<%BANNER%>

Influential Factors on Brand Choice and Consumption Behaviors

Permanent Link: http://ufdc.ufl.edu/UFE0021644/00001

Material Information

Title: Influential Factors on Brand Choice and Consumption Behaviors An Exploratory Study of College Students and Beer
Physical Description: 1 online resource (130 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: beer, behavior, brand, choice, college, consumption, dave, ritter, student
Journalism and Communications -- Dissertations, Academic -- UF
Genre: Advertising thesis, M.Adv.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: This study examined factors influencing beer brand choice and beer consumption behavior among college students. It was determined after using six situational groupings that situational variation did not have a significant impact on brand choice of beer for college students. Relationships among brand benefits, exploratory shopping behaviors, susceptibility to interpersonal influence, product category involvement, and demographics were studied in order to determine what factors had the most significant impact on brand choice. It was determined that price and risk-taking behaviors had a significant impact on beer brand choice. Consumption behaviors among college students were influenced by emotions. In addition, an association between situation and consumption behavior existed among college students consuming their favorite beer. Bars, clubs, or parties demonstrated heavy to moderate drinking behavior. Situations where college students are relaxing such as at the pool/beach, at home with friends, and not a party demonstrated light beer drinking behaviors.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (M.Adv.)--University of Florida, 2008.
Local: Adviser: Kim, Hyojin.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021644:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021644/00001

Material Information

Title: Influential Factors on Brand Choice and Consumption Behaviors An Exploratory Study of College Students and Beer
Physical Description: 1 online resource (130 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: beer, behavior, brand, choice, college, consumption, dave, ritter, student
Journalism and Communications -- Dissertations, Academic -- UF
Genre: Advertising thesis, M.Adv.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: This study examined factors influencing beer brand choice and beer consumption behavior among college students. It was determined after using six situational groupings that situational variation did not have a significant impact on brand choice of beer for college students. Relationships among brand benefits, exploratory shopping behaviors, susceptibility to interpersonal influence, product category involvement, and demographics were studied in order to determine what factors had the most significant impact on brand choice. It was determined that price and risk-taking behaviors had a significant impact on beer brand choice. Consumption behaviors among college students were influenced by emotions. In addition, an association between situation and consumption behavior existed among college students consuming their favorite beer. Bars, clubs, or parties demonstrated heavy to moderate drinking behavior. Situations where college students are relaxing such as at the pool/beach, at home with friends, and not a party demonstrated light beer drinking behaviors.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (M.Adv.)--University of Florida, 2008.
Local: Adviser: Kim, Hyojin.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021644:00001


This item has the following downloads:


Full Text





INFLUENTIAL FACTORS ON BRAND CHOICE AND CONSUMPTION BEHAVIORS: AN
EXPLORATORY STUDY ON COLLEGE STUDENTS AND BEER
















By

DAVE RITTER


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

UNIVERSITY OF FLORIDA

2008




































2008 Dave Ritter





























To my family, friends, educators, and colleagues, cheers! Without your love, support, and
guidance none of this would be possible.









ACKNOWLEDGEMENTS

I would like to thank my entire family for supporting my education and research at the

University of Florida. I would like to thank Lakeland College, the University of Illinois, and the

University of Florida for providing me with the skills and education possible to complete this

document. I also would like to express my gratitude to all of my committee members for their

help, guidance, and comments throughout this process. In addition, I would like to give a special

thank-you to my thesis chair, Hyojin Kim. Finally, I would like to thank God for giving me faith,

ability, and opportunity to persevere.









TABLE OF CONTENTS

page

A CKN OW LED GEM EN TS .............. ........... ...........................................................4

LIST OF TABLES ............ ........... ......... ........7.....

LIST OF FIGURES .................................. .. ..... ..... ................. .8

A B S T R A C T ......... ....................... .................. .......................... ................ .. 9

CHAPTER

1 INTRODUCTION ............... .............................. ............................. 10

2 L IT E R A TU R E R E V IE W ......................................................................... ........................ 13

B ran d C h oice ................................................................................13
D desired Brand Benefits ................................................... .............. ......... 19
Perform ance/Q quality .................. ........................................ .............. 20
P rice/V alu e for M money .......................................................................... ....................22
E m options .................................................................................................23
N orm ative/P personal (Social) ........................................ ............................................23
Environm ent (Stew hardship) ............................................................. ............... 24
H e a lth ........................................................ ...................................2 5
Situational F actors ............................................................................27
C on su m er F actors ...........................................................................2 9
Exploratory Shopping B ehaviors .............................................................................30
Interpersonal Influence .................. ...................................... ............... 30
C on su m option B eh av iors .................................................................................................. 3 1
Product Category Involvem ent ............................................ ...... ....................... 32
D em og rap h ics............................ ............................................................ ............... 32
T theory of R seasoned A action ................................................................................. ...........33
R research Q uestions.......... ............................................................................ ......... .... 36

3 M E T H O D .............. .... ...............................................................3 7

P re te st .............................................................................................................. ....... 3 7
Subj ects ........................................... .. ..............................37
D design and Procedure ....................................................................................... ............... 38
M e a su re s .......... .. ................. .................................................................................................. 3 8
B rand C choice .............................................................................................................38
Situational V ariation.............................................. 39
D esired Brand B enefits................................................... 39
Consum ption Behavior.................................................. 39
D em graphics .............................................. 40
M ain Stu dy ......... .................................................................40










Subj ects ......... ........... ..... ..... ....................... 40
D design and Procedure ........... .................. .. ......... ........ .. ...... 42
M measures ......... ........ .... ................................ ............... 43
B rand C choice .............................................................................................................43
Situational V ariation.............................................. 43
D esired Brand B enefits................................................... 46
Exploratory Shopping Behaviors ................. ...............................46
Interpersonal Influence .................................................................... .... ......... .......... ........ 47
Consum ption Behavior................................................... 47
Product Category Involvement .............. ......... ................. 49
D em graphics .............................................. 50

4 RESULTS ............. .... ...... ... ................................... ............... 51

Brand Choice among Situational Variation ......... ...... ............ .......... ...... ............. 51
Consumption Behavior among Situation Variation ........................................... ....... 58
Variables Affecting Beer Brand Choice .................................................67
V ariables Affecting Beer Consum ption.............................................. ........... ............... 79

5 DISCUSSION ................... ...................................... .90

6 LIMITATIONS ................... .... ................................. 96

APPENDIX

A PR E TE ST .............................................................. .................. 99

B M A IN STUD Y ...... .......................................... .......................... 109

REFEREN CES ............................................ .............. .....................120

B IO G R A PH ICA L SK ETCH ......... ......... .....................................................................130





















6









LIST OF TABLES


Table page

2-1 B rand C choice Studies............ .................................................................... ...... .. .... 16

3-1 Beer Brand Choice............ ........ ...................... ......... 41

3-2 Demographic Frequency Distribution.................... ......... ........................ ............... 42

3-3 Situation Frequency D distribution ............................................................ .....................45

3-4 Average Beers Consumed and Average Occasions Drinking Beer..............................48

3-5 Consumption Behavior Frequency .............................................................................. 49

4-1 Situational Groups and Brand Choice........................................ ............................ 53

4-2 Situational Groups and Consumption Behavior ..................................... .................61

4-3 Multiple Discriminant Analysis: Beer Brand Choice....................................................73

4-4 Structure M atrix for Beer Brand Choice....................................... ......................... 75

4-5 Group Classification for Beer Brand Choice: Multiple Discriminant Analysis ...............75

4-6 Mean and Standard Deviation based on Brand Choice ............................................. 76

4-7 Stepwise Discriminant Analysis: Beer Brand Choice ................................................. 77

4-8 Group Classification: Beer Brand Choice Stepwise Discriminant Analysis ...................79

4-9 Multiple Discriminant Analysis: Beer Consumption Behavior............... ...................84

4-10 Structure Matrix for Beer Consumption Behavior ............... ..................................86

4-11 Group Classification: Beer Consumption Behavior Multiple Discriminant Analysis.......86

4-12 Mean and Standard Deviation: Beer Consumption Behavior.......................................87

4-13 Stepwise Discriminant Analysis: Beer Consumption Behavior .....................................88

4-14 Group Classification: Beer Consumption Behavior Stepwise Discriminant Analysis ......89









LIST OF FIGURES


Figure page

2-1 Theory of Reasoned Action: Brand Choice/Consumption of Beer ............................. 35









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 Advertising

INFLUENTIAL FACTORS ON BRAND CHOICE AND CONSUMPTION BEHAVIORS: AN
EXPLORATORY STUDY OF COLLEGE STUDENTS AND BEER

By

Dave Ritter

May 2008

Chair: Hyojin Kim
Major: Advertising

This study examined factors influencing beer brand choice and beer consumption

behavior among college students. It was determined after using six situational groupings that

situational variation did not have a significant impact on brand choice of beer for college

students. Relationships among brand benefits, exploratory shopping behaviors, susceptibility to

interpersonal influence, product category involvement, and demographics were studied in order

to determine what factors had the most significant impact on brand choice. It was determined

that price and risk-taking behaviors had a significant impact on beer brand choice. Consumption

behaviors among college students were influenced by emotions. In addition, an association

between situation and consumption behavior existed among college students consuming their

favorite beer. Bars, clubs, or parties demonstrated heavy to moderate drinking behavior.

Situations where college students are relaxing such as at the pool/beach, at home with friends,

and not a party demonstrated light beer drinking behaviors.









CHAPTER 1
INTRODUCTION

Studies have shown that there are vast numbers of heavy users for consumption of alcohol

among college students in the United States. According to a study, 31.5% of 17 to 25 year olds

reported heavy usage once in the past thirty days (Faden, 2006). Over 80% of college students

drink alcohol (Johnston et al. 2004), and more than 43% of these students have been reported as

exhibiting heavy episodic drinking behaviors (Wechsler et al. 2003). Demographically, college

students that have a greater likelihood of being a heavy user are white, around 23 years old or

younger, and are in a fraternity or sorority. Among members of Greek organizations, 75.1% of

fraternity members and 62.4% of sorority members are heavy users. Out of all heavy users, 91%

of women and 78% of men do not consider themselves such. Instead, they classify themselves as

moderate or light drinkers (Boulard 2005).

