of how of consumers feel and think about the product (Peter and Donnelly, 2003). Respondents were asked to rate the relative importance of factors such as price specials, convenience, safety, cholesterol, and fat to meat purchases. Participants were also asked how they viewed goat meat. The responses could have ranged from very positive to very negative. It was expected that this portion of the survey would reveal consumers knowledge about goat meat. Information regarding respondents consumption of other meats such as chicken, beef, and seafood was also collected because according to demand theory, consumption is expected to be affect by substitutable and complementary products. Participants were asked the frequency and then quantity of the alternative meat consumed. First, respondents were asked to specify if the selected food item was consumed everyday, more than once a week, once a week, more than monthly, monthly, on special occasions, or never. If the respondent indicated he/she had consumed the meat, respondent was then asked to identify the quantity consumed per sitting. When consumers make a decision to try or consume a product, they usually evaluate the alternative products available. This study incorporated a variable that captured the consumption of other meats to evaluate the relationships between other selected meats and the willingness to try goat meat. The final section of the questionnaire solicited socio-economic and demographic information, such as household income, household size, educational attainment, gender, and race. Based on previous empirical studies, this information is expected to be significant when analyzing the factors that influence goat meat consumption.

PAGE 34

CHAPTER 4 DATA The data for this study were attained through a telephone consumer survey in conjunction with the SARE #LS502-138, An Investigation of the General Goat Meat Demand and the Sustainability of Goat Production. Adult consumers within the state of Florida were surveyed via telephone to establish consumer preferences for goat meat and the level of consumption and potential levels of consumption. The Survey Research Center at the University of Georgia collected the data during the Spring of 2004. Florida was targeted because the strongest demand for meat goats is found along the East Coast, especially within Florida (McKenzie-Jakes, 2004). In addition, Hispanics were targeted because these consumers are expected to provide a more stable demand for goat meat products throughout the year (Spaugh, 1997). Thus, understanding the factors that effect their demand for goat meat is imperative. Florida households were surveyed utilizing random digit dialing to assess the factors influencing their consumption decisions of goat meat. As previously mentioned, telephone surveying is an effective method of data collection that allows a large amount of data to be collected in a short period of time. Additionally, random digit dialing ensures that all residents with a landline telephone have an equal chance of being included in the sample. Selection response is minimized, and inferences about Floridas adult population can be made with greater assurance from the results obtained in the survey. Four hundred thirty-five consumers participated in the survey, and data is summarized in Appendix B. However, due to incomplete demographic and 24

PAGE 35

25 socioeconomic information such as age, educational attainment, and gender, sample observations were deleted. Thus, the information from 365 responses is summarized in this analysis. When the sample results are compared with US Census data for Florida (US Census Bureau, 2000) Hispanics were over represented within the sample, accounting for 41.4 percent while making up only 16.1 percent Floridas population. However, Caucasians and other races were underrepresented in the sample making up 50.1 percent and 15.5 of the sample, respectively. An emphasis was placed on Hispanic consumers because these individual are perceived to have a historical preferences towards goat meat; therefore, the results were expected to be biased towards Hispanics. Among the Hispanic respondents, few inconsistencies exist between their geographic origins for their respective populations and sample, making the two reasonably comparable. For instance, Cubans represent 31.1 percent of the Hispanic population in Florida and 25.8 percent of the sample, Mexicans account for 13.6 percent of the population and 16.6 percent of the sample, Puerto Ricans represent 18 percent of the population and 16.6 percent of the sample, and all other Hispanics descents account for 37.4 percent of the population and 41.1 percent of the sample. The majority of the survey respondents were female, 70.1 percent. The results are biased towards females because the questionnaire targeted the primary grocery shopper as a means of accurately estimating consumption history, the willingness to try, and psychographic characteristics sought by consumers and potential consumers of goat meat. According to literature, 70 percent of all females in the United States were considered the primary grocery shopper (Progressive Grocer, 2002).

PAGE 36

26 Figure 4-1 illustrates the percentage of Floridas population and survey respondents in various age groups. The sample tolerance for this study is +/5.1 percent. With this in mind, the sample and population are comparable despite the small discrepancies that exist. The samples median age fell in the 35-44 age category, comparing to the median age in Florida, 38.6. The samples average household consisted of three people, which exceeds the states average household size of 2.46 individuals. Additionally, the average household size for Floridas Hispanic population was 3.12 persons, slightly less than that of the sample, 3.32 persons. Finally, survey respondents were more educated than average. According to survey response, 47.7 percent of the respondents possessed either a high school diploma or lower, 26.5 percent had some college education, and nearly 25.8 percent had at least a college degree, which compares to the populations 49.8 percent, 29.6% and 20.6 % for the respective categories. Likewise, 52.3 percent of the Hispanic respondents had more than a high school diploma, only 40.1 percent of the Hispanic population had surpassed this mark. Furthermore, 25.8 percent of the Hispanic sample and 15.6 percent of the population had received degrees from a four year college or more advanced degree (Figure 4-2). Among the 365 respondents, 43.6 percent expressed a willingness to try goat meat if it were available in food stores (159 respondents), whereas, 56.4 percent were uninterested in the product (206 respondents). Male respondents were more likely to try goat than females, 52.3 percent and 39.8 percent, respectively. Hispanics were more likely than Caucasians, but less likely than other races to express a willingness to try goat meat. Over 60 percent of respondents that perceived goat meat in a positive manner

PAGE 37

27 indicated that they would tr y goat meat. However, only 6.1 percent of respondent with a negative view of goat meat were willing to try the product. Other explanatory variables incorporated in the study include the frequency at which goat meat substitutes are consumed, psychographic characteristics, and various demographic and socioeconomic variables. De scriptive statistics fo r the survey sample are found in Table 4-2 and Tabl e 4-3. Since a separate an alysis involving Hispanic respondents will be conducted, reasoning is ex plained in the next chapter, descriptive statistics for only the Hispanic respon dents are shown in Tables 4-4 and 4-5. Figure 4-1. Comparison of Fl orida census data and survey respondents by age.

PAGE 38

28 Figure 4-2. Comparison of Fl orida census data and Hisp anic survey respondents by educational attainment levels. Table 4-2. Summary of demographic information. Triers NonTriers Overall Sample Number of Observations 159 206 365 Ethnicity % % % Hispanic 45.9 37.9 41.4 Caucasian 44.0 54.9 50.1 Other races/ethnicities 10.1 7.3 8.5 Gender Percent Female 64.2 74.8 70.1 Education HS Diploma or less 37.7 37.9 37.8 Some College 26.4 25.2 25.8 College Degree and above 35.9 36.9 36.4 Household Size 1 Only 9.4 11.2 10.4 2 People 34.0 33.0 33.4 3 People 18.2 17.0 17.5 4 People 17.0 25.2 21.6 5 and above 21.4 13.6 17.0 Age of Respondents 18-24 7.6 8.7 8.2 25-34 23.9 17.5 20.3 35-44 13.8 19.9 17.3 45-54 18.9 27.5 18.1 55-64 18.2 12.1 14.8 65 and older 17.6 24.3 22.2

PAGE 39

29 Table 4-3. Descriptive statistics on additional factors that influence the willingness to try. Triers Non-Triers Overall Sample Frequency of Chicken Consumed % % % More than once a week 78.0 79.1 78.6 Once a week 17.6 13.6 15.3 Special Occasions 4.4 5.3 5 Never 0.0 1.9 1.1 Frequency of Beef Consumed More than once a week 52.8 49.0 50.7 Once a week 24.5 30.1 27.7 Special Occasions 18.9 15.1 16.7 Never 3.8 5.8 4.9 Frequency of Pork Consumed More than once a week 23.3 17.0 19.7 Once a week 23.3 24.8 24.1 Special Occasions 41.5 40.3 40.8 Never 12.0 18.0 15.3 Frequency of Fish/Seafood Consumed More than once a week 43.4 36.4 39.5 Once a week 34.6 23.3 28.2 Special Occasions 18.9 31.6 26.0 Never 3.1 8.7 6.3 Frequency of Lamb Consumed Previously Consumed 51.6 25.3 36.7 Never 48.4 74.8 63.3 Frequency of Turkey Consumed Previously Consumed 88.7 85.0 86.6 Never 11.3 15.1 13.4 View of Goat Meat Positive 60.4 16.0 35.3 Neutral 33.3 38.4 36.2 Negative 6.3 45.6 28.5 Importance of Cholesterol to Meat Purchasing Important 83.6 78.9 80.2 Unimportant 16.4 21.2 19.8 Importance of Fat in Purchasing Meat Important 86.6 87.4 86.8 Unimportant 13.8 12.6 13.2 Importance of Convenience in Purchasing Meat Important 67.3 61.7 64.1 Unimportant 32.7 38.4 35.9 Importance of Price to Meat Purchasing Important 83.0 65.1 72.9 Unimportant 17.0 35.0 27.1 Importance of Safety to Meat Purchasing Important 95.6 90.8 92.9 Unimportant 4.4 9.2 7.12

PAGE 40

30 Table 4-4. Summary of demographic information for Hispanic respondents. Triers Nontriers Overall Sample Number of Observation 73 78 151 Descent % % % Mexican 17.8 15.4 16.6 Cuban 28.9 23.1 25.8 Puerto Rican 15.1 18.0 16.6 Other Descents 38.7 43.6 8.5 Generation in U.S. First generation 65.8 60.3 62.9 Other generation 34.2 40.7 37.1 Gender Percent Female 67.1 74.4 70.9 Education HS Diploma or less 49.3 46.2 47.7 Some College 24.7 28.2 26.5 College Degree and above 26.0 25.6 25.8 Household Size 1 Only 5.5 9.0 7.3 2 People 24.7 21.8 23.2 3 People 17.8 24.4 21.2 4 People 23.3 29.5 26.5 5 and above 28.8 15.4 21.9 Age of Respondents 18-24 11.0 5.1 8.0 25-34 28.8 28.2 28.5 35-44 16.4 20.5 18.5 45-54 20.6 16.7 18.5 55-64 11.0 11.5 11.3 65 and older 12.3 18.0 15.2