Beer is the majority of overall consumption for alcoholic beverages involving heavy

consumption with this consumer segment (Coate and Grossman, 1988). According to a Fall 2005

report from MRI data, nearly half (43.9%) of consumers aging from 18 to 24 purchase beer/ale

products. This group is 50% more likely (index of 150) to have purchased beer/ale in the past six

months in comparison to all other age groups. In addition, this age group is 21% more likely

(index of 121) to consume at least 5 beers in the last 7 days, which is also the most for any other

age group ("Fall," 2006).

There are factors that contribute to the opportunity for beer to be a marketable product in a

college market. First, according to a Student Monitor's Lifestyle and Media study, college

students consider drinking beer one of the top "in" things on their campuses. This study found

drinking beer rated as the second "in" thing on campuses at 71%, tied with the college

networking website Facebook.com (Snider, 2006). Second, there is a relationship among the









availability and consumption of alcoholic beverages. Therefore, as availability increases, then the

consumption of alcoholic beverages increases (Gruenewald et al. 1992; Gruenewald et al. 1993;

Jones-Webb et al. 1997). Third, the college market has a large, disposable income. In addition, as

beer availability often exposes this market to beer brands for the first time, consumers begin

trying brands and formulating their consideration set for this product category. Brand preferences

for beer often begin in college, and college students have plenty of income to spend on it with

$231 billion annually ("Beer," 2005; Ballantyne et al, 2006).

In efforts to reach the college market, marketers have used many different strategies

("Beer," 2005). First, pricing strategies have been used based on the assumption that as the price

decreases, then consumption increases (Levy and Sheflin, 1985). Second, marketers have utilized

advertisements at the point of purchase and happy hours, offering discounts on drinks in bars

during certain allocated periods of time (Kuo et al. 2003; Christie et al. 2001). In addition, some

companies use bar promotions such as theme nights or brand sales representatives in the

establishment of purchase to promote their brands. Third, marketers have utilized college

sporting events for reaching college students with event sponsorships ("Alcohol," 2005). Finally,

companies have been more aggressive using innovative techniques such as viral marketing

campaigns on popular websites such as www.youtube.com and targeting areas just off campus

for promotions and advertising ("Viral," 2006; "Domestic," 2006).

There are some problems involved with the beer market and college students. First of all,

despite high advertising expenditures (advertising spending of $478,000,000 for the first half of

the year), the beer product category has been losing market share ("U.S.," 2007). There has been

a decrease in the consumption of beer based on increased usage of mixed drinks and other

alcoholic beverages. Second, there are many health concerns associated with beer on college









campuses. Companies have a responsibility to promote safe drinking practices based on heavy

usage (Boulard, 2005). As the level of awareness for beer on college campuses increases, this

could potentially lead to negative consequences for sponsorships on or off campuses (Kuo et al.

2003). Third, marketers are currently battling pricing and promotional wars with competitors

based on heavy product proliferation. These promotions have little consistency among a

branding strategy. Therefore, strategies that do not contribute to a brand could lead to increased

brand switching for consumers ("Beer," 2005; Kuo et al. 2003).

If marketed effectively, there is a tremendous opportunity in the college segment.

According to the 2000 U.S. Census Bureau, this market is comprised of 17,472,000 people (Day

and Jamieson, 2005). It has been said that this group exhibits low brand loyalties. However,

according to Wood (2004), "To suggest that young consumers have low loyalty would be to miss

the richness of their complexity" (p. 13). This study is aimed at helping marketers have a better

understanding of the current beer and college market. It also aids in finding what factors

influence brand choice and consumption behavior for this product category. Information from

this study can help marketer's position beer brands and help them segment the market for college

students based on the benefit needs of this group (Smith, 1956; Bagozzi, 1986). Marketers can

use information from this study to develop the most cost effective advertising and promotional

messages for this market (Vazquez et al. 2002). From a research perspective, there are few

studies involving brand choice, the college market, and the product category of beer. Therefore,

this study could lead to further research studies in the future. This study tests the validation of the

desired brand benefits and choice model as a predictor of brand choice (Orth, 2005). Finally, this

literature adds to existing studies on brand choice.









CHAPTER 2
LITERATURE REVIEW

Brand Choice

Understanding and predicting brand choice decisions by consumers has been a topic of

interest to both marketers and researchers. Brand choice investigation involves understanding

consumer behaviors in their selection of brands among various product categories (Bentz and

Merunka, 2000). In the past, brands have been perceived as products with different attributes;

however, brands are now viewed as personalities, identities, and have special meanings intrinsic

to consumers (Ballantyne et al. 2006). Brand choice research has been investigated for many

years and has intensified as product categories have become more proliferated. For example, 30

years ago there were only a handful of beer brands in grocery stores. Now, there are several

brands of beer with brand extensions featuring light beers, imports, ice beers, as well as many

others. Consumers have more options and many different brands to choose from (Leger and

Scholz, 2004).

Much of brand choice research has been through probability models to test the impact of

marketing mix variables as a predictor of brand choice (Wagner and Taudes, 1986; Chib et al.

2004; Bentz and Merunka, 2000). These variables (referred in most research studies as the 4 P's)

are elements such as product features, displays (i.e. advertising, sales promotions), availability

(stock of inventory), and price (Chib et al. 2004, May; Bentz and Merunka, 2000; Wager and

Taudes, 1986). When used in probability modeling, marketing mix variables are considered non-

stationary and heterogeneous among the population (Wagner and Taudes, 1986).

There are other areas that have been researched with brand choice as well. Researchers

have examined the casual effects of brand related variables on brand choice. These variables

include situational factors, consumer personality, social benefits, emotions, quality, brand









credibility, product attributes, seasonality, and trends. The studies used within brand choice

researches have involved experiments and surveys of key marketing variables to measure impact

on brand choice (Charlton and Ehrenberg, 1973; Simonson et al. 1994; Erdem and Swait, 2004;

Wagner and Taudes, 1986; Orth, 2005). Table 2-1 demonstrates these brand choice studies.

Among specific marketing mix variables, pricing appears to have the most consistent

impact in studies. Promotions such as sales promotions have shown influence on brand choice

which ultimately effect bottom-line prices for consumers. For example, pricing promotions could

involve coupons or simply a reduction of price within the product category (Singh et al. 2005;

Papatla and Krishnamurthi, 1996; Wagner and Taudes, 1986; Orth, 2005). In probability

modeling studies, it has been shown that displays and features have some impact on brand

choice, but this evidence is not as overwhelming or as consistent as other factors among brand

choice research studies (Chib et al. 2004; Papatla and Krishnamurthi, 1996; Alvarez and

Casielles, 2005). Product attributes have high importance on discovering what areas of the

product can be altered in order to make their brand more appealing to the consumer. According

to current research, it has been found that the greater the number of brand attributes for a

product, then the more likely the consumer is to make that particular band choice (Greenwald et

al. 1986; Romaniuk, 2003). Product attributes are important to marketers in order to differentiate

products from their competitors (Aaker et al. 1992; Belch and Belch, 1995).

Non-marketing mix variables have been researched in order to discover external factors

that impact brand choice. Seasonality and trends have been researched with brand choice.

However, their outcomes depend upon the product category. For example, a product such as

laundry detergent will most likely have better sales figures in the summertime when the weather

is more favorable and people are outside more (Wagner and Taudes, 1986). Personality factors









have shown an impact based on what brands consumers buy. Brand credibility has shown

significance in determining brand choice as well (Erdem and Swait, 2004; Fry, 1971). Other

areas such as purchase time, purchase order, and product name have been researched but have

not been deemed to be main factors in determining a brand choice decision (Charlton and

Ehrenberg, 1973). These studies allow marketers to understand consumer switching behaviors

and allow for market share penetration, which give marketers a better understanding of what

elements effect a particular brand or product category (Chib et al. 2004; Wagner and Taudes,

1986).

Several product categories have been used in order to study brand choice. The majority of

product categories include low consumer involvement retail products. Some examples of

products studied in the past with brand choice are laundry detergent, soda, athletic shoes,

ketchup, coffee, snack foods, and bar soaps. Table 2-1 provides a listing of the various product

categories used in previous brand choice researches (Wagner and Taudes, 1986; Chib et al. 2004;

Erdem and Swait, 2004; Baumgartner, 2003; Papatla and Krishnamurthi, 1996; Alvarez and

Casielles, 2005; Berne et al. 2004; Singh et al. 2005; Auger et al. 2003).

Among previous brand choice literature, there have been very few studies involving the

product category of beer. Woodside and Fleck Jr. (1979) conducted a qualitative study regarding

brand choice of beer drinkers. The methodology for this study consisted of two in-depth personal

interviews with two beer drinkers. The researchers concluded that involvement, normative,

situational, and product attributes all influenced brand choice in the study. Charlton and

Ehrenberg (1973) conducted an experiment with the product category of beer where variables

manipulated were price, purchase time, purchase order, product name, and brand name. More

recently a study was conducted (Orth et al. 2004) which examined craft beer preference and the









relationship of brand benefits with consumer demographics. Brand benefits were considered to

be significant drivers of consumer preferences in this product category. Brand benefits were

shown to be an effective predictor in the product category of beer for brand choice.

Table 2-1 Brand Choice Studies
Author Independent Dependent Product Methods
Variables Variables Categories
Studied
Orth (2005) Situations (Host, Brand Choice Wine Electronic
Gift, Self) Survey
Quality*
Social Benefits*
Price*
Emotional*
Health
Environment
Wagner and Marketing Mix Purchase Rate Laundry Testing of
Taudes (1986) (Advertising, Detergent Multivariate
Price)* Brand Choice Polya Process
Seasonality* Probability Model using
Trends* Consumer
Panel Purchase
Data
Chib et al. Marketing Mix Brand Choice Soda Testing of
(2004) (Price, Feature, (Beverage) Model of Brand
Display)t Choice with
Scanner-Panel
Data
Erdem and Brand Credibility* Brand Athletic Shoes, Survey
Swait (2004) (Expertise, Consideration, Cellular
Trustworthiness, Brand Choice Providers,
Perceived Quality, Headache
Perceived Risk, Medication,
Information Cost Personal
Saved) Computer,
Shampoo
Papatla and Price* Brand Choice Laundry Testing the
Krishnamurthi Detergent Utility Model
(1996) Sales Promotion Using
(Display*, Household
Feature*) Scanner Data
Miller and Situation* Brand Choice Fast Food Survey from
Ginter (1979) Attributes* Restaurants Mail Panel









Table 2-1 Continued
Author Independent Dependent Product Methods
Variables Variables Categories
Studied
Simonson et Quality Rating (Brand Choice) Brownie Mix Experiment
al. (1994) Quality (Three Studies)
Brand Name 35 mm Film
Promotion
Price CD Player
Premiums Consumer
Need
Product
Features
Romaniuk Product Brand Choice Fast Food Survey
(2003) Attributes* Market

Benefit
Attributes*

Situation-Based
Attributes*
Fry (1971) Personality Brand Cigarettes Experiment
Variables* (Sex, Preference with a Field
Social Class, Study Panel
Self-Confidence,
etc.)
Alvarez and Sales Promotions Brand Choice Soda Testing of
Casielles (Price*, (Beverage) Brand Choice
(2005) Reference Price, Models using
Losses and Gains, Logit Models
Sales Promotion from Consumer
Techniques) Panel Data
Bern et al. Price Brand Choice Ground Coffee Testing of
(2004) Brand Brand choice
Logit Models
Coffee Type using
(Blend, Natural, Consumer
Special) Panel Data

Promotional
Discount

Consumer Type
(Regular or
Occasional
Shopper)t









Table 2-1 Continued
Author Independent Dependent Product Methods
Variables Variables Categories
Studied
Baumgartner Price Brand Choice Ketchup Testing the
(2003) Multinomial
Promotion Brand Loyalty Coffee Logit Model
for Time
Goodwill Variations with
Brand Choice
Using Panel
Data.
Auger et al. Basic Product Brand Bar Soaps Experiment
(2003) Features (i.e. Preference
Weight, Ankle Athletic Shoes
Support, Price)

Ethical Features
(Tested on
Animals, Child
Labor,
Biodegradable)

Consumer
Personality

Demographics
Singh et al. Product Brand Choice Pretzels Testing of
(2005) Attributes Potato Chips Multicategory
(Price*, Feature, Tortilla Chips Brand Choice
Display, Flavor, Mayonnaise Model using
No Salt/Light*, Sliced Cheese Household
Pack Sizes*, Panel Data
Brand Names*)
Charlton and Price Brand Choice Beer Experiment
Ehrenberg
(1973) Purchase Time

Purchase Order

Product Name

Brand Namet









Table 2-1 Continued
Author Independent Dependent Product Methods
Variables Variables Categories
Studied
Orth et al. (2004) Brand Name Consumer Craft Beer Online Survey
Preferences from Consumer
Functional Panel Data
Benefits

Price/Value

Social Benefit

Positive
Emotional
Benefit

Negative
Emotional
Benefits
Bentz and Marketing Mix Brand Choice Instant Coffee Testing of the
Merunka (2000) Variables (Price Store Purchases Multinomial
per Quantity, Logit Model in
Promotional Chocolate Combination
Price Cut as a with Neutral Net
Percentage of work Model
Normal Price) using Panel
Scanner Data
Product
Characteristics

Household-
Specific
Variables (Brand
and Size
_Loyalties)t
* Indicates variables were found to be significantly associated with brand choice.
f Indicates these studies involved interaction effects among dependent variables but had no main
effects on individual dependent variables.

Desired Brand Benefits

Researchers do not always account for separation of effects for brand name with product

attributes. Keller (1993) suggests that the brand name creates added benefits separate from the









product for consumers. Benefits are personal values that consumers associate with a product or

service. It is what the consumer believes the product or service can do for them (Park et al.

1986). Brand benefits create a value by the brand name (i.e. logo, design), which transcends the

functional value of the product. Brand benefits focus on the needs that the product fulfills for the

consumer (Lancaster, 1971; Haley, 1968; Farquar, 1989, p. 24; Orth et al. 2004). Some

researchers believe that consumers make purchases based on the product benefits and not the

attributes offered. However, consumers evaluate purchase decisions based on product attributes

with the promise of benefits received from product attributes (Haley, 1968; Aaker et al. 1992;

Belch and Belch, 1995; Puth et al. 1999). A recent study has shown that consumers do not

always seek both attributes and benefits in products. Consumers tend to seek benefits when

involved with simple products: low involvement (i.e. food products). Consumers seek attributes

when dealing with a more technical product: high involvement (i.e. television, automobile)

(Bozinoff and Roth, 1984).

More recently research has focused on both brand name and brand benefits that led to a

brand choice by consumers. Brand benefits have been analyzed in terms of dimensions that

impact brand choice. Findings have been discovered by researching brand benefits that brands

outperform and are more actionable than previous research studies in product attributes.

Dimensions that have been researched in the past include performance/quality, value-for-money,

emotion, health, social, and environmental benefits (Orth 2005; Orth et al. 2005). Six brand

benefit measures were shown to be significant in measuring brand choice in a previous study

(Orth 2005).

Performance/Quality

Quality refers to the degree of excellence in a product or service (Xianhua and Germain,

2003). Therefore, quality is one of the most important factors influencing customer satisfaction









(Fornell et al. 1996) and is considered the ability of a product or service to perform its specific

task (Ennew et al. 1993). The success of a brand in customer satisfaction is quality. Companies

conform to requirements set by consumers (Berden et al. 2000). Quality is significant on the

performance of a product (Calantone and Knight, 2000). The interaction of a product meeting or

exceeding consumer expectations based on its performance is how quality is evaluated (Fomell

et al. 1996; Reeves and Bednar, 1994). Performance specifications generally define how quality

is judged for products (Ennew et al. 1993). Findings from research indicate that marketing

strategies, differentiation, cost leadership, and focus are drivers of quality (Calantone and

Knight, 2000).

Product quality adds many benefits for a company. Product quality allows companies to

charge higher prices to consumers. In addition, having a higher product quality gives a

competitive advantage which leads to gains in profit margins and market share. However,

research has shown that quality may not equate to success without the proper marketing

techniques in order to reach and communicate with consumers (Calantone and Knight, 2000;

Choi and Coughlan, 2006).

Quality is not defined as a situation of spending money to make money. Often times a

product's quality can be improved by reducing waste, fewer dissatisfied consumers, and being

more efficient in the production of the product. There has been research to support the theory

that companies do not have to incur costs to make their product superior in order to have superior

quality. Instead, attention to quality as a differentiating approach in dealing with competitors

often can make a larger overall impact on quality (Calantone and Knight, 2000; Berden et al.

2000). Quality is important for impacting brand choice because it is the portion of personal risk









that a consumer takes on the decision making processing in evaluating the purchase of a product

(Berden et al. 2000; Hoyer and Maclnnis, 2004).

Price/Value for Money

In retail markets, consumers are value driven, where value is considered a tradeoff among

price and value. Price can serve as an indicator of quality for consumers. The higher the price of

a product, the more perceived risk a consumer incurs (Quester and Smart, 1998). In general,

consumers often associate a high-priced retail product with higher quality than those of lower

pricing (Lambert, 1972). However, some researchers believe that this quality and price

relationship is too simplistic (Sweeney and Soutar, 2001). Prices are used by marketers in retail

stores in order to appeal to different consumers on different levels. The consumer uses

comparative judgments in order to evaluate a potential purchasing decision. The consumer

utilizes reference prices in order to make these comparisons (Alvarez and Casielles, 2005).

Reference pricing is a subjective price level that is used by the consumers to determine if the

product is at an acceptable price for purchase (Mayhew and Winer, 1992). Brands in most

product categories have a wide range of different prices. These prices vary for a vast number of

reasons (advertising, lower economies of scale, premium brand positioning, generics, and several

other factors). These prices demonstrate information perceived in many different ways by

consumers. A consumer might perceive a lower priced product to be considered "cheap" or

having low quality, whereas a different consumer could potentially see the low cost as a good

value (Hruschka, 2002; Lambert, 1972).

Therefore, price is a major factor in determining brand choice. First, several studies have

been conducted in order to determine the effect of price on alcohol consumption. Studies have

found an inverse relationship for sales and pricing. For example, as price of alcohol beverages

increase, then sales for these products decrease and vice versa (Osterberg 1995; Levy and









Sheflin, 1983). Second, the consumer wants the best product at the best price. Therefore, a higher

priced item will have more economic risk, but higher priced goods are more visible to others

socially. For example, some consumers choose to never purchase generic products because they

believe the quality of the product to be inferior. In addition, they have a social fear that others

will perceive that they are not economically well off (Hoyer and Maclnnis, 2004).

Emotions

Consumers can develop emotional feelings for products, specifically brands. These

emotions toward brands can have a major influence based on brand choice. Research has shown

that emotions lead to an interaction with the product on a personal level (Bowlby, 1979; Hazan

and Shaver, 1994). These emotions can lead to brand loyalty, paying premiums, and influencing

others to purchase the brand. Therefore, a consumer's emotional attachment to a brand may be

able to predict their commitment and willingness to make sacrifices to obtain it. Some basic

ideals that are associated with this emotional involvement for brands are a positive brand

attitude, high involvement in the product category, brand loyalty (willingness to pay a premium),

affection, passion, connection, and the overall satisfaction associated from the brand (Thompson

et al. 2005).

Normative/Personal (Social)

Social influences consist of influential factors determined by family and friends. College

students have more of a propensity to drink the brands that their parents and friends consume on

a regular basis. When children leave their parent's home to join the workforce or go off to

college, then a majority of them are taking their parents' purchasing behaviors with them. These

behaviors may diminish over time as the young adult is separated from their family, but the

influence is still apparent (Feltham, 1998). In addition, adolescents are exposed to peer-pressure

and group-think mentalities, which lead them to consuming brands that their friends and peers









consume (Collins et al. 2003). This social influence stems from persuasion by attitudes and

behaviors of fellow peers (Jessor, 1981; Kandel, 1980). Therefore, normative influences can

have an affect on brand choice for the beer product category. Throughout research on social

behavior, other individuals' behaviors may serve as cues which could increase the potential for

behavior. In addition, the behavior of others might remind the individual that alternatives to their

own behavior are available (Bandura, 1977).

Social influence has an affect on brands that consumers choose. There is a social risk

associated with every purchase decision a consumer makes. Opinion leaders, family/friend

influence, reference groups, social class, culture, and subculture can affect the brands that a

consumer purchases. This social risk is often associated with what the consumer believes are

acceptable brands based on the brand perceptions in the individual's social group. For example, a

consumer may purchase a higher priced, upscale brand in order to identify and be accepted by a

higher social class (Hoyer and MacInnis, 2004).