PAGE 41

31 Table 4-5. Descriptive statistics on additional factors that influence the Hispanic respondents willingness to try. Triers Non-Triers Overall Sample Frequency of Chicken Consumed % % % More than once a week 84.9 83.3 84.1 Once a week 11.0 7.7 9.3 Special Occasions 4.1 5.1 4.6 Never 0.0 3.9 2.0 Frequency of Beef Consumed More than once a week 54.8 38.5 46.4 Once a week 27.4 34.7 31.1 Special Occasions 16.4 15.4 15.9 Never 1.4 11.5 6.6 Frequency of Pork Consumed More than once a week 24.7 18.0 21.2 Once a week 26.0 18.0 21.9 Special Occasions 34.3 32.0 33.1 Never 15.0 32.0 23.8 Frequency of Fish/Seafood Consumed More than once a week 42.5 35.9 39.1 Once a week 37.0 26.9 31.8 Special Occasions 15.0 25.6 20.5 Never 5.5 11.5 8.6 Frequency of Lamb Consumed Previously Consumed 42.5 24.4 33.1 Never 57.5 75.6 66.9 Frequency of Turkey Consumed Previously Consumed 79.5 69.2 74.2 Never 20.5 30.7 25.8 View of Goat Meat Positive 72.6 29.5 50.3 Neutral 23.3 39.7 31.8 Negative 4.1 30.8 17.9 Importance of Cholesterol to Meat Purchasing Important 93.2 88.5 90.7 Unimportant 6.8 11.5 9.3 Importance of Fat in Purchasing Meat Important 89.0 88.5 88.7 Unimportant 11.0 11.5 11.3 Importance of Convenience in Purchasing Meat Important 72.6 60.3 66.2 Unimportant 27.4 40.7 33.8 Importance of Price to Meat Purchasing Important 87.7 62.8 74.8 Unimportant 12.3 37.2 25.2 Importance of Safety to Meat Purchasing Important 98.6 92.3 95.4 Unimportant 1.4 7.7 4.6

PAGE 42

CHAPTER 5 THEORETICAL MODEL AND MODEL SPECIFICATIONS Theoretical Model Neoclassical demand theory indicates that the determinants of demand may be categorized under four headings: (1) population size and its distribution by various demographic and geographical characteristics, (2) consumer incomes, (3) prices and availability of complements and substitutes, and (4) consumer tastes and preferences. The homogeneity condition provides a theoretical basis for consumer behavior. The condition states that the sum of the own-price elasticity, cross-price elasticity and the income elasticity for a given commodity equals zero (Tomek and Robinson, 1990): E ii + E i1 + E i2 + + E iy = 0, where E ii is the own price elasticity, E i1 E i2 are the cross-price elasticities and E iy is the income elasticity. The condition implies that the substitution and the income effect on own-price change must be consistent with the cross-price and income price elasticites for a particular commodity. Goat meat prices are very volatile and have been unavailable for many years; therefore, this research does not attempt to estimate demand nor calculate the elasticities. The prices of substitutes were not included in the study; however, information about the relationship between goat meat and its substitutes may give insight on the cross-price elasticites. The central focus of this study is to develop an understanding of factors influencing goat consumption and the willingness to try goat meat. Sensitivity of willingness to try goat meat with price and income is assessed in qualitative terms. 32

PAGE 43

33 This study will examine factors influencing willingness to try goat meat among Hispanic and non-Hispanic consumers using data obtained from a telephone survey administered in Spring 2004. The survey instrument has been discussed in the previous chapter. This study utilizes a probit analysis to estimate the factors that influence the dependent variable, willingness to try goat meat. Since research suggested Hispanics have a historic preference for goat meat, two models are used: (1) an estimation of factors that influence consumption for the entire sample and (2) one that evaluates the factors that influence the willingness to try goat meat among Hispanics. Probit Model Due to their popularity, linear regressions models may be one of the most misused analytical techniques in the social sciences (Aldrich and Nelson, 1984). Linear regression models assume that the dependent variable is continuous; therefore, when the endogenous variable is qualitative, the estimates from the regression analysis may be robust in errors, causing inaccurate statistical inferences (Aldrich and Nelson, 1984). The endogenous variable in this study is a yes or no variable, the willingness to try goat meat. When the regressand is discrete rather than continuous a different analytical technique is needed. Probit models estimate the probability of the binary dependent variable, y, occurring given K observable, explanatory variables, k = 1,, K. Each of the observations on y, y 1 y 2 ,,y N are statistically independent of each other, ruling out serial correlation. Additionally, the model assumes that data are generated from a random sample of size of N observations, with each sample point indicated by i, i = 1,, N (Aldrich and Nelson, 1984). Probit analysis requires that there is no exact linear

PAGE 44

34 dependence among ik s. This implies the number of observations exceed the number of explanatory variables, N>K, that there is variation among each explanatory variable across the observations, and that no two or more ik s are perfectly correlated. The expected outcomes of the dependent variable, y i are considered to be mutually exclusive and exhaustive (Gujarati, 2003). Probit models assume the dependent variable depends on a latent variable, y i *, which is and observed and determined by one or more independent variables. y i = i + i i ~ NID(0, 2 ) The larger y i *, the greater the probability of an event, y, occurs. An event is assumed to occur if the utility differences exceed a certain threshold level. Probit analysis follows a cumulative normal probability distribution with the same mean and variance, providing information on the nature of the latent variable and its parameters. The dependent variable, y, may take on values of zero or one and if the latent variable is defined as y*, then the probit model is described as follows: y i = i + i i ~ NID(0, 2 ) y i = 1 if y* > 0 = 0 if y* 0, The point of interest relates to the probability of the event occurring, Y=1. Utilizing the information above, we have: P(y i = 1) = P(y i > 0) = P( i + i > 0) = P( i < i ) = ( i ), where denotes the cumulative distribution of i (Verbeek, 2004). Maximum likelihood estimation techniques are used to obtain the value of the parameters, that maximize the probability of observing the outcome, y. The maximum

PAGE 45

35 likelihood estimation model is nonlinear and asymptotic, producing better results as the sample size increases. It produces estimates that are nonbiased (estimates are centered around the true values on average), efficient (no other unbiased estimator has lower sampling variance) and normal (we can know how to perform hypothesis testing and draw other inferences) (Aldrich and Nelson, 1984). Maximizing and then taking the first derivative of the log likelihood function produce the parameters for each explanatory variable. The log likelihood function is as follows: log L() = y i log F( i ) + (1 y i ) log y i log F(1 ( i )), where is included in the probabilities to accentuate that the likelihood function is a function of The parameters derived from the log likelihood function are known as marginal effects or marginal probabilities. The marginal probabilities measure the change in probabilities resulting from a unit change in one of the regressors while holding the other regressors constant. Predicted marginal probabilities assist in understanding the relationship between the dependent and independent variables and the signs of the parameter estimates and their statistical significance indicate the direction of the relationship (Gujarati, 2003; Verbeek, 2004). A goodness of fit measure is a summary statistic suggesting the accuracy with which the model approximates the observed data (Verbeke et al., 2000). When the dependent variable is qualitative the accuracy of the model is determined by comparing the fit between the calculated probabilities and observed response frequencies or through the models ability to forecast observed responses. Goodness of fit measures are usually based on a comparison between a model that contains only a constant as the independent variable. The pseudo R 2 takes into account the two likelihood values, log L 1 and log L 0

PAGE 46

36 where L 1 represents the maximum log likelihood value of the model of interest and L 0 stands for the maximum value of the log likelihood function when the intercept is the only parameter value that is not equal to zero. The difference between the log L 1 and log L 2 serves as an indicator of the explained variation of the underlying latent variable caused by the additional parameters (Laitila, 1993; Verbeek, 2004). In summary, the pseudo R 2 is a tool used to evaluate the explained variation in a model. It is important to mention that this measure has two shortcomings: (1) the pseudo R 2 usually decreases as additional parameters are included in the model and (2) the measure does not adjust for the degrees of freedom of the model (Laitila, 1993). Model Specification Existing empirical studies provide the basis for the variables selected in the model. Earlier studies conducted by McLean-Meyinsse (2003) and Degner and Lin (1993) used a probit analysis to evaluate the factors that influence consumers willingness to consume goat meat and goat meat products. Their studies indicated that race, age, household size, geographical location, and gender affect the willingness to consume or try goat meat. For example, the studies found non-Caucasians, men, and those living in larger households were most likely consumers of goat. Studies conducted to assess the factors that influence the consumption of specialty meats McLean-Meyinsse (1999) and McLean-Meyinsse (2000) found ethnicity, education, household size, and gender influenced consumers attitude toward exotic animal food item and their willingness to consume. Finally, studies by Hui et al. (1995) and Chen et al. (2002) used probit model simulations evaluated the impact selected meat attributes had on meat consumption among various demographic, socioeconomic, and geographical characteristics. The results from the Hui et al. (1995) study suggested that female and non-white consumers