Environment (Stewardship)

Stewardship involves a company or brand taking an active responsibility for the

environmental impact of their product. This can come from the design, manufacturing, usage

from consumers, disposal of the product, and literature of the product to stay within the

boundaries of government law, industry standards, and consumer standards (Bruen, 2002). In

addition, product stewardship involves the environmental concerns involving health and safety in

all phases of the brand's life cycle (Braglia and Petroni, 2000; Hickle and Stitzhal, 2003; White

and Pomponi, 2003).

Environmental management has become a major issue in today's business world. New

"green" strategies are making companies rethink the production and waste practices involved

with company operations (Braglia and Petroni, 2000; Hickle and Stitzhal, 2003). Brands that do









not exhibit stewardship miss out on "green" marketing opportunities. Results from a study

analyzing the effects of stewardship concluded, firms that are more environmentally aware

derive more benefits from their "green" activities (Braglia and Petroni, 2000).

Consumer benefits that come from implementing stewardship are enhancing brand

reputation and image. In addition, stewardship can gain access to new markets, in some cases

reduction of costs (not paying fines and other environmental liabilities), and comply with

regulations. Firms that do not practice stewardship might have cost and regulatory benefits over

firms that do (Braglia and Petroni, 2000). An example of stewardship involving the product

category of beer would be if the can or bottle is recyclable or was recycled itself before

consumption.

Health

Historically, there has always been debate based on whether alcohol has been considered a

good thing or a bad thing for health purposes. Abraham Lincoln suggested in reference to

alcohol, "Many were greatly injured by it, but none seemed to think the injury arose from the use

of a bad thing but from the abuse of a very good thing" (Basler, 1953, p. 275).

Today, consumers are more health conscious than they have been in previous

generations. This trend has made marketers conform to these health concerns by catering

products and brands to meet consumer criteria. American's obsession with obesity has reached

the beer market. Low carbohydrate diets are becoming more and more popular, informing

consumers to cut carbohydrates from their diet. Beer marketers have recognized this need and

have begun to follow suit by marketing light beers aggressively (Coxe, 2004; Walker, 2004).

Research studies have shown that drinking alcohol in moderate amounts has consumer

benefits with cardiovascular health through blood thinning properties in alcohol. Also, alcohol

has been proven to ward off conditions such as bone health, cancer, hot flashes, heart attacks and









ischemic strokes. In addition, some studies actually determined that it showed protection against

dementia (Woods, 2005; Rhodes, 2005; Klatsky, 2006; Kondo, 2004). In addition, past studies

have shown that alcohol consumption can lower the risk of developing diabetes by improving

insulin sensitivity ("Is," 2004). The main medical health benefit of alcohol is a lowered risk of

coronary heart disease (Kondo, 2006; Klatsky, 2006).

With these recent consumer benefits, there are also risks associated with heavy

consumption of alcoholic beverages. There has been some evidence that drinking beer and other

alcoholic beverages in moderation (one drink a day) can increase rates of breast cancer in

postmenopausal women by 30% (Helliker and Ellison, 2005; Andorfer, 2005). Drinking beer

does add calories to a diet, and if these calories are not burned off, then the cardiovascular

benefits are void (Tufts University Health and Nutrition Letter, 2004). The beer market has

started a shift toward lower carbohydrate beers in order to try and meet the needs of these

consumers ("Beer's," 2006).

Hops, an ingredient in beer products, provide cancer prevention and treatment of

osteoporosis. However, to get these benefits, an individual would have to drink large quantities

of beer, which add separate health risks (Loftus, 2005). Heavy drinking has many consequences

associated with it based on the alcohol content in beer. Heavy consumption can lead to liver

disease, high blood-pressure, increase the risk of cancer, and pancreas damage (Kondo, 2006).

Some beer companies are beginning to mention health benefits through marketing efforts.

However, it is something that is handled very delicately for liability purposes. There is research

that is supporting health benefits from drinking in moderation (Cioletti, 2006). There is a

perception that wine is healthier than beer. The University of Western Ontario found that one

beer has the same antioxidants as a glass of red wine. Therefore, beer companies are making









efforts to investigate and change these perceptions ("Is," 2004). Indicating health benefits to

consumers, such as promotion of drinking in moderation as a health benefit and low

carbohydrate beers, influences a brand choice decision for health conscious consumers (Coxe,

2004; Walker, 2004; Woods, 2005; Kondo, 2002). Many of the research studies involving health

benefits of beer are still in the exploratory stages and have not produced consistent results in

some areas. Therefore, companies and researchers are not suggesting for consumers to begin

beer consumption for health benefits. Instead they are merely pointing out that there are some

health benefits for those that currently consume beer (Klatsky, 2006).

Situational Factors

Benefits sought out by consumers can differ based on the situation that the consumer is in

(Yang et al. 2002). According to Belk (1974), "Situations may be defined as those factors

particular to a time and place of observation which have a demonstrable and systemic effect on

behavior" (p. 157). Consumers evaluate brands in different manners based on the situation

(Vazquez et al. 2002). It is suggested from previous research that situational factors are a better

predictor for consumer behavior than measures involving consumer attitudes. Research has

indicated that consumer preferences change according to their environment (Quester and Smart,

1998; Lai, 1991, Belk, 1974).

According to Lai (1991), there are three types of situations that are used in marketing

strategy among situational factors: communication situation, purchase situation, and

consumption situation. Situational drivers should have a frequent number of customers per

situation. In addition, each situation must be clearly different than the other in order to account

for variance measures. Therefore, effects from environmental factors are not homogenous but

rather heterogeneous (Miller and Ginter, 1979; Yang et al. 2002).









A consumer might choose a brand based on being in different situations and will therefore,

be motivated to drink a certain brand (Yang et al. 2002). According to drinking studies, around

80% of young people's total alcohol consumption occurs at a public place (Knibbe et al. 1991).

The greatest occurrences of drinking are in the home or in bars (Wilks and Callan, 1990). In

addition, heavy and light drinkers tend to drink twice as much during "happy hours" in bars than

they do during times that are not involved in such promotions. Therefore, there are some

interaction effects of brand benefits based on situational factors (Babor et al. 1978; Orth, 2005).

Consumers may face similar environments, but there are several motivating conditions that

play a role on brand choice depending on the consumer (Yang et al. 2002). Several studies have

shown this idea of situational influences proving that individuals prefer to drink different brands

based on different occasions (Bearden and Etzel, 1982). For example, Quester and Smart (1998)

used the purchase of a bottle of red wine for a drink during the week (alone or with one's family)

over dinner, for a dinner party at a friend's house on a weekend (with 5 to 6 close friends), and as

a gift for an employer or respected friend. Orth (2005) evaluated three different situations based

on drinking red wine with the same scale from Quester and Smart. Miller and Ginter (1979)

explored situational impacts on brand choice with respect to fast food restaurants. The situation

variations analyzed were lunch on a weekday, snack during a shopping trip, evening meal when

rushed for time, and evening meal with the family when not rushed for time. All of the studies

involving situational factors demonstrated significance based on impacting brand choice (Orth,

2005; Miller and Ginter, 1979).

Areas that have been studied with situational drivers include product involvement, brand

choice, and product attributes. High product involvement was considered a factor that influences

behaviors with the interaction of situational drivers. Product factors have different levels of









importance to consumers based on situation. Brand choice has been found to be impacted

significantly by situational factors (Orth, 2005; Quester and Smart, 1998; Miller and Ginter,

1979; Yang et al. 2002).

It is important for marketers to understand where brands are effective in given situations.

This gives marketers insights as to where the brands are being effectively communicated,

purchased, and consumed (Miller and Ginter, 1979; Quester and Smart, 1998). However, one

study has argued with these notions. Results from a research study using a probability models to

determine preferences indicated that marketers do not have to make their brands congruent to

consumers or their environment. It is suggested that the source of brand preferences must be

understood in order to have an impact on situational factors that influence brand choice (Yang, et

al. 2002).

Situation variation depends on the product category used for research (Belk, 1974). Beer is

an important category to use because it is a narrowly defined product category in accordance

with researching situational drivers (Miller and Ginter, 1979). Drinking beer is considered an

activity that may occur in distinct situations. Therefore, there should be a clear variance

according to their changing environment (Yang et al. 2002).

Consumer Factors

Marketers and researchers have studied consumer factors in order to have an understanding

of what characteristics and traits (such as consumer demographics, susceptibility to interpersonal

influence, product category involvement) impact purchasing decisions. In addition, consumer

factors are utilized to target and segment populations (Park and Lessig, 1977; Bearden et al.

1989; Quester and Smart, 1998). Consumer behaviors (such as exploratory behavior, product

usage, and frequency of purchase) have been researched in the past in order to have an

understanding of choice (Raju, 1980, Redman et al. 1987; Uncles and Ehrenberg, 1990). These









variables have been linked historically in research as potential drivers of situational variation

based on brand choice with brand benefits (Orth, 2005; Orth et al. 2004)

Exploratory Shopping Behaviors

It is important to understand brand switching and exploratory behaviors in consumer brand

choice decisions. Previous research has indicated a link between situational drivers and

behaviors based on personality traits of a consumer. According to research by Raju (1980),

consumers take a risk, seek variety, and have curiosity in purchase behaviors such as brand

switching. Risk taking involves the consumer's need for innovation and alternatives in which

they are not familiar with. There is more risk perceived with this behavior. When a consumer is

variety-seeking, they are looking for alternatives that they are familiar with. The final consumer

exploratory tendency is curiosity-motivated behavior, which the consumer seeks out information

about a product or service through shopping or interpersonal communications (Raju, 1980;

Wahlers et al. 1986).

Product Categories such as beer are a good fit for exploratory behaviors based on the

desire for variety and brand-related factors. Consumers begin to have boredom with a brand and

seek alternatives in product categories similar to beer. In extreme cases where consumers have a

high level of involvement, these exploratory behaviors may not exist. For example, if the

consumer has a level of brand loyalty and habitually purchases the same brand irrelevant of

factors influencing choice, then this consumer will most likely not seek alternative brands (Van

Trijp, 1994; Van Trijp et al. 1996).

Interpersonal Influence

Social influence is major driver of brand choice. However, it is important to understand an

individual consumer's susceptibility to social influence in order to have an understanding of how

much social influence impacts their purchase decisions. If there is a high degree of social









influence, then the consumer could potentially change perceptions and purchasing behaviors

(Batra et al. 2001). Interpersonal influence is the individual traits or characteristics that impact

social influence within an individual (McGuire, 1968). Consumer susceptibility to interpersonal

influence has an impact on brand choice behaviors (Witt, 1969; Stafford and Cocanougher,

1977). Previous research indicates that interpersonal influence has a direct effect on personal

behaviors (McGuire, 1968). There are varying degrees of social impact within each individual

consumer. These levels indicate how much social influence an individual is susceptible to in

contrast with other individuals (Bearden and Etzel, 1982).

Consumption Behaviors

There are several behavioral factors that play a role in determining brand choice for

consumers. Product usage is among one of these factors and plays a major influential role in

impacting consumer behavior (Ram and Jung, 1989). Product usage consists of two dimensions:

usage variety and usage frequency (Zaichkowsky, 1985). Variety usage is how the product is

used and depends upon the product category and situation. Market share for product brands

could increase based on an event or situation. For example, sales and volume for specific brands

of beer could fluctuate before and after the super bowl (Ram and Jung, 1989). Therefore, brand

choice measurements should take into account for temporal changes in brand choice behavior

(Wilkie, 1986). There have been studies that have compared the differences in drinking

consumption of males and females. These studies included usage measures consisting of quantity

per occasion, average volume, and frequency of drinking (Green et al. 2004). According to

previous studies, there are three categories of drinkers in order to classify users: heavy,

moderate, and light users (Redman et al. 1987).

Frequency of purchases, the second dimension of product usage, deals with the amount of

a single item purchased during a given time period. According to a study, frequency of purchases









can provide insight on brand choice (Ram and Jung, 1989; Uncles and Ehrenberg, 1990). In

addition, expenditures on the product category itself also have some insights based on how

consumer's select a brand based on a product category (Orth, 2005).

Product Category Involvement

Park and Mittal define involvement as a state of mental readiness that impacts cognitive

resources for an action, object, or decision with consumer consumption (1985). Product category

involvement has a major effect on consumer decision making. If a consumer feels strongly

positive about the product category, then they are more prone to seek increased value, pay more

attention, and try to find the most product benefits among the product category (Quester and

Smart, 1998; Richins and Bloch, 1986). Product category involvement has been shown to

demonstrate considerable influence over consumer decision processing (Laurent and Kapferer,

1985).

Demographics

Demographic variables have been proven to be indicators for brand choice. Factors such as

age and gender play a role in how consumers evaluate and ultimately purchase brands in several

different product categories (Walsh and Mitchell, 2005). Based on studies involving

demographics and drinking behaviors, males tend to drink in larger quantities in same sex

groups, whereas women drink with mixed crowds or with a male (Hartford et al. 1983). Age is

also a variable to be explored for college students because there are those of legal age and others

that are obtaining beer illegally. There are a number of these college students that purchase beers

illegally via a false ID or by having an older peer purchase it for them (Schwartz et al. 1998). In

addition, there is very little known about demographic issues such as gender, age, and education

(year in college) with particular subject matter as it relates to this segment and brand choice.









Theory of Reasoned Action

Ajzen and Fishbein's (1980) Theory of Reasoned Action (TRA) is one of the most

researched models that describes the psychological processes of decision making. It is comprised

of three main components in order to predict behavior. The three components are attitude,

subjective norms, and intention. This model has been applied to many different areas of study

such as alcohol, marijuana, and purchasing consumer products (Eagly and Chaiken, 1993).

In this model, attitude involves the positive or negative associations an individual has on

specific behavior. Subjective norms deal with the normative and social influences that impact an

individual's behavior. Social influence on an individual and susceptibility to interpersonal

influence are factors that measure subjective norms. In a given population, there may be cases

that lean more towards attitude providing more influence in terms of behavior. However, in other

cases, subjective norms might potentially lead to a different behavior (Trafimow and Fishbein,

2001). Other influential factors could be intrinsic and extrinsic. They result from situational

and/or interpersonal factors (Chatzisarantis and Biddle, 1998; Bagozzi et al. 1992).

The two main factors involved in TRA, attitude and subjective norms, lead to intention.

Intention is the likelihood of completing a certain behavior, and the relative importance of

normative influence and attitudinal considerations. Intention is utilized for understanding

judgment based on how a final decision is made (Ajzen and Fishbein, 1980). Consumer factors

such as demographics and consumption behaviors provide an understanding of intention.

Intention can give marketers an idea of how a consumer will behave toward particular brands

(Bagozzi et al. 1992).

According to previous research studies, other variables aside from attitude and subjective

norm can have an overall impact on behavior (Trafimow and Fishbein, 2001). Susceptibility to

interpersonal influence and social influence lead to subject norms in an individual. Quality, price,









emotion, environment, health benefits, and product category involvement deal with the

individual's attitude toward the brand. In addition, importance of subjective norms and attitudes

can vary depending upon the situation (Bagozzi et al. 1992). All of these components, either

weighing more heavily on subjective norm or attitude, lead to intention. This intention results in

an individual beer brand choice and beer consumption behavior.

There have been several studies involving the TRA model and alcohol research studies.

These studies involved predicting alcohol consumption behavior (O'Callaghan et al. 1997;

Trafimow, 1996; Wall et al. 1998). Most of these particular studies were used in efforts to

understand and predict drinking behaviors prevalent among college students. These studies were

ultimately used in order to curve drinking behaviors.

The TRA model has been utilized in this study to conceptualize research questions

involving beer brand choice and beer consumption behaviors. Figure 2-1 illustrates the TRA

model used for this study. The model has been extended in order to demonstrate all measures

involved in this study. This modified model is exploratory in nature in order to gather a

theoretical understanding among variables used in the study. The model lists the organization of

variables as they relate to the concepts and relationships in the model.















Price, Quality, Emotion,
Environment, Health, and
Product Category Involvement


Susceptibility to Interpersonal
Influence


Attitude


Intention


H


Behavior

Beer Brand Choice

Beer Consumption


Subjective Norms


Social Brand Benefits


Figure 2-1 Theory of Reasoned Action: Brand Choice/Consumption of Beer. [Reprinted with permission from Ajzen and Fishbein
(1980). Theory of Reasoned Action. Master thesis (Page 37, Figure 2-1). University of Florida, Gainesville, Florida


Consumer Factors
(Demographics, Exploratory
Shopping Behaviors and
Consumption Behavior)

Situational Variation


W-


I----------------


m,









Research Questions

Based on the previous sections, research questions were developed in order to have a

theoretical understanding of the relationship between brand choice, brand benefits, interpersonal

influence, consumption behaviors, situation, product category involvement, exploratory

behaviors, and demographics.

RQ1: How does beer brand choice vary by situation?

RQ2: How does beer consumption behavior vary by situation?

RQ3: Which factors (perceived brand benefits, exploratory shopping behaviors,

susceptibility to interpersonal influence, product category involvement, and

demographics) have the most significant impact on beer brand choice?

RQ4: Which factors (perceived brand benefits, exploratory shopping behaviors,

susceptibility to interpersonal influence, product category involvement, and

demographics) have the most significant impact on beer consumption behaviors?









CHAPTER 3
METHOD

Pretest

A pretest was administered in order to improve accuracy of measures and to test

reliability of pre-existing measures (Appendix A). Brand choice, consideration set formulation,

situation, and brand benefits were all asked to respondents with a predetermined brand,

Budweiser. This brand was chosen based on its awareness, popularity, and market share in the

beer product category. Beer consumption questions and demographics were asked in order to

gain knowledge in these areas before developing the main study.

Subjects

Participants for the pretest were selected from two Junior/Senior-level courses. These

courses were an advertising course and a health communications course at the University of

Florida. College students in these courses were offered extra credit in exchange for their

participation (N =46). The pretest results provided a profile of college student participants. The

average age of respondents was 21.61 (standard deviation = .95). The minimum age was 20, and

the maximum age was 25 years old. Out of the sample of 46, 30 were female (65.2%) and 16

were male (34.8%). Among students, 27 were employed (58.70%) and 19 were unemployed

(41.30%). The respondents had little Greek membership from this sample. There were 38

students that were not Greek members (82.6%) and 7 students were Greek members (15.2%).

Based on past month beer consumption, the top five brands in the pretest were Bud Light,

Blue Moon, Corona, Coors Light, and Miller Lite. Participants also answered questions

regarding beer brands that they were most likely to consume in the next month. These brands

were ranked, and the top brands were the exact same as the past beer consumption rankings.

These beer brand choice questions created some confusion among respondents. Therefore, it was









determined to make beer brand choice an open-ended question for the main survey in order to

account for individual consideration sets. Based on an opened-ended situational question, there

were 20 situation items resulting from this question. Brand benefits scales had satisfactory

reliability and were utilized for the main study. However, two health item questions were added

to the health benefits scale based on exploratory data from an open-ended question. These health

items added were "It is a good way to relieve your stress" and "It glorifies unhealthy drinking

behaviors". These questions were evaluated based on agreement or disagreement on a 7-point

Likert scale. Beer consumption behavior questions provided accurate and clear results to

measure the construct.

Design and Procedure

A paper questionnaire was used in order to collect data. The survey was administered and

collected during class time. Subjects completed separate forms in order to keep their anonymity.

Subjects were given 15 minutes to complete the survey. Students were asked regarding the

constructs of brand choice, consumption behavior, demographics, situational variation, and brand

benefits.

Measures

Brand Choice

Brand choice questions (Appendix A: Question 1 and 13) were asked with comprehensive

but not exhaustive lists of 31 beer brands. In addition to these choices, the respondent had the

option of choosing an "Other" category and if so, then to specify the brand. A "None of the

Above" option was used to account for non-beer drinkers. In the first question, respondents were

asked to check the brands of beer they have consumed in the past month. This question was used

in order to get a perspective of consideration set for participants. In the second brand choice









question (13), respondents were asked to rank order the brands of beer they were most likely to

consume.

Situational Variation

Situations (Appendix A: Question 3) were asked on a 7-point Likert scale based on where

a respondent would consume Budweiser beer (1 being Not Very Likely and 7 being Very

Likely). Eight factors were used, exploratory in nature, to get an idea of discovering strength of

situations among college students. In conjunction with this information, an open-ended question

was asked (Appendix A: Question 4) regarding other occasions or situations that were not listed.

Desired Brand Benefits

Previous brand benefit scales (Appendix A: Questions 6, 8-12) used by Orth (2005) were

tested for their reliability. Measurement for desired brand benefits were based on six dimensions:

quality/performance, price/value for money, social/normative, emotion, environment, and health

benefits (Orth 2005; Vasquez et al. 2002; Sweeney and Soutar, 2001; Orth et al. 2004).