PAGE 47

37 are more concerned with fat, cholesterol, and price. Chen et al. (2002) suggested segments within Asian populations have specific taste and preferences and are not a homogenous group. In this study, probit models are used to estimate the willingness to try goat meat for Hispanic and non-Hispanic populations with respect to several explanatory variables. The entire survey sample model will included demographic and socioeconomic factors (i.e. age, gender, income, household size, and educational attainment), perception of goat meat, frequency of other meat consumption, and consumer characteristics. It is believed that segments within the Hispanic population have different variables affecting their willing to try goat meat. Additional independent variables that will be included only in the Hispanic model are descent and generation in the U.S. Specification of the probit for the entire survey sample is as follows: Y ki = k1 ETH2 + k2 ETH3 + k3 GENDER + k4 AGE1 + k5 AGE2 + k6 AGE3 + k7 AGE4 + k8 AGE5 + k9 HSIZE1 + k10 HSIZE2 + k11 HSIZE3 + k12 HSIZE4 + k13 BEEF1 + k14 BEEF2 + k15 CHICK1 + k16 CHICK2 + k17 FISH1 + k18 FISH2 + k19 PORK1 + k20 PORK2 + k21 TURK + k22 LAMB + k23 VIEW1 + k24 VIEW2 + k25 FAT + k26 SAFETY + k27 CONVEN + k28 PRICE Specification of the Hispanic model is as follows: Y ji = j1 MEXICAN + j2 PTRICAN + j3 OTHERD + j4 GENERA + j5 AGE1 + j6 AGE2 + j7 AGE3 + j8 AGE4 + j9 AGE5 + j10 HSIZE1 + j11 HSIZE2 + j12 HSIZE3 + j13 HSIZE4 + j14 BEEF1 + j15 BEEF2 + j16 CHICJ1 + j17 CHICJ2 + j18 FISH1 + j19 FISH2 + j20 PORJ1 + j21 PORJ2 + j22 TURJ + j23 LAMB + j24 VIEW1 + j25 VIEW2 + j26 FAT + j27 SAFETY + j28 CONVEN + j29 PRICE

PAGE 48

38 1 if respondent is willing to try goat meat 0 if respondent is unwilling to try goat meat The probit model estimates the influence the selected explanatory variables have on consumers preferences of goat meat. The analysis also predicts the probabilities of the consumers willingness to try goat meat under several variable levels. A description of the variables used in this study can be seen in Table 5-1. Y i =

PAGE 49

39 Table 5-1. Probit model variables and description. Variant Variable Name Description Willingness to try goat meat TRY 1 if willing to try goat meat Ethnicity ETH 1 if Hispanic, 0 otherwise Hispanic Origin MEXICAN 1 if of Mexican, 0 otherwise CUBAN 1 if of Cuban, 0 otherwise PTRICAN 1 if of Puerto Rican, 0 otherwise OTHERD 1 if of Mexican, 0 otherwise Gender GENDER 1 if male, Age AGE1 1 if 35 or older, 0 otherwise Education EDU1 High School diploma or less EDU2 Some College EDU3 College 4 year degree and beyond Household Size HSIZE1 1 if one person HSIZE2 1 if two people HSIZE3 1 if three people HSIZE4 1 if four people HSIZE5 1 if five or more people Perception of Goat Meat VIEW1 1 if positive view VIEW2 1 if neutral view VIEW3 1 if negative view Consumer Attributes FAT 1 if fat is important to the consumer SAFETY 1 if safety is important to the consumer CHOLES 1 if convenience is important to the consumer CONVEN 1 if convenience is important to the consumer PRICE 1 if price specials are important to consumer Substitutes Consumed BEEF1 1 if consumed more than once a week BEEF2 1 if consumed once of week BEEF3 1 if consumed during monthly or less frequently BEEF4 1 if never consumed CHICK1 1 if consumed more than once a week CHICK2 1 if consumed once of week CHICK3 1 if consumed during monthly or less frequently CHICK4 1 if never consumed LAMB1 1 if consumed, 0 otherwise PORK1 1 if consumed more than once a week PORK2 1 if consumed once of week PORK3 1 if consumed during monthly or less frequently PORK4 1 if never consumed TURK1 1 if consumed, 0 otherwise

PAGE 50

CHAPTER 6 EMPIRICAL RESULTS Using data collected from a consumer survey, the specification set forth in the previous chapter and maximum likelihood procedures, two independent probit models were estimated with the dependent variable representing the consumers willingness to try goat meat. Due to the nature of the dichotomous dependent variable, a probit analysis was utilized to predict the likelihood of trying goat meat given various exogenous variables. The probit model coefficients and marginal probabilities from the two models one for the survey population and one for only Hispanic respondents are shown in Tables 6-1 and 6-2, respectively. According to Greene (2003), marginal probabilities should be used to draw inferences about the relationship between the dependent and independent variables rather than coefficient estimates. The marginal probabilities measure the change in probability of the willingness try goat meat from a unit change in one of the explanatory variables, while holding the other regressors at their sample means. The results from these models are discussed in this chapter, beginning with the whole population model and followed by the Hispanic only model. Entire Survey Sample Probit Estimates The entire survey sample probit model (Table 6-1) correctly predicted 75.8 percent of consumers responses (incorrectly predicting both a consumers willingness to try goat meat and non-willingness to try 12.1 percent of the time). This compares to a nave, which resulted in correct prediction 56.4 percent of the time. The chi-squared 40

PAGE 51

41 value is 137.0 is statistically significant to the .01 confidence level, which implies good predictive power of the variables included in the model. Table 6-1. Empirical results from the whole survey probit model. Variable Estimated Coefficient Standard Error Marginal Effects ETH2 0.0554 0.2020 0.0217 ETH3 0.0424 0.3031 0.0165 GENDER -0.2985 0.1783 -0.0830 AGE1 0.0404 0.3484 0.0159 AGE2 0.0862 0.2719 0.0317 AGE3 -0.0493 0.2898 -0.1922 AGE4 0.0780 0.2536 0.0307 AGE5 0.4215 0.2606 0.1668*** EDU2 0.0883 0.2006 0.0347 EDU3 0.0761 0.3031 -0.2973 HSIZE1 -0.6960** 0.3234 -0.2235** HSIZE2 -0.4697*** 0.2451 -0.1790** HSIZE3 -0.5380** 0.2606 -0.1984** HSIZE4 -0.6253* 0.2508 -0.2295* VIEW1 1.6244* 0.2286 0.5829* VIEW2 0.7787* 0.2070 0.3019* FAT -0.5495** 0.2818 -0.2165** SAFETY -0.2617 0.3042 -0.1037 CHOLES 0.1972 0.2471 0.0760 CONVEN -0.0671 0.1691 -0.0263 PRICE 0.5373* 0.5376 0.2017** BEEF1 -0.1805 0.2132 -0.0709 BEEF2 -0.2505 0.2305 -0.0967 CHICK1 -0.7486** 0.3215 -0.2918* CHICK2 -0.7544** 0.3728 -0.2662** PORK1 0.0635 0.2136 0.0249 PORK2 -0.1158 0.1955 -0.0451 FISH1 0.3839** 0.1951 0.1505** FISH2 0.4458** 0.2103 0.1756** TURK 0.0635 0.2502 0.0318 LAMB 0.4887** 0.1690 0.1915* LogLikelihood = -181.45 R 2 (Psuedo) = 27.2 % Chi-squared = 137.03 % of corrected predictions = 75.9% *,**,*** indicate significance at .01, .05, and .10 levels, respectively

PAGE 52

42 The results indicate that there is no significant difference between the willingness to try amongst demographic characteristics such as ethnicity, gender, nor educational attainment levels. Caucasians or consumers of other races levels of willingness to try were compared to those of Hispanics consumers, the base. Unexpectedly, the willingness to try goat meat amongst ethnic groups was not statistically different, even though apparent differences existed between each groups responses. Raw statistics indicated that 24.5 percent of Hispanics currently ate goat meat, compared to 12.0 percent of Caucasians. However, 31.6 percent of non-goat meat consuming, Hispanics were willing to goat meat, compared to 29.8 percent of non-consuming Caucasians. Statistically significant demographics variables include household size and age. As hypothesized, if less than five individuals were present in a household the chances of trying goat meat diminishes (Figure 6-1). For example, as the household changes from five or more people to one person, the willingness of trying goat meat decreased by 22.3 percent. Additionally, as household size changes from the base, five or more individuals, to a two, three, or four person household the likelihood of trying goat decreased 17.9, 19.8, and 23 percent, respectively. Respondents age also had a significant effect on the willingness to try goat meat, with respondents between the ages of 55 and 64 more 42.2 percent likely the participants 65 years and older to try goat meat. The study failed to investigate the reasons respondents were willing or not willing to try goat meat and goat meat attribute that influence consumption such as, but not limited to freshness, price, presence of chemical additive, and various nutritional attributes. However, the survey did focus on the psychographic characteristics that are important to consumers when making all meat purchases.

PAGE 53

43 Figure 6-1. Changes in the probability of tryi ng goat meat with respect to household size. (All household sizes are compared to the base, households containing five or more individuals). The psychographic factors were included in the model to provide insight on potential goat meat consumers. Of convenience, price sp ecials, fat content, cholesterol, and safety, price specials and fat content were the only significant variables. The relationships between price specials and fa t content with willingness to try were the opposite of the original hypotheses. Consumer s that viewed price specials as important were 20 percent more likely than consumers that rated price sp ecial as unimportant to be willing to try goat meat. Additionally, consumers that pe rceived fat content as relevant were 21.6 percent less likely to be willing to try goat meat than those that view fat levels as insignificant. These results reveal that many consumers are not knowledgeable of the on the characteristics, of goat meat, including pr ice. As mentioned in previous chapters, goat meat is a lean meat and a high source of proteins. Therefore, consumers that believe fat levels are important should be more inclined to try goat meat. One potential