Quality/performance consisted of 6 items. Price/value for money consisted of 4 items. Emotions

were measured with 4 items. Social/normative scales consisted of 11 total items. Environmental

brand benefits were measured with 3 items. Health benefits consisted of 2 items. All of these

items were used on a 7-point Likert scale involving the agreement of disagreement of statements

involving each benefit (1 = Strongly Disagree and 7 = Strongly Agree). Scale items were tested

for reliability and confirmed to be used in the main study based on their results. An open-ended

health benefit question was asked in order to potentially add more items to the health scale in

order to increase reliability (Appendix A: Question 7).

Consumption Behavior

Two questions were used to have an understanding of college student beer consumption

behavior (Appendix A: Question 2 and 5). These two questions were inquiries regarding number









of days drinking (per week) and quantity of drinks per occasion. These two values provided

product usage information.

Demographics

Demographic questions (Appendix A: Questions 14-20) were asked in order to have an

understanding of the respondent profile before the main study is conducted. Age, sex,

employment, Greek membership, living situation, academic year as a student, and college major

were asked to gain insight into college student respondents.

Main Study

Subjects

Participants for this study were selected among college students in Public Relations and

Advertising courses at the University of Florida. College students that participated in the study

were rewarded extra credit for their completion of survey materials. A sample of 222 participants

was used to complete the main survey. The total number of brands chosen for this study was 38

different brands of beer (Table 3-1). Among these brands, it was determined that with a 95%

confidence interval (Equation 3-1) only three brands would be used for this study:

95% CI = +/- 1.96 / (1/2/1 /N) (3-1)

The three brands extracted from the main study were the following: Blue Moon, Bud

Light, and Corona. Based on participants only choosing the three brands, in conjunction with

subjects that did not have consistent beer consumption behavior (have not consumed a beer in

excess of the last two weeks), the sample size was reduced to 98 participants. Based on these 98

participants, the average age was 20.69 (Minimum = 18, Maximum = 31, and Standard

Deviation = 2.28). A frequency table demonstrates the results from these demographic questions

(Table 3-2). The gender distribution was 75.5% females and 24.5% males. There were 55.1% of

respondents that were not employed and 44.9% were employed. There were 30.6% of Greek









membership and 69.4% that were not Greek members. Most of the college students (63.3%)

lived with a roommate or roommates in an apartment or house off campus.


Table 3-1 Beer Brand Choice (N


222)


Beer Brand Frequency Percent
Sam Adams 1 .5%
Busch Light 1 .5%
Bud Light 43 19.4%
Michelob Ultra 11 5.0%
Stella 3 1.4%
Coors Light 7 3.2%
Heineken 12 5.4%
Budweiser 9 4.1%
Corona 28 12.6%
Miller Lite 13 5.9%
Blue Moon 27 12.2%
Yuengling 8 3.6%
Heineken Light 1 .5%
Natural Light 8 3.6%
Amber Boch 4 1.8%
Guinness 3 1.4%
Bud Select 2 .9%
Killians 2 .9%
Sapporo 1 .5%
Hardcore 1 .5%
Shipyard Blueberry Ale 1 .5%
Labatt Blue 1 .5%
New Castle 2 .9%
Red Stripe 8 3.6%
President 2 .9%
Smimoff 2 .9%
Khalik Gold 1 .5%
Pabst Blue Ribbon 1 .5%
Michelob Light 1 .5%
Milwaukee's Best 1 .5%
Hornsby 1 .5%
Miller High Life 1 .5%
Grolsch Light 2 .9%
Goldwesser 1 .5%
Shiner Bock 1 .5%
Corona Light 1 .5%
Molson 1 .5%
Coopers 1 .5%
Total 214 96.4%









Table 3-2 Demographic Frequency Distribution
Frequency Percentage
Gender Male 24 24.5%
Female 74 75.5%
Age 18 5 5.1%
19 24 24.5%
20 29 29.6%
21 19 19.4%
22 10 10.2%
23 2 2.0%
24 5 1.0%
27 1 1.0%
29 1 1.0%
30 1 1.0%
31 1 1.0%
Employed Yes 44 44.9%
No 54 55.1%
Greek Yes 30 30.6%
Membership No 68 69.4%
Academic Freshman 10 10.2%
Year Sophomore 39 39.8%
Junior 28 28.6%
Senior 20 20.4%
Master's Graduate Student 1 1.0%
Academic Public Relations 25 25.5%
Major Advertising 27 27.6%
Journalism 2 2.0%
Other 44 44.9%
Living Live with Parents 1 1.0%
Situation Live Alone (apt/home) on Campus 1 1.0%
Live Alone (apt/home) off Campus 9 9.2%
Live with Roommate(s) (apt/home) on Campus 8 8.2%
Live with Roommate(s) (apt/home) off Campus 62 63.3%
Fraternity or Sorority House 6 6.1%
Live in Dorm Alone 2 2.0%
Live in Dorm with Roommate(s) 8 8.2%
Other 1 1.0%
N=98

Design and Procedure

A paper questionnaire was used in order to collect data for the main study. The survey was

administered and collected during class time. Subjects completed separate forms in order to keep

their anonymity. Subjects were given 30 minutes to complete the study. Students were asked









brand choice, consumption behavior, demographic, susceptibility to interpersonal influence,

product category involvement, exploratory shopping behaviors, situational variation, and brand

benefit questions.

Measures

Brand Choice

To measure brand choice, respondents were asked a fill in a blank question (Appendix B:

Question 1) asking them to write in their favorite brand of beer. This information allows the

respondent to select their favorite brand among that particular individual's consideration set

when it comes to the product category of beer.

Situational Variation

Based on pretest results, it was discovered that 20 situations existed (Appendix B:

Question 2) in which college students consume their favorite beer. Modeling for situational

variation was developed from Orth's study (2005). Consumption situations were varied by the

company of the individual drinking. Situations were varied with three possible scenarios:

drinking beer with a group of friends, drinking with a date, and drinking alone. Among the 20

situations possible, 13 situations were chosen by respondents in the main study (Table 3-3). Six

groupings were created with multiple categories in an attempt to limit the disparity among

situation data. Each grouping dealt with a respondent answering what situation they last

consumed their favorite beer. Groupings were assessed in order to find a significant relationship

based on similarities of the situation, frequencies of situations selected, and social interaction

based on situations. These situational groupings can be seen in left column on Table 4-1.

Group 1 consisted of 5 categories: in a bar or club with friends, at home with friends not a

party, a house/apartment party not in your home with friends, partying at home or pool/beach,

and other. At the University of Florida, most housing for college students has a pool. Therefore,









party at home was combined with pool/beach. This combination was utilized in order to

distribute the data more evenly among the categories. The other category contains eight

situations to contrast category size among these categories.

Group 2 consisted of 4 categories: in a bar or club with friends, at home with friends not a

party, a house/apartment party not at home with friends, and other. The first three categories

remained unchanged. However, the other category consists of ten situations combined in order to

contrast with the high numbers selected among the first three situations.

Group 3 was made up of in a bar or club, a restaurant or at home not partying, and

partying/special events. This first category consists of an environment when college students go

out to a bar or club and have the social interactions and situations that go along with the bar/club

scene on a college campus. In addition, the category of restaurant and at home not partying deals

with the college student that wants to relax and have a few drinks in a comfortable situation. The

final category deals with a special event such as a party, beach/pool, or sporting/concert. College

students participate in activities throughout the school year such as tailgating at football games,

music concerts, pool parties, and the occasional house/apartment party. In these environments,

students are attending with the intent of having a good time and enjoying these special events.

Group 4 contains two categories. The first category deals with college students drinking in

a bar, club or party situation. These students are looking to have a good time and social

interaction. The second category involves students not in bars, clubs, or parties. These students

are in a more relaxed, comfortable environment where there may not be as much social

interaction or pressure to consume large quantities of beer.

Group 5 has four categories consisting of social, date, alone, and relax with friends. The

social category contains situations where there is high social interaction. The date situation is









when the respondent was drinking beer with a date. The alone category dealt with situations

when the individual drank beer by themselves. Relax with friends was used in order to

demonstrates college students that were just drinking their favorite beer to relax and not

necessarily drinking to have a good time or socially interact.

Group 6 involved the same social, date, and alone categories as listed in group 5. Data

from relax with friends was omitted from this group. The three respondents that chose other on

the main survey were also omitted from groups 5 and 6 based on a lack of frequency for this

selection.

Table 3-3 Situation Frequency Distribution
Situation Frequency

In a bar or club with friends 35

In a bar or club with a date 1

In a bar or club alone 1

In restaurant with friends 4

At home with friends (not a party) 17

At home with a date (not a party) 2

At home alone (not a party) 2

Sporting event/concert with friends 1

Beach/pool with friends (not a party) 8

Party (not at home) with friends 20

Party (not at home) with a date 2

Party at your home 3

Other 3

N= 97









Desired Brand Benefits

Measurement for desired brand benefits (Appendix B: Questions 5-10) was consistent

with items in the pretest. However, respondents were answering statements for their favorite beer

and not a forced choice (Budweiser as used in the pretest). Brand benefit index scores were

calculated by using averages for each of the six brand benefit dimensions. Scale items were

tested for reliability. Quality/performance consisted of 6 items (Chronbach's alpha = 0.89).

Price/value for money consisted of 4 items (Chronbach's alpha = 0.85). Emotions were measured

with 4 items (Chronbach's alpha = 0.94). Social/normative scales consisted of 11 total items

(Chronbach's alpha= .90). Environmental brand benefits were measured with 3 items

(Chronbach's alpha = 0.86). Health benefits consisted of 4 items (Chronbach's alpha = .50).

When one item was removed, "It glorifies unhealthy drinking behaviors", then reliability

improved (Chronbach's alpha= .66). Finally, another item was removed ("It is a good way to

relieve your stress") in order to improve reliability scores (Chronbach's alpha = 0.74).

Exploratory Shopping Behaviors

Exploratory shopping behaviors (Appendix B: Questions 11-13) are a measure that was

adopted from many previous measures (Raju, 1980; Wahlers et al. 1986). The constructs used in

this study for exploratory shopping behavior were risk-taking, variety-seeking, and curiosity-

motivated (Orth, 2005). Items were measured on a 7-point Likert scale based on agree with

statements (1 = Strongly Disagree and 7 = Strongly Agree). Index scores were calculated for

each of the three exploratory shopping behaviors by using averages. Among these measures, risk

taking behavior consisted of 5 items (Chronbach's alpha = 0.83). Variety-seeking behavior was

made up of 4 items (Chronbach's alpha = 0.79). Curiosity-motivated behavior was a measure

made up of 3 items (Chronbach's alpha = .47). However, one item was removed ("When I hear









about a new store or restaurant, I take advantage of the first opportunity to find out more about

it") in order to improve reliability (Chronbach's alpha = 0.53).