PAGE 54

44 explanation for the opposite result is that people believe goat meat in high in fat. Also, goat meat prices are usually higher than traditional meats (McLean-Meyinsse (1999) found that goat prices range from $1.79 to $2.79 per pound in Louisiana), suggesting those that believe price specials are important should have been less willing to try goat meat. The same potential explanation, lack of knowledge about goat meat, could explain this result. It is not alarming that convenience was not significant. If respondents lack information on preparation methods, they may be unaware if goat meat is easy to cook or not. The probit model indicated that consumers perception of goat meat had a statistically significant impact on the likelihood of trying goat meat. As the consumers perception changed from negative to positive, the probability of willingness to try goat meat increased 58.3 percent. Likewise, if consumers possessed a neutral view of goat meat rather than a negative view the likelihood of consuming goat meat increased 30.2 percent. Based on the results, one can assume that as the consumers attitude towards goat meat becomes more positive, their chances of trying goat meat increases at significant rates. Finally, the relationship between trying goat meat and the frequency of consumption of other meat substitutes was examined. Chicken, fish, and lamb consumption were found to significantly affect the willingness to try goat meat. Respondents that had previous consumed lamb were more willing to try goat meat. In fact, participants that had consumed lamb were 48.9 percent more likely to be willing try goat meat than individuals that had not eaten lamb. The probit estimates also indicated that fish and other seafood consumption positively effect the likelihood of trying goat

PAGE 55

45 meat. Participants that consumed seafood more than once a week were 15 percent more likely to express a willingness to try goat meat. Whereas, respondents that consumer fish weekly were 17.6 percent more incline to be willing to try goat meat. Respondents that consume chicken more than once a month are less likely to be willing to try goat meat. If a respondent consumed chicken more than once a week, the probability of consumer goat meat decreased by 29.1 percent. Likewise if the respondent consumed chicken at least once a week, the chances of trying goat meat declined 26.6 percent. For the results, one can infer that goat meat consumption will occur less often as the frequency of chicken consumed increases. On the other hand, fish and lamb consumption may serve as an indicator of potential goat consumption. Hispanic Model The probit model that focused on the Hispanic respondents only correctly predicted consumers willingness to try goat meat 76.2 percent of the time (incorrectly predicting a consumers willingness to try goat meat 12.5 percent of the time and non-willingness to try 11.3 percent of the time). This is better than nave prediction, 51.7. The chi-squared value is 64.3 and is statistically significant to the .01 confidence level, which implies this model has good predictive ability of forecasting the willingness to try. Although ethnicity was statistically insignificant in the general population model, a separate probit analysis was conducted using Hispanic respondents to reveal if different factors affected the willingness to consumers between Hispanics and the general population.

PAGE 56

46 Table 6-2. Empirical results from the Hispanic probit model. Variable Estimated Coefficient Standard Error Marginal Effects MEXICAN -0.4283 0.4764 -0.1673 PTRICAN -0.7114 0.4405 -0.2690*** OTHERD -0.2024 0.3570 -0.8056 GENERA -0.1514 0.2801 -0.0603 GENDER 0.1312 0.3171 0.0522 AGE1 2.0406* 0.7452 0.5417* AGE2 0.1318 0.4631 0.0525 AGE3 0.1973 0.5136 0.0785 AGE4 0.3302 0.4392 0.1307 AGE5 0.3676 0.5432 0.1448 EDU2 0.0520 0.3108 0.0275 EDU3 0.1250 0.3545 0.4983 HSIZE1 -0.3877 0.5957 -0.1510 HSIZE2 -0.2943 0.3831 -0.1164 HSIZE3 -0.8757* 0.4041 -0.3267* HSIZE4 -0.5672 0.3782 -0.2206 VIEW1 1.4971* 0.4403 0.5457* VIEW2 0.8390 0.4239 0.1521 FAT -0.6843 0.5699 -0.2603 SAFETY -1.0338 0.7084 0.3605* CHOLES 0.3370 0.6091 0.1321 CONVEN 0.1072 0.2887 0.0427 PRICE 1.2540* 0.3541 0.4469* BEEF1 0.3021 0.3544 0.1201 BEEF2 0.3916 0.3902 0.1551 CHICK1 -1.5846* 0.6051 -0.5132* CHICK2 -1.6177** 0.7845 -0.4825* PORK1 0.5338 0.3672 0.2087 PORK2 0.3141 0.3477 0.1245 FISH1 0.7546** 0.3583 0.2936** FISH2 0.3966 0.3689 0.1570 TURK 0.2715 0.3294 0.1076 LAMB 0.3545 0.299 0.1406 LogLikelihood = -72.41 R 2 (Psuedo) =15% Chi-squared = 64.44 % of corrected predictions = 76.2% *,**,*** indicate significance at .01, .05, and .10 levels, respectively

PAGE 57

47 Additionally, the second probit analysis was conducted to examine if the levels of trying goat meat vary amongst the consumers of various Hispanics origins and as the generations the respondent had spent in the United States (and expected acculturation level) increases. Consistent with the whole survey sample results, there is no significant difference between the willingness to try amongst genders and educational attainment levels. When willingness to try levels for individuals of Puerto Rican, Mexican, and other descents were compared to those of Cuban descent a statistical significant difference was found when the individual are of Puerto Rican descent. The results revealed that if the respondent was of Puerto Rican rather than Cuban descent, the respondents willingness to try levels decreased 25.9 percent. According to probit results there was not a significant difference in the willingness to try goat meat between first generation Hispanics and Hispanics whose families have been in the United States for more than one generation. This may suggest that as acculturation increases, consumption patterns for goat meat remain unchanged, a positive indicator for the goat meat industry. When only considering Hispanic respondents, the results of the demographic variable did change. Household size remained significant, but now only the household size of three was statistically different from the size of five, (32.7 percent less likely to be willing to try), unlike the larger model, where all sizes were significantly different. Again only one age variable was significant, but this time it was the youngest age group (18 and 24). Respondents in this group were 51.4 percent more likely to indicate a willingness to try goat. Psychographic factors were again were slightly different in the Hispanic only model. Safety and price specials were the only significant variables. Consumers that

PAGE 58

48 rated safety as important were 36 percent less likely than individuals that feel safety is unimportant to try goat meat. Similar to the entire sample, as the significance of price specials changed from important to unimportant, the chances of trying goat meat increased 44.7 percent. Respondents perception of goat meat again had a statistically significant impact on the willingness of try goat meat. As the respondents perception changed from negative to positive, the likelihood of willingness try goat meat increased 54.6 percent. Unlike the general population model, there is no significant difference between the levels of trying goat meat as the consumers opinion towards the products change from negative to neutral. Finally, the analysis examined the relationship between trying goat meat and the frequency of consumption of other meat substitutes. Chicken and fish and seafood consumption were found to significantly affect the willingness to try goat meat. Similar to the entire survey sample model, respondents that consume chicken at least once a week were less likely to exemplify a willingness to consume goat meat. Respondents that ate chicken once a week were 48.2 percent less likely to be willing to try goat meat than those that consume chicken on special occasions or less frequently. When the frequency of consuming chicken changed from special occasions to more than once a week the chances of being willing to try goat meat decreased 51.3 percent. Fish and other seafood consumption positively effected goat meat only when they were consumed more than once a week. The willingness to try goat meat increased 29.4 percent as the frequency of fish and seafood consumption varied from more than once a week to on special occasion and less frequently.

PAGE 59

CHAPTER 7 CONCLUSIONS AND IMPLICATIONS Summary This thesis focuses on developing an understanding of factors that influence willingness to consume goat meat in Florida. Researchers believe Florida has the potential for a meat goat market that is profitable and the demand for goat meat products in Florida is expected to continue growing as target populations within the niche market increase. However, the lack of consumer information has hindered producers, processors, and marketers ability to fully delineate the profitability and viability of Floridas meat goat industry. Consumer information is a critical component to totally understand the possible economic impact of goat meat production and marketing in Florida. Thus, the primary objective of this research is to identify factors that influence and barriers that reduce consumption of goat meat. A probit analysis of willingness to try goat meat indicted that factors influencing willingness try goat meat differed between the whole survey model and the Hispanics model. Hispanic consumers were unaware of the safety standards, which may be a result of the lack of grades and standards in the meat goat industry. The general population sample participants were uninformed of goat meats nutritional attributes (i.e. low levels of fat and cholesterol). Results from a partial nutrient analysis (Johnson, 1995), suggested that goat meat was comparable to chicken in total grams of fat, percent calories from fat and cholesterol. Also, the nutrient profile indicated that goat meat was similar beef in iron content. According to probit model results, safety and fat content had a 49

PAGE 60

50 negative influence on the willingness to try. Indicating that overall, consumer awareness of goat meat attributes in low, and more information should be made available to consumers. Both groups of shoppers indicated that price specials were important when purchasing meat to the grocery store. Price specials increased that likelihood of trying goat meat; however, specialty meat price are usually more expensive than traditional meat prices. Therefore, marketers may want implement various pricing strategies, pricing the meat lower than other meats to increase sales. This study did not identify reasons for not consuming goat meat. Further research should be conducted to identify these factors so that the industry can address and attempt to rectify these issues that restrict consumption. Perception of goat meat influenced respondents willingness to try goat meat. As the respondents view of goat meat became more positive, the likelihood of trying goat meat increased at least 30 percent...Unlike the results the entire survey model, the was not a significant relationship between the willing to try goat meat and the perception as the respondent view changed from negative to neutral. Results show that the frequency of chicken and fish consumption effect the willingness to try goat meat in both models. In addition to those meats, respondents that consumed lamb were also more likely to be willing try goat meat in the model including the entire sample. Prices of substitutes were not included in the model, as price for goat meat were not available either. However the relationship between the willingness to try goat meat and chicken, fish, and lamb indicated that goat meat has substitutes and complements. Therefore, members of the goat meat industry could develop strategies that differentiate goat meat from its competitors in an effort to increase market share. The

PAGE 61

51 likelihood of trying goat meat increased when participants consumed fish on a weekly basis and if consumers had previously consumed lamb; thus, meat marketing strategies should be aimed at these individuals. Income was not included in the model due to a high refusal rate in answering the question; therefore income elasticities were not calculated. However, education was used as a proxy for income and was found to be insignificant. The most unanticipated result from this study was that ethnicity or race was insignificant, which implies that the willingness to try goat meat is the same for Hispanics, Caucasians, and other races is statistically the same. This notion means that marketers should develop marketing strategies that target all consumers, and not focus on one particular group per se Hispanics. Furthermore, our results indicate that there is opportunity for growth in the goat meat industry. Gender was not significant, however, females are usually the primary grocery shoppers; goat meat should be promoted in a manner than accentuates the characteristics that have been identified as important to female shoppers. Other demographic factors that influenced the willingness to try goat meat were household size and age. Consistent with previous research, larger households, those containing five or more individuals were most likely to try goat for the entire sample. Larger households may be more willing to try goat meat because consumers perceive it to be inexpensive; thus, making goat meat an affordable meat alternative when feeding a large family. The consumers in the youngest age group, 18-24, were most likely to express a willingness to try goat meat among Hispanic consumers, where as there was consumes between the age of 55-64 years were most likely for the entire sample. This is

PAGE 62

52 an important finding for the goat meat industry, because if an individual develops a fondness for goat meat at an earlier age, more than likely these individuals will become life long consumers of the product, which may result an overall in goat meat consumption. Conclusions This research suggests opportunities for expanding the goat meat industry exists in Florida, with 43.6 percent of the sample indicating a willingness to try goat meat if it was available in supermarkets. Demographic characteristics such ethnicity, gender, nor educational attainment levels did not affect the dependent variable, but other demographic characteristics did affect the willingness to try. The results revealed that as the descent of the respondent changed from Cuban to Puerto Rican, the willingness to try decreased 25.9 percent. Psychographic factors that effect the willing to try included price specials, fat content, safety, and the consumers perception of the product. Consumers indicated fat content and safety were important to their purchase decisions were less likely to be willing to goat meat, while perception of goat meat and price specials has a positive relationship with the willingness to try. As goat meat is low in fat and often more expensive than other meats, these results seem counter-intuitive. However, this may all be an indicator that knowledge about the attributes of goat meat are low and in industry may benefit from educational efforts. Finally, the analysis describes relationship between the willing to trying goat meat and the frequency of consumption of other meats. Chicken, fish, and lamb consumption were found to significantly effect the willingness to try goat meat. Respondents that had previously consumed lamb were more willing to try goat meat. If the participants consumed fish and seafood a minimum of once a week, the probability of trying goat

PAGE 63

53 meat increases. Consumers that ate chicken more than once a month were less likely to try goat meat. Pork and beef consumption were found to have no significant effect on trying goat meat. One can conclude that goat meat consumption will occur less often as the frequency of chicken consumed increases. On the other hand, fish and lamb consumption may serve as an indicator of potential goat consumption. Implications In recent years the demand for goat meat has increased in the United States and it appears that opportunities for expansion exist for Floridas goat meat industry. The findings from this study can be used by the industry to develop marketing strategies that will provide assistance in increasing the demand for the product. The results also suggest potential consumers that should be targeted by the industry. This study identified new consumers for goat meat. In order for the goat meat industry to expand, market development strategies, which involve targeting new consumers with a present product, can be used to enhance the demand for goat meat. It is easier to target consumers that have an interest in trying goat meat than to market the product to those that are unwilling to try it. If the goat meat industry wishes to increase the demand for goat meat it is necessary to expand its target market beyond ethnic populations and promote the products or all individuals. This study suggests that this opportunity exists because there was no difference in willingness to try among the ethnic/ racial groups suggesting that the industry should tap into the Caucasian market. Research suggests that the industry should direct its marketing efforts to individuals living in larger households and those between the ages of 18-24. It is imperative for the industry to attract long term consumers in order to have long run success and by targeting younger consumers this is possible. If younger consumers develop a preference for goat meat

PAGE 64

54 products at an earlier age, they are more than likely to consumer the product as they grow older. Additionally, since ethnicity/race was an insignificant factor in this study, the industry should also target Caucasian females. Other potential consumers are those that live in larger households and lamb consumers. It seems a major barrier that hinders the prosperity of the goat meat industry is that consumers lack knowledge on goat meat. Consumers are becoming more health conscious and they are consuming products that possess nutritional qualities such as chicken and fish more frequently. Goat meat is very healthy, low in fat and cholesterol and high in proteins, however consumers are unaware of these qualities; therefore, educational information that increases consumers awareness of goat meat. If this information was known by consumers the industry may be able to repositioning the product in consumers mind. It seems that consumers perceive the product as cheap and containing high levels of fat, but the in untrue. In store demonstrates, educational advertising, recipes are promotional strategies that would increase consumers awareness of goat meat, resulting in increased consumption. As a result of this study it is evident that further research that involves knowledge testing is needed. This type of research would reveal the familiarity levels consumers have for goat meat. These results would inform industry official on the subject matter that needs to be discussed in the educational advertisements. Additionally, since ethnicity/race was non-significant and more balance sample, on the does not place an emphasis on Hispanics, should be conducted to investigate the reasons for consuming and not consuming goat meat is needed. The Floridas goat meat industry has the opportunity

PAGE 65

55 to flourish; however, effective marketing strategies are needed to increase consumers awareness and the availability of goat meat.

PAGE 66

APPENDIX A SOUTHERN REGION GOAT CONSUMPTION SURVEY March 15, 2004 Hello, this is (NAME) calling from the University of Georgia in Athens. The Survey Research Center is conducting a study this evening in conjunction with Mack C. Nelson, a professor from Fort Valley State University (GA) concerning the use of goat meat and wed like to talk to the primary food shopper of your household. Do you have a few minutes right now to complete an interview? 1. Yes (CONTINUE) 2. No (GET SRS NAME, ARRANGE CALLBACK; APPLY PERSUADERS. EVEN IF RESPONDENT DOESNT EAT GOAT MEAT, WE WANT THEM TO COMPLETE THE SURVEYIT WILL BE SHORTER. ) [INTERVIEWER NOTE: IF PRIMARY FOOD SHOPPER DID NOT HEAR INITIAL INTRODUCTION BUT DOES COME TO THE PHONE, REPEAT INTRO] S1 Are you 18 years or older? 1. Yes (CONTINUE) 2. No (ASK TO SPEAK TO ADULT 18 YEARS OR OLDER. RETURN TO INTRO. IF NECESSARY, GET SRS NAME AND SET CALLBACK.) Great! Before we begin, I need to let you know that the interview is completely voluntary. All of the information you provide will be kept strictly confidential and you dont have to answer any questions you dont want to. Also, my supervisor may listen to part of the interview to be sure that Im not making any mistakes. Q1. Have you or any member of your immediate family ever eaten goat meat? 1. Yes [SKIP TO Q4] 2. No 3. Dont know 9. Ref/NA Q2. DELETED 56

PAGE 67

57 Q2.1 If goat meat was available in your area food stores, do you think you would try it? 1. Yes 2. No 9. Ref/DK/NA Q2.2 If there were a cooperative that sold the meat of animals grown organically, would you be willing to join? 1. Willing 2. Not willing 9. Ref/DK/NA [ALL ANSWERS SKIP TO Q14] Q3. DELETED Q4. Whats your preference in goat meat? Would it be the kid, small male, small female, wether or something else? 1. Kid 2. Small male 3. Small female 4. Wether (castrated male) 5. Other [Specify] _________________ 9. Ref/DK/NA Q5. What live weight do you prefer? 1. Less than 30 pounds 2. 30 50 pounds 3. 51 69 pounds 4. 70 pounds or more 9. Ref/DK/NA Q6. Do you prefer a certain cut of meat? 1. Yes 2. No [SKIP TO Q7] 9. Ref/DK/NA [SKIP TO Q7]

PAGE 68

58 Q6.1 How much do you prefer the shoulder? Do you prefer the shoulder very much, somewhat, not much or not at all? 1. Very much 2. Somewhat 3. Neutral (Doesnt matter) 4. Not much 5. Not at all 9. Ref/DK/NA Q6.2 How much do you prefer the ribs? Would it be very much, somewhat, not much or not at all? 1. Very much 2. Somewhat 3. Neutral (Doesnt matter) 4. Not much 5. Not at all 9. Ref/DK/NA Q6.3 How much do you prefer the hind leg? 1. Very much 2. Somewhat 3. Neutral (Doesnt matter) 4. Not much 5. Not at all 9. Ref/DK/NA Q6.4 How much do you prefer loin chops? 1. Very much 2. Somewhat 3. Neutral (Doesnt matter) 4. Not much 5. Not at all 9. Ref/DK/NA

PAGE 69

59 Q6.5 How much do you prefer loin cubes? 1. Very much 2. Somewhat 3. Neutral (Doesnt matter) 4. Not much 5. Not at all 9. Ref/DK/NA Q6.6. Are there any other cuts of meat that you prefer to eat? 1. Name cuts of meat ________________________________ 9. No, Ref/DK/NA Q7. Do you normally buy a whole goat? 1. Yes 2. No 3. Dont know 9. Ref/NA Q8. Are there certain seasons of the year that you eat more goat meat? 1. Yes 2. No [SKIP TO Q9] 9. Ref/DK/NA [SKIP TO Q9] Q8.1 What seasons are those? [INTERVIEWER NOTE: CHOOSE ALL THAT APPLY] [PROGRAMMER NOTE: YES/NO TOGGLE] 1. Winter 2. Spring 3. Summer 4. Fall 5. Ref/DK/NA 6. Exit

PAGE 70

60 Q9 Do you eat goat meat on special occasions? 1. Yes 2. No [SKIP TO Q10] 9. Ref/DK/NA [SKIP TO Q10] Q9.1 Which special occasions are those? [INTERVIEWER NOTE: CHOOSE ALL THAT APPLY] [PROGRAMMER NOTE: YES/NO TOGGLE] 1. Christmas 2. 4 th of July 3. Family re-unions 4. Marriages 5. Ramadan 6. Cinco de Mayo 7. Other [Specify] _________________ 8. Ref/DK/NA 9. Exit Q10. About how many pounds of goat meat do you think your family eats each year? [INTERVIEWER NOTE: USE PERSUADERS IF NECESSARY: I JUST NEED A BALLPARK FIGURE.] pounds 998 998 or more 999 Ref/DK/NA [RANGE: 1 999] Q11. Would your family eat more goat meat if it was available in your local grocery stores? 1. Yes 2. No [SKIP TO Q12] 3. Dont know [SKIP TO Q12] 9. Ref/NA [SKIP TO Q12]