Interpersonal Influence

Interpersonal influence measures were 12 items (Appendix B: Question 14) used from

previous research in order to measure the level of social influence within each individual

(Bearden et al. 1989). Items were measured on a 7-point Likert scale based on agree with

statements (1 = Strongly Disagree and 7 = Strongly Agree). A susceptibility to interpersonal

influence index score was calculated by using an average of the 12 items. These 12 items were

found to be reliable (Chronbach's alpha = 0.89).

Consumption Behavior

Beer consumption behavior consisted of three questions. The first question (Appendix B:

Question 3) was a question inquiring about basic beer consumption. This question was utilized to

discover respondents that were not ordinarily beer drinkers and remove them from the sample.

Beer consumption behavior was measured by creating a consumption index from two separate

measures (Appendix B: Questions 15 and 16). The first measure was the amount of beer on

average that an individual drinks in one occasion. The second measure consisted of the average

number of days a week in which an individual drinks. These two measures have been illustrated

in a cross tabulation on Table 3-4.

These measures were multiplied in order to create an index score. Index scores were

evaluated and grouped into light, moderate, and heavy beer drinkers based on their consumption

behaviors. These groups were divided equally into 3rds based on the frequency distribution

illustrated on Table 3-5. Light beer drinkers had a drinking index of 0 2. This means that a

college student that had a consumption index of 2 drank either 2 twelve ounce beers in one night

on average in a week, or they drank one beer on average for two nights a week. Moderate beer









drinkers had consumption indexes from 3 8, and heavy beer drinkers had consumption indexes

of 9 54. It is important to note that these beer consumption behavior levels differ in contrast to

the general population. Due to high beer consumption levels of this particular college market,

these measures should not be used for direct comparisons involving consumption behavior for

other beer markets.

Table 3-4 Average Beers Consumed and Average Occasions Drinking Beer
Average Average Number of Occasions per Week Drinking Beer Total
Number 0 1 2 2.5 3 4 5 6
of Beers
Drank
per
Occasion
0 7 0 0 0 0 0 0 0 7

1 6 4 0 0 0 0 0 0 10

2 4 7 2 0 1 0 0 0 14

3 3 5 10 0 2 0 0 0 20

4 1 7 3 0 5 1 0 0 17

5 0 4 3 1 1 0 1 0 10

6 1 0 3 0 2 1 0 0 7

7 0 1 0 0 3 0 0 0 4

8 1 1 0 0 0 1 0 0 3

9 1 1 0 0 0 0 0 1 3

10 0 0 0 0 0 1 0 0 1

14 0 0 0 0 1 0 0 0 1

17+ 0 1 0 0 0 0 0 0 1

Total 24 31 21 1 15 4 1 1 98











Consumption Behavior Frequency Percent (%)
Light Beer Drinkers
0 24 24.5
1 4 4.2
2 7 7.1
Total 35 35.7
Moderate Beer Drinker
3 5 5.1
4 9 9.2
5 4 4.1
6 11 11.2
7 1 1
8 4 4.1
Total 34 34.7
Heavy Beer Drinker
9 3 3.1
10 3 3.1
12 8 8.2
12.5 1 1
15 1 1
16 1 1
17 1 1
18 2 2
21 3 3.1
24 1 1
25 1 1
32 1 1
40 1 1
42 1 1
54 1 1
Total 29 29.6%
N=98

Product Category Involvement

Three items (Appendix B: Question 17) used from previous research were used in order

to measure product category involvement (DeWolf, 2001). Items were measured on a 7-point

Likert scale based on agree with statements (1 = Strongly Disagree and 7 = Strongly Agree).

Index scores of the three items resulted in satisfactory reliability (Chronbach's alpha = 0.94).


Table 3-5


Con slim nti on R eh avi or Frenllen cv









Demographics

Demographic questions used from the pretest were also utilized in the main study in order

to retrieve information in order to profile and identify respondents (Appendix B: Questions 18-

24).









CHAPTER 4
RESULTS

Brand Choice among Situational Variation

Chi-square analysis was used in order to explore the possible association between beer

band choice and the situation in which college students consumed their favorite beer. All data

was tested at level of p .05. Beer brand choice was a categorical variable consisting of three

brands: Bud Light, Blue Moon, and Corona. Situation was a categorical variable consisting of six

groups with categories for each situation. Table 4-1 illustrates the cross tabulation of these

results for the chi-square value, frequencies among groups with brand choice, and degrees of

freedom.

In grouping one with five situations, 45.71% that chose drank in a bar or club with friends

chose (16 out of 35). Bud Light was also chosen more when at home with friends not at a party

(41.18%, 7 out of 17), party not at home with friends (55%, 11 out of 20), and other (50%, 7 out

of 14). When college students drank beer partying at home or pool/beach area, then Corona was

the highest chosen (63.64%, 7 out of 11). These results were found to not be statistically

significant. Therefore, there is no significant relationship between categories in group one with

brand choice.

Group two with four situations yielded somewhat similar results as group one. College

students drinking in a bar or club chose Bud Light (45.71, 16 out of 35). Bud Light was most

chosen in all categories: bar or club with friends (45.71, 16 out of 35), at home with friends not a

party (42.18%, 7 out of 17), party not at home with friends (55%, 11 out of 20), and other (36%,

9 out of 25). Results were found to not be statistically significant and no relationship for these

four categories and brand choice exists.









In grouping three with three situations, results were not statistically significant. College

students that drank in a bar or club preferred Bud Light (45.95%, 17 out of 38). College students

that drank in a restaurant/home not partying preferred Blue Moon (39.29%, 11 out of 28) slightly

over Bud Light (35.71%, 10 out of 28). In a partying/special event, college students chose Bud

Light (50%, 16 out of 32) closer over Corona (37.5%, 12 out of 32).

In grouping four with two situations, college students in a party, bar, or club were more

prone to choose Bud Light (50%, 30 out of 60). College students not in a party, bar, or club

drank Bud Light (35.14%, 13 out of 37), Blue Moon (32.43%, 12 out of 37), and Corona

(32.43%, 12 out of 37). Results for grouping 4 with two situations indicated no statistical

significance.

In grouping five with four situations, college students in social situations drank Bud Light

more frequently (46.15%, 30 out of 65). College students that drank their favorite beer with a

date preferred Bud Light (80%, 4 out of 5). College students that drank their favorite alone were

equally distributed among brand choice at 33.33% (1 out of 3) for each brand. College students

that drank their favorite beer to relax with friends chose Blue Moon (42.86%, 9 out of 21). These

results did not provide statistical significance.

In grouping six with three situations, results were exactly the same as group five other than

omitting the relax with friends category. Therefore in group six, Bud Light was the most

dominantly chosen brand for these situations, but as in group five, results were also found to not

be statistically significant.

The chi-square analysis indicated that situation groups one (x2 = 12.17, p > .05), two (x2 =

2.55, p > .05), three (x2 = 6.15, p > .05), four (x2 = 2.11, p > .05), five (x2 = 6.89, p > .05), and









six (x2 = 3.06, p > .05) were not associated with beer brand choice. Results for all situation

groups were found to be not statistically significant with brand choice.

Table 4-1 Situational Groups and Brand Choice
Situational Items Included in Bud Light Blue Moon Corona Total
Grouping Grouping

1 Bar or club with friends 16 10 9 35
(45.71%) (28.57%) (25.71%)
At home with friends 7 5 5 17
(not a party) (41.18%) (29.41%) (29.41%)

Party (not at home) with 11 3 6 20
friends (55%) (15%) (30%)

At home alone (not a 2 2 7 11
party) (18.18%) (18.18%) (38.89%)

Beach/Pool with friends
(not a party)

Party at your home

In a bar or club with a 7 6 1 14
date (50%) (42.86%) (7.14%)

In a restaurant with
friends

Other

In a bar or club alone

At home with a date (not
a party)

Sporting event/concert
with friends

Party (not at home) with
date









Table 4-1 Continued
Situational Items Included in Grouping Bud Light Blue Moon Corona Total
Grouping

2 Bar or club with friends 16 10 9 35
(45.71%) (28.57%) (25.71%)
At home with friends 7 5 5 17
(not a party) (41.18%) (29.41%) (29.41%)

Party (not at home) with 11 3 6 20
friends (55%) (15%) (30%)
At home alone (not a party) 9 8 8 25
(36%) (32%) (32%)
Beach/Pool with friends
(not a party)

Party at your home

In a bar or club with a date

In a restaurant with friends

Other

In a bar or club alone

At home with a date (not a
party)

Sporting event/concert with
friends

Party (not at home) with
date









Table 4-1 Continued
Situational Items Included in Bud Light Blue Moon Corona Total
Grouping Grouping


In a bar or club with
friends

In a bar or club with a
date

In a bar or club alone

In a restaurant with
friends

At home with friends (not
party)

At home with a date (not a
party)

At home alone (not a
party)

Other

Sporting event/concert
with friends

Beach/pool with friends
(not a party)

Party (not at home) with
friends

Party (not at home) with a
date

Party at your home


17
(45.95%)


10
(35.71%)


16
(50%)


11
(29.73%)


11
(39.29%)


4
(12.5%)


9
(24.32%)


7
(25%)


12
(37.5%)









Table 4-1 Continued
Situational Items Included in Grouping Bud Light Blue Moon Corona Total
Grouping


In a bar or club with friends

In a bar or club with a date

In a bar or club alone

Party (not at home) with
friends

Party (not at home) with a
date

Party at your home

In a restaurant with friends

At home with friends (not
party)

At home with a date (not a
party)

At home alone (not a party)

Sporting event/concert with
friends

Beach/pool with friends (not
a party)

Other


30
(50%)


13
(35.14%)


14
(23.33%)


12
(32.43%)


16
(26.67%)


12
(32.43%)









Table 4-1 Continued
Situational Items Included in Grouping Bud Light Blue Moon Corona Total
Grouping

5 In a bar or club with friends 30 14 21 65
(46.15%) (21.54%) (32.31%)
Sporting event/concert with
friends

Beach/pool with friends (not
a party)