PAGE 71

61 Q11.1 Which meat product would you eat less of if you increased the family consumption of goat meat? [INTERVIEWER NOTE: IF NECESSARY, READ: BEEF, PORK, SEAFOOD, LAMB, CHICKEN OR TURKEY?] 1. Enter response 9. Ref/DK/NA Q12. Please tell me how important the following attributes are in your decision to purchase goat meat products. Would you say fresh, never frozen product is very important, important, not very important or not at all important? 1. Very important 2. Important 3. Neutral (Doesnt matter) 4. Not very important 5. Not at all important 9. Ref/DK/NA Q12.1 How important is the color of the meatvery important, important, not very important or not at all important? 1. Very important 2. Important 3. Neutral (Doesnt matter) 4. Not very important 5. Not at all important 9. Ref/DK/NA Q12.2 How important is the government inspection label? 1. Very important 2. Important 3. Neutral (Doesnt matter) 4. Not very important 5. Not at all important 9. Ref/DK/NA

PAGE 72

62 Q12.3 How important is it that the goat meat is organically grown? 1. Very important 2. Important 3. Neutral (Doesnt matter) 4. Not very important 5. Not at all important 9. Ref/DK/NA Q12.4 How important is it that there are a variety of cuts available? 1. Very important 2. Important 3. Neutral (Doesnt matter) 4. Not very important 5. Not at all important 9. Ref/DK/NA Q12.5 How important is it that there are prepackaged cuts? 1. Very important 2. Important 3. Neutral (Doesnt matter) 4. Not very important 5. Not at all important 9. Ref/DK/NA Q12.6 How important are cooking instructions? 1. Very important 2. Important 3. Neutral (Doesnt matter) 4. Not very important 5. Not at all important 9. Ref/DK/NA

PAGE 73

63 Q12.7 How important are marinade cuts? 1. Very important 2. Important 3. Neutral (Doesnt matter) 4. Not very important 5. Not at all important 9. Ref/DK/NA Q12.8 How important are convenience foods, such as sausage? 1. Very important 2. Important 3. Neutral (Doesnt matter) 4. Not very important 5. Not at all important 9. Ref/DK/NA Q12.9 How important is the price in your decision to purchase goat meat products? 1. Very important 2. Important 3. Neutral (Doesnt matter) 4. Not very important 5. Not at all important 9. Ref/DK/NA Q13. When you cook your goat meat, how often do you make soup? Would you say frequently, some of the time, not very often or never? 1. Frequently 2. Some of the time 3. Not very often 4. Never 9. Ref/DK/NA

PAGE 74

64 Q13.1 When you cook your goat meat, how often do you make meat sauce? Would you say frequently, some of the time, not very often or never? 1. Frequently 2. Some of the time 3. Not very often 4. Never 9. Ref/DK/NA Q13.2 How often do you make chili when you cook goat meat? 1. Frequently 2. Some of the time 3. Not very often 4. Never 9. Ref/DK/NA Q13.3 How often do you make meat loaf when you cook goat meat? 1. Frequently 2. Some of the time 3. Not very often 4. Never 9. Ref/DK/NA Q13.4 When you cook your goat meat, how often do you broil it? Would you say . 1. Frequently 2. Some of the time 3. Not very often 4. Never 9. Ref/DK/NA Q13.5 When you cook goat meat, how often do you oven roast it? 1. Frequently 2. Some of the time 3. Not very often 4. Never 9. Ref/DK/NA

PAGE 75

65 Q13.6 When you cook goat meat, how often do you have sausage made? 1. Frequently 2. Some of the time 3. Not very often 4. Never 9. Ref/DK/NA Q13.7 When you cook your goat meat, how often do you barbeque the meat? 1. Frequently 2. Some of the time 3. Not very often 4. Never 9. Ref/DK/NA Q13.8 Are there any other ways that you cook your goat meat? 1. Enter response ________________________ 9. Ref/DK/NA Q14. This section is about which meats you eat most often. Would you say that you eat beef every day, more than once a week, once a week, more than once a month, once a month, on special occasions or never? 1. Every day 2. More than once per week 3. Once per week 4. More than once per month 5. Once per month 6. Special occasions 7. Never [SKIP TO Q15] 9. Ref/DK/NA [SKIP TO Q15]

PAGE 76

66 Q14.1 When you eat beef how much do you usually eat? Would you say less than pound, pound, pound, 1 pound or more than one pound? 1. Less than one-fourth pound 2. One-fourth pound 3. One-half pound 4. One pound 5. More than one pound 9. Ref/DK/NA Q15 And how often would you say that you eat chicken? Would it be . 1. Every day 2. More than once per week 3. Once per week 4. More than once per month 5. Once per month 6. Special occasions 7. Never [SKIP TO Q16] 9. Ref/DK/NA [SKIP TO Q16] Q15.1 When you eat chicken how much do you usually eat? Would you say less than pound, pound, pound, 1 pound or more than one pound? 1. Less than one-fourth pound 2. One-fourth pound 3. One-half pound 4. One pound 5. More than one pound 9. Ref/DK/NA Q16 What about turkey, how often do you eat turkey? 1. Every day 2. More than once per week 3. Once per week 4. More than once per month 5. Once per month 6. Special occasions 7. Never [SKIP TO Q17] 9. Ref/DK/NA [SKIP TO Q17]

PAGE 77

67 Q16.1 When you eat turkey how much do you usually eat? Would you say less than pound, pound, pound, 1 pound or more than one pound? 1. Less than one-fourth pound 2. One-fourth pound 3. One-half pound 4. One pound 5. More than one pound 9.Ref/DK/NA Q17 And how often do you eat lamb? 1. Every day 2. More than once per week 3. Once per week 4. More than once per month 5. Once per month 6. Special occasions 7. Never [SKIP TO Q18] 9. Ref/DK/NA [SKIP TO Q18] Q17.1 When you eat lamb how much do you usually eat? Would you say less than pound, pound, pound, 1 pound or more than one pound? 1. Less than one-fourth pound 2. One-fourth pound 3. One-half pound 4. One pound 5. More than one pound 9. Ref/DK/NA Q18 How often do you eat goat meat or chevon? 1. Every day 2. More than once per week 3. Once per week 4. More than once per month 5. Once per month 6. Special occasions 7. Never [SKIP TO Q19] 9. Ref/DK/NA [SKIP TO Q19]

PAGE 78

68 Q18.1 When you eat goat/chevon how much do you usually eat? Would you say less than pound, pound, pound, 1 pound or more than one pound? 1. Less than one-fourth pound 2. One-fourth pound 3. One-half pound 4. One pound 5. More than one pound 9. Ref/DK/NA Q19 How often do you eat fish or seafood? 1. Every day 2. More than once per week 3. Once per week 4. More than once per month 5. Once per month 6. Special occasions 7. Never [SKIP TO Q20] 9. Ref/DK/NA [SKIP TO Q20] Q19.1 When you eat fish or seafood, how much do you usually eat? Would you say less than pound, pound, pound, 1 pound or more than one pound? 1. Less than one-fourth pound 2. One-fourth pound 3. One-half pound 4. One pound 5. More than one pound 9. Ref/DK/NA Q20 And finally, how often do you eat pork? 1. Every day 2. More than once per week 3. Once per week 4. More than once per month 5. Once per month 6. Special occasions 7. Never [SKIP TO Q23] 9. Ref/DK/NA [SKIP TO Q23]

PAGE 79

69 Q20.1 And when you eat pork how much do you usually eat? Would you say less than pound, pound, pound, 1 pound or more than one pound? 1. Less than one-fourth pound 2. One-fourth pound 3. One-half pound 4. One pound 5. More than one pound 9. Ref/DK/NA Q21 Question deleted (duplicate of Q11) Q22 DELETED Q23. In your household, has your spouse consumed goat meat products? 1. Yes 2. No 3. Do not have spouse 8. Ref/DK/NA [PROGRAMMER NOTE: IF Q23 = 3, SKIP Q36] Q23.1 Have your children consumed goat meat products? 1. Yes 2. No 3. Do not have children [SKIP TO Q26.1] 9. Ref/DK/NA [SKIP TO Q26.1] [PROGRAMMER NOTE: IF Q23.1 = 3, SKIP Q37] Q24. Now Id like to ask you the gender of your children. Remember, this is confidential and the answers will not be connected with your phone number in any way. Q24.1 DELETED Q24.2 Gender of first child under 18? 1. Male 2. Female 9. Ref/DK/NA [SKIP TO Q25]

PAGE 80

70 Q24.3 DELETED Q24.4 Gender of second child under 18? 1. Male 2. Female 3. No more children [SKIP TO Q25] 9. Ref/DK/NA [SKIP TO Q25] Q24.5 DELETED Q24.6 Gender of third child under 18? 1. Male 2. Female 3. No more children 9. Ref/DK/NA Q25. If you and/or your spouse eat goat meat but your children dont, what are the reasons your children dont consume the product? [INTERVIEWER NOTE: CHOOSE ALL THAT APPLY] [PROGRAMMER NOTE: YES/NO TOGGLE] 1. They dont like it 2. They were not reared where it was consumed regularly 3. It wasnt available 4. Their friends dont eat it, so they dont 5. Others (please list) _______________________________________________ 6. Ref/DK/NA 7. Exit Q26. DELETED Q26.1 To your knowledge, have any of the following family members eaten goat meat products? [INTERVIEWER NOTE: READ RESPONSES. CHOOSE ALL THAT APPLY]

PAGE 81

71 [PROGRAMMER NOTE: YES/NO TOGGLE] 1. Mother 2. Father 3. In-Laws 4. Others [Please specify _____________________] 5. Ref/DK/NA 6. Exit Q27 I would like to know how you see goat meat products. Would you say your view is: 1. Very positive 2. Positive 3. Somewhat positive 4. Neutral 5. Somewhat negative 6. Negative 7. Very negative 9. Ref/DK/NA For the next few items, Id like you to rate the importance of several factors in your decision to buy or not to buy a goat meat product. Even if you do not currently consume goat meat, please tell me how important the following items are in your decision to buy goat meat and other kinds of meat. Q28. How important are food page advertisements in your decision to buy goat meat? 1. Very important 2. Important 3. Neutral 4. Not very important 5. Not at all important 10. Ref/DK/NA Q28.1 How important are food page advertisements in your decision to buy other meats? 1. Very important 2. Important 3. Neutral 4. Not very important 5. Not at all important 9. Ref/DK/NA

PAGE 82

72 Q29. How important are store displays in your decision to buy goat meat? 1. Very important 2. Important 3. Neutral 4. Not very important 5. Not at all important 9. Ref/DK/NA Q29.1 How important are store displays in your decision to buy other meats? 1. Very important 2. Important 3. Neutral 4. Not very important 5. Not at all important 9. Ref/DK/NA Q30. How important are price specials in your decision to buy goat meat? 1. Very important 2. Important 3. Neutral 4. Unimportant 5. Not important at all 9. Ref/DK/NA Q30.1 How important are price specials in your decision to buy other meats? 1. Very important 2. Important 3. Neutral 4. Unimportant 5. Not important at all 9. Ref/DK/NA Q31. How important are in-supermarket taste tests in your decision to buy goat meat? 1. Very important 2. Important 3. Neutral 4. Unimportant 5. Not important at all 9. Ref/DK/NA

PAGE 83

73 Q31.1 How important are in-supermarket taste tests in your decision to buy other meats? 1. Very important 2. Important 3. Neutral 4. Unimportant 5. Not important at all 9. Ref/DK/NA Q32. How important are safety assurances such as USDA inspections of goat meat? 1. Very important 2. Important 3. Neutral 4. Unimportant 5. Not important at all 9. Ref/DK/NA Q32.1 How important are safety assurances such as USDA inspections of other meats? 1. Very important 2. Important 3. Neutral 4. Unimportant 5. Not important at all 9. Ref/DK/NA Q33. How important are convenience products such as sausage or ground meat from goat meat? 1. Very important 2. Important 3. Neutral 4. Somewhat unimportant 5. Not important at all 9. Ref/DK/NA Q33.1 How important are convenience products such as sausage or ground meat from other meats? 1. Very important 2. Important 3. Neutral 4. Somewhat unimportant 5. Not important at all 9. Ref/DK/NA

PAGE 84

74 Q34. How important is the fat content in goat meat? 1. Very important 2. Important 3. Neutral 4. Somewhat unimportant 5. Not important at all 9. Ref/DK/NA Q34.1 How important is the fat content in other meats? 1. Very important 2. Important 3. Neutral 4. Somewhat unimportant 5. Not important at all 9. Ref/DK/NA Q35. How important is the cholesterol content of goat meat? 1. Very important 2. Important 3. Neutral 4. Somewhat unimportant 5. Not important at all 9. Ref/DK/NA Q35.1 How important is the cholesterol content of other meats? 1. Very important 2. Important 3. Neutral 4. Somewhat unimportant 5. Not important at all 9. Ref/DK/NA

PAGE 85

75 For each of the following that I read to you, please rate how important each would be in your decision to purchase goat meat products, just as you make decisions to buy beef, poultry, or pork. Please use a scale of 1 to 5 where is very important and is not important at all. Q36. First, how important is your spouses opinion on your decision to purchase or not purchase goat meat products? 1. Very important 2. Important 3. Neutral 4. Somewhat unimportant 5. Not important at all 9. Ref/DK/NA [PROGRAMMER NOTE: IF Q23 = 3, THIS QUESTION IS SKIPPED (Q36)] Q37. How important are your childrens opinions on your decision to purchase or not purchase goat meat products? 1. Very important 2. Important 3. Neutral 4. Somewhat unimportant 5. Not important at all 9. Ref/DK/NA [PROGRAMMER NOTE: IF Q23.1 = 3, THIS QUESTION IS SKIPPED (Q37) Q38. How important is your own opinion on your decision to purchase or not purchase goat meat products? 1. Very important 2. Important 3. Neutral 4. Somewhat unimportant 5. Not important at all 9. Ref/DK/NA [PROGRAMMER NOTE: IF Q1 > 1, SKIP TO Q47] Q39. DELETED Q40. DELETED

PAGE 86

76 Q41. How often does your family eat goat meat? 1. Weekly 5. Quarterly 2. Bi-Weekly 6. Special occasions 3. Monthly 7. Refused 4. Less than quarterly 8. Dont know Q42. Do you get your goat meat from . .? [INTERVIEWER NOTE: READ LIST, CHOOSE ALL THAT APPLY] [PROGRAMMER NOTE: YES/NO TOGGLE 1. Supermarket 2. Farmer 3. A friend not a farmer 4. Farmers Market 5. Grow our own 6. Restaurant 7. Other place (SPECIFY) ____________ 8. Ref/DK/NA 9. Exit Q43. Would you be willing or not willing to join a consumer-farmer cooperative if prices for goat meat were lower? 1. Willing 2. Not willing 9. Ref/DK/NA Q44. Would you be willing or not willing to join a co-op if it had a more dependable source of goat meat? 1. Willing 2. Not willing 9. Ref/DK/NA Q45. Would you be willing or not willing to join a cooperative that grew its animals organically? 1. Willing 2. Not willing

PAGE 87

77 9. Ref/DK/NA 46. Would you be willing or not willing to join a cooperative that sold its members fresh, never frozen goat meat? 1. Willing 2. Not willing 9. Ref/DK/NA Q47. How many people live in your home? ____________ number of people 99 Ref/DK/NA [RANGE: 1 99] Q48 How many of the people are less than 18 years old? _______________ less than 18 years old 99 Ref/DK/NA [RANGE: 0 99] Now I just need to ask a few questions about you personally so that we can compare your answers with different types of people. Q49. What do you consider your race to be? 1. White [SKIP TO Q54] 2. African-American/Black [SKIP TO Q54] 3. Black Non African American [SKIP TO Q54] 4. Hispanic 5. Asian [SKIP TO Q54] 6. Multi-Racial (SPECIFY)( ) [SKIP TO Q54] 9. Ref/DK/NA [SKIP TO Q54]

PAGE 88

78 Q50. Are you of . .? [INTERVIEWER NOTE: READ RESPONSES. CHOOSE ONLY ONE] 1. Mexican descent 2. Cuban descent 3. Puerto Rican descent 4. Spaniard 5. Or Other (Please list) __________________ 7. Ref/DK/NA Q51. Are you the 1 st generation, 2 nd generation or another generation to live in the contiguous U. S. States? 1. 1 st generation 2. 2 nd generation 3. Other 9. Ref/DK/NA Q52. DELETED Q53. Which of your family members are Hispanic? [INTERVIEWER NOTE: CHOOSE ALL THAT APPLY] [PROGRAMMER NOTE: YES/NO TOGGLE] 1. Mother 2. Father 3. Spouse 4. Aunts or uncles 5. Cousins 6. Other 7. Ref/DK/NA 8. Exit

PAGE 89

79 Q54. What is your age range? 1. Less than 20 years 7. 60 64 years 2. 20 24 years 8. 65 74 years 3. 25 34 years 9. 75 84 years 4. 35 44 years 10. 85 years plus 5. 45 54 years 11. Refused 6. 55 59 years Q55. What is the highest grade of school or year of college you completed? 1.
PAGE 90

80 4. GA 5. MS 6. NC 7. OK 8. LA 9. SC 10. TN 11. TX 99 Ref/DK/NA This completes the survey and I want to thank you for taking the time to answer these questions. Have a nice evening. Good bye. IMPORT FIPS IMPORT MSA/NON-MSA QUOTA: 250 FOR EACH STATE

PAGE 91

APPENDIX B SUMMARY OF DEMOGRAPHIC INFORMATION 010203040506070Percent Had ConsumedWilling to TryHad not Consumedand Unwilling Try Hispanics Cacausians Others Note: Chi-squared probability < .10. Figure B-1. Respondents consumption preference for goat meat by ethnicity/race. 010203040506070Percent Had ConsumedWilling to TryHad not Consumedand Unwilling Try Male Female Note: Chi-squared probability < .05. Figure B-2. Respondents consumption preference for goat meat by gender. 81

PAGE 92

0102030405060 HadConsumedWilling to TryHad notConsumed andUnwilling Try HS Diploma or less Some College College Degree and above Figure B-3. Respondents consumption preference for goat meat by educational attainment levels. 010203040506070Percent Had ConsumedWilling to TryHad notConsumed andUnwilling Try Mexican Puerto Rican Cuban Other Figure B-4. Hispanic respondents consumption preference for goat meat by origin. 82

PAGE 93

83 0102030405060Percent Had ConsumedWilling to TryHad notConsumed andUnwilling Try Male Female Figure B-5. Hispanic respondents consumption preference for goat meat by gender. 0102030405060Percent Had ConsumedWilling to TryHad notConsumed andUnwilling Try HS Diploma or less Some College College Degree and above Figure B-6. Hispanic respondents consumption preference for goat meat by educational attainment levels.