Party (not at home) with
friends

Party at your home

In a bar or club with a date 4 1 0 5
(80%) (20%) (0%)
At home with a date (not a
party)

Party (not at home) with a
date

At a bar or club alone 1 1 1 3
(33.33%) (33.33%) (33.33%)
At home alone (not a party)

In a restaurant with friends 7 9 5 21
(33.33%) (42.86) (23.81%)
At home with friends (not a
party)









Table 4-1 Continued
Situational Items Included in Grouping Bud Light Blue Moon Corona Total
Grouping

6 In a bar or club with friends 30 14 21 65
(46.15%) (21.54%) (32.31%)
Sporting event/concert with
friends

Beach/pool with friends (not
a party)

Party (not at home) with
friends

Party at your home

In a bar or club with a date 4 1 0 5
(80%) (20%) (0%)
At home with a date (not a
party)

Party (not at home) with a
date

At a bar or club alone 1 1 1 3
(33.33%) (33.33%) (33.33%)
At home alone (not a party)

1: x2 = 12.17, df = 8, N = 97, p-value =.14
2: x2 = 2.55, df = 6, N = 97, p-value = .86
3: x2 = 6.15, df = 4, N= 97, p-value = .19
4: x2 = 2.11, df = 2, N=97, p-value = .35
5: x2 = 6.89, df = 6, N = 94, p-value = .33
6: x2 = 3.06, df = 4, N = 73, p-value = .55
*p-value < .05
(%) Percentage of situation for a specific beer brand choice

Consumption Behavior among Situation Variation

An additional chi-square test was used in order to examine the relationship between

consumption behavior and situation variation. The same six situational groupings were used as in

the previous analysis. Beer consumption behavior was a categorical variable (beer consumption









levels of light, moderate, and heavy). Table 4-2 illustrates the results for the chi-square value,

frequencies among groups with brand choice, and degrees of freedom.

For group one consisting of five situations, college students that drank in a bar/club with

friends consisted of 40% moderate drinkers (14 out of 35), and 42.67% heavy drinkers (15 out of

35) drank in a bar/club with friends. Respondents that drank at home with friends were

dominantly light drinkers at 52.94% (9 out of 17). College students that drank their favorite beer

at a party not in their home with friends were light drinkers (35%, 7 out of 20) and moderate

drinkers (40%, 8 out of 20). Respondents drinking beer partying at home or pool/beach area

consisted of light drinkers at 72.73% (8 out of 11). Other situations were mostly comprised of

moderate drinkers (42.86%, 6 out of 14) that consumed their favorite beer in these situations.

These results were found to be statistically significant (p < .05).

In group two with four situations, the first three situations (bar/club with friends, at home

with friends not a party, and at a party not at home with friends) are the exact same as in group

one. However, the other sections were added with the 4th and 5th situation from group one.

Therefore, in other situations light drinkers at 48% (12 out of 25) consumed their favorite beer.

However, unlike group the first grouping, these results were not statistically significant.

Therefore, the situations cannot be associated with consumption behaviors.

Group three contains three situations, and college students drinking a bar or club were

mostly heavy drinkers (42.24%, 16 out of 37) and moderate drinkers (40.54%, 15 out of 37).

Respondents that drank their favorite beer in a restaurant/home not partying were 50% light

drinkers (14 out of 28). At a party/special event the light drinkers were comprised of 43.75% (14

out of 32). These results were statistically significant (p < .05), and the situations have

associations with the drinking behaviors during them.









Group four has two situations. In a party, bar, or club moderate drinkers at 40% (24 out of

60) and heavy drinkers at 36.67% (22 out of 60) drank their preferred beer. The other group,

drinking beer but not in a party, bar, or club consisted mainly of light drinkers at 54.05% (20 out

of 37). This data was considered to be significant (p < .05). Therefore, there is an association

with these drinking behaviors and the two situations.

Group five contains four situations, college students drinking in a social environment were

distributed moderately even among consumption behaviors: light drinkers at 29.69% (19 out of

65), moderate drinkers at 38.46% (25 out of 65), and heavy drinkers at 32.31% (21 out of 65).

College students that drank with a date were mostly heavy drinkers with 60% (3 out of 5).

Students that drank alone were mostly light drinkers 66.67% (2 out of 3). Respondents that drank

their favorite beer relaxing with friends were light drinkers (47.62%, 10 out of 21). Group five

data was not statistically significant so no associations can be made.

Group six had three situations. These situations come directly from group five. However,

the relax with friends situation was omitted form this grouping. Situations in group six were as

follows: social, date, and alone. This information had the same results as in group five. In

addition, this group was also not statistically significant.

Chi-square test results for all situational groupings were as follows: group one (x2 = 15.50,

p < .05), two (x2 = 10.22, p > .05), three (x2 = 10.44, p < .05), four (x2 = 9.67, p < .05), five (x2 =

6.28, p > .05), and six (x2 = 4.07, p > .05). Groups one, three, and five were significant (p < .05).

Therefore, associations can be made from these situations and consumption behaviors. College

students that drank in bars, clubs, and party situation had heavy and moderate drinking

behaviors. Students that just stayed with friends not in a party situation were light drinkers.

Situations including special events and restaurants exhibited light drinking behaviors as well in









college students. When respondents tended to go to parties not in their home, college students

drank in light to moderate drinking behaviors.

Table 4-2 Situational Groups and Consumption Behavior
Situational Items Included in Light Moderate Heavy Total
Grouping Grouping Drinker Drinker Drinker


Bar or club with friends


6
(17.14%)


14
(40%)


15
(42.86%)


At home with friends (not 9 4 4 17
a party) (52.94%) (23.53%) (23.53%)

Party (not at home) with 7 8 5 20
friends (35%) (40%) (25%)

At home alone (not a 8 2 1 11
party) (72.73%) (18.18%) (9.09%)

Beach/Pool with friends
(not a party)

Party at your home


In a bar or club with a
date

In a restaurant with
friends

Other

In a bar or club alone

At home with a date (not a
party)

Sporting event/concert
with friends

Party (not at home) with
date


4
(28.57%)


6
(42.86%)


4
(28.57%)









Table 4-2 Continued
Situational Items Included in Light Moderate Heavy Total
Grouping Grouping Drinker Drinker Drinker

2 Bar or club with friends 6 14 15 35
(17.14%) (40%) (42.86%)
At home with friends (not 9 4 4 17
a party) (52.94%) (23.53%) (23.53%)

Party (not at home) with 7 8 5 20
friends (35%) (40%) (25%)

At home alone (not a 12 8 5 25
party) (48%) (32%) (20%)

Beach/Pool with friends
(not a party)

Party at your home

In a bar or club with a
date

In a restaurant with
friends

Other

In a bar or club alone

At home with a date (not a
party)

Sporting event/concert
with friends

Party (not at home) with
date









Table 4-2 Continued
Situational Items Included in Light Moderate Heavy Total
Grouping Grouping Drinker Drinker Drinker


In a bar or club with
friends

In a bar or club with a
date

In a bar or club alone

In a restaurant with
friends

At home with friends (not
party)

At home with a date (not a
party)

At home alone (not a
party)

Other

Sporting event/concert
with friends

Beach/pool with friends
(not a party)

Party (not at home) with
friends

Party (not at home) with a
date

Party at your home


6
(16.22%)


14
(50%)


14
(43.75%)


15
(40.54%)


8
(28.57%)


11
(34.38%)


16
(43.24%)


6
(21.43%)


7
(21.88%)









Table 4-2 Continued
Situational Items Included in Light Moderate Heavy Total
Grouping Grouping Drinker Drinker Drinker

4* In a bar or club 14 24 22 60
with friends (23.33%) (40%) (36.67%)

In a bar or club
with a date

In a bar or club
alone

Party (not at
home) with friends

Party (not at
home) with a date

Party at your home
In a restaurant 20 10 7 37
with friends (54.05%) (27.03%) (18.92%)

At home with
friends (not party)

At home with a
date (not a party)

At home alone
(not a party)

Sporting
event/concert with
friends

Beach/pool with
friends (not a
party)

Other









Table 4-2 Continued
Situational Items Included in Light Moderate Heavy Total
Grouping Grouping Drinker Drinker Drinker


In a bar or club
with friends

Sporting
event/concert with
friends

Beach/pool with
friends (not a
party)

Party (not at
home) with friends

Party at your home


19
(29.23%)


25
(38.46%)


21
(32.31%)


In a bar or club 1 1 3 5
with a date (20%) (20%) (60%)

At home with a
date (not a party)

Party (not at
home) with a date

At a bar or club 2 1 0 3
alone (66.67%) (33.33%) (0%)

At home alone
(not a party)


In a restaurant
with friends

At home with
friends (not a
party)


10
(47.62%)


6
(28.57%)


5
(23.81%)









Table 4-2 Continued
Situational Items Included in Light Moderate Heavy Total
Grouping Grouping Drinker Drinker Drinker

6 In a bar or club 19 25 21 65
with friends (29.23%) (38.46%) (32.31%)

Sporting
event/concert
with friends

Beach/pool with
friends (not a
party)

Party (not at
home) with
friends

Party at your
home

In a bar or club 1 1 3 5
with a date (20%) (20%) (60%)

At home with a
date (not a party)

Party (not at
home) with a date

At a bar or club 2 1 0 3
alone (66.67%) (33.33%) (0%)

At home alone
(not a party)

1: x2 = 15.50, df = 8, N = 97, p-value =.05*
2: x2 = 10.22, df = 6, N = 97, p-value = .12
3: x2 = 10.44, df = 4, N = 97, p-value = .03*
4: x2 = 9.67, df = 2, N = 97, p-value = .01*
5: x2 = 6.28, df = 6, N = 94, p-value = .39
6: x2 = 4.07, df = 4, N = 73, p-value = .40
*p-value < .05
(%) Percentage of situation for a specific drinking behavior









Variables Affecting Beer Brand Choice

A multiple discriminant analysis was used in order to determine a linear combination of

continuously measured IVs that best classify cases in one of several known groups. The

continuous independent variables used in this analysis were brand benefits (quality, price,

emotion, social, environment, and health), exploratory shopping behaviors (risk-taking, variety-

seeking, and curiosity-seeking), susceptibility to interpersonal influence, product category

involvement, gender, employment, Greek membership, and age. The categorical dependent

variable was brand choice among the three brands of Bud