PAGE 94

84 0102030405060Percent HadConsumedWilling to TryHad notConsumed andUnwilling Try First Generation Second Generation or later Figure B-7. Hispanic respondents consumption preference for goat meat by generation in United States.

PAGE 95

LIST OF REFERENCES Aldrich, John H., and Forrest D. Nelson. Linear Probability, Logit, and Probit Models. Edited by Michael S. Lewis-Beck. Vol. 45, Quantitative Applications in the Social Sciences. Newbury Park, CA: Sage Publications, Inc., 1984. Associated Press. 2005." Polling the 'Cell Phone Only' Crowd: Pollisters, Researchers Tackle Hard-to-Track Population." In, Cable News Network, http://www.cnn.com/2005/TECH/02/25/polling.cell.users.ap/index.html (accessed February 25, 2005). Chen, Kevin, Murad Ali, Michele Veeman, Jim Unterschultz, and Theresa Le. "Relative Importance Rankings for Pork Attributes by Asian-Origin Consumers in California: Applying an Ordered Probit Model to a Choice-Based Sample." Journal of Agricultural and Applied Economics 34 (2002): 67-79. Davis, Ernie and Zane Willard. 1996. "Goat Meat the New Lean Meat." In, Texas A&M University, Texas A&M Agriculture News, http://agnews.tamu.edu/dailynews/stories/AGEC/Dec2096a.htm (accessed May 22, 2003). Degner, L. Robert. "Should You Market Chevon, Cabrito, or Goat Meat." Paper presented at the Regional Small Farm and Trade Show Conference, Tallahassee, FL, November 7, 1991. Degner, L. Robert and J. David Locascio. "Distribution of Goat Meat in Selected Metropolitan Florida Markets." Gainesville, FL: University of Florida, Food Resource Economics Department, Florida Agricultural Market Research Center, 1988. Degner, L. Robert, and C. T. Jordan Lin. "Marketing Goat Meat: An Evaluation of Consumer Perceptions and Preferences." University of Florida, Food Resource Economics Department, Florida Agricultural Market Research Center, 1993. Degner Robert, L., and Susan D. Moss. "The Florida Market for Goat Meat: 1999 Survey of Florida Meat Wholesalers." Gainesville, FL: University of Florida, Food Resource Economics Department, Florida Agricultural Market Research Center, 1999. Dillman, Don. Mail and Telephone Surveys: The Total Design Method. New York: Wiley-Interscience Publication, 1978. 85

PAGE 96

86 Frey, James, A. Survey Research by Telephone. 2nd ed. Vol. 150. Newbury Park, CA: Sage Publications, 1989. Gipson, Terry A. 1999. Demand for Goat Meat: Implications for the Future of the Industry. In Demand for Goat Meat: Implications for the Future of the Industry, Langston University, E(Kika)de la Garza Institute for Goat Research, http://www2.luresext.edu/goats/library/field/goat_meat_demand99.htm (accessed January 20, 2004). Greene, William H. Econometric Analysis. 5th ed. Upper Saddle River, NJ: Pearson Education, Inc., 2003. Gujarati, Damodor. Basic Econometrics. International Edition 2003 ed. New York, NY: McGraw Hill Irwin, 2003. Harwell, Lynn. "Goat Marketing Strategies Along the East Coast." In Florida's Goat Meat Industry, edited by James R. Simpson, 15-21: Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida, 1995. Hui, Jianguo, Patricia E. McLean, and Dewitt Jones. "An Empirical Investigation of Importance Rating of Meat Attributes by Louisiana and Texas Consumers." Journal of Agricultural and Applied Economics 27 (1995): 636-43. James, N. A. and B. W. Berry. "Use of Chevon in the Development of Low-Fat Meat Products." Journal of Animal Science 75 (1997): 571-77. Johnson, Dwain. "Composition and Quality of Goat Meat Produced." edited by James R. Simpson, 39-41: Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida, 1995. Latalia, Thomas. "A Pseudo R 2 Measure for Limited and Qualitative Dependent Variable Models." Journal of Econometrics 56 (1993): 341-56. Lavrakas, Paul J. Telephone Survey Methods: Sampling, Selection, and Supervision. Edited by Leonard Bickman. Vol. 7, Applied Social Research Method Series. Newbury Park, CA: Sage Publications, Inc., 1987. McKenzie-Jakes, Angela. 2004. "Markets for Goat Meat." In, Florida A&M University, http://smallfarms.ifas.ufl.edu/Liverstock/goats.htm (accessed July 18, 2003).

PAGE 97

87 McLean-Meyinsse, Patricia E. "Assessing Factors Affecting Consumers' Decisions to Shop at Stores Offering Specialty Meat." Journal of Food Distribution Research 78 (1999): 134-39. . "Assessing the Market Outlook for Rabbit Meat in Louisiana and Texas." Journal of Food Distribution Research 31 (2000): 139-44. . Factors Influencing Consumption or Willingness to Consume a Variety of Goat-Meat Products." Journal of Food Distribution Research 34 (2003): 72-79. Melton, Bryan, E., Wallace E. Huffman, Jason F. Shogren, and John A. Fox. "Consumer Preferences for Fresh Food Items with Multiple Quality Attributes: Evidence from an Experimental Auction of Pork Chops." American Journal of Agricultural Economics 78 (1996): 916-23. Miller, Pat. "Marketing Strategies for Goat Meat." In Florida's Meat Goat Industry, edited by James A. Simpson, 22-26: Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida, 1995. Nelson, Mack C., N.B. Brown, Jr., S. Mobini, S. Gelaye, and S. Leak. "Production and Marketing Challenges for Goat Producers: Implication for Supply and Demand." Paper presented at the Professional Ag Workers Conference (PAWC), December 1999. Ohio Cooperative Development Center. "Market Analysis of Meat Goat in Ohio". Ohio Cooperative Development Center. Available at http://ocdc.osu.edu/pdf/anal_proposal.pdf Accessed on. (accessed February 19, 2004). Peter, J. Paul, and James H. Donnelly, Jr. A Preface to Marketing Management. 9th ed. New York, NY: McGraw-Hill Irwin, 2003. Pinkerton, Frank, Lynn Harwell, William Drinkwater, Nelson Escobar. "Marketing Meat Goats: Channels, Supply, and Demand." Langston University, OK: Oklahoma Cooperative Extension Service, 1994. Pinkerton, Frank, David Scarfe, and Bruce Pinkerton. "Meat Goat Productions and Marketing." Langston University, E(Kika)de la Garza Institute for Goat Research, 1992. Putnam, Judith Jones, and Jane E. Allshouse. "Food Consumption, Prices, and Expenditures, 1970-97." edited by Economic Research Service (USDA-ERS) U.S. Department of Agriculture, 196, April 1999. Progressive Grocer. Progressive Grocer, 69th Annual Report of the Grocery Industry." April 2002.

PAGE 98

88 Rhee, K. S., C.E. Meyers, and D.F. Waldron. "Consumer Sensory Evaluation of Plain and Seasoned Goat Meat and Beef Products." Meat Science 65 (2003): 785-89. Rhee, K. S., M. Oltman, and J. Han. "A Consumer Survey of Goat Meat: Perception, Knowledge, and Use." Sheep and Goat Research Journal 16 (2000): 111-16. Russell, Jeremy. "Targeting Hispanic Consumers." National Provisioner 216 (2002): 44-53. Salant, Priscilla, and Don A. Dillman. How to Conduct Your Own Survey. New York: Wiley, 1994. Schupp, Alvin, Jeffery Gillespie, and Debra Reed. "Consumer Choice among Alternative Red Meats." Journal of Food Distribution Research 29 (1998): 35-43. Sherman, David. 2002. Current Tends in Goat Production in the United States. In, http://www.neahi.org.nejlah/articles/articles_goatmeat.html (accessed May 22, 2003). Simpson, James. "Introduction to Florida's Meat Goat Industry." In Florida's Goat Meat Industry, edited by James R. Simpson, 1-4: Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida, 1995. Smith, C. G., M. I. Pike, and Z. L. Carpenter. "Comparison of the Palatability of Goat Meat and Meat from Four Other Animal Species." Journal of Food Science 39 (1974): 1145-46. Spaugh, Beth C. 1997. "Goat Carcass Characteristics." In, Cornell Cooperative Extension, http://www.cce.cornell.edu/clinton/ag/goat-carcass.html (accessed August 20, 2004). Tomek, William G., and Kenneth L. Robinson. Agricultural Product Prices. 3rd ed. Ithaca, New York: Cornell University Press, 1990. U.S. Census Bureau. 2000. U.S. Census 2000. In, http://factfinder.census.gov/home/saff/main.html?_lang=en (accessed August 20, 2004). U.S. Department of Agriculture, National Agricultural Statistical Service (USDA-NASS). 2002. Census of Agriculture 2002. In, USDA-NASS, http://www.nass.usda.gov (accessed December 17, 2004). Verbeek, Marno. A Guide to Modern Econometrics. Southern Gate, Chichester, West Sussex, England: John Wiley & Sons, 2004.

PAGE 99

89 Verbeke, Wim, Ronald W. Ward, and Jacques Viaene. "Probit Analysis of Fresh Meat Consumption in Belgium: Exploring BSE and Television Communication Impact." Agribusiness 16 (2000): 215-34. Walsh, Robb. "Getting Your Goat." Natural History 104 (1995): 48-49. Zachery, Nikki, and Mack C. Nelson. "Consumers Knowledge and Use of Goat Products: An Atlanta Case Study." Paper presented at the Professional Ag Workers Conference (PAWC), Tuskegee University, December 1992.

PAGE 100

BIOGRAPHICAL SKETCH Erika Knight was in Warner Robins, GA. After graduating from Warner Robins High School, she attended Fort Valley State University, Fort Valley, GA, and earned a Bachelor of Science degree in agricultural economics. In August 2004, Erika began the Food and Resource Economics Master of Science program and specialized in marketing. 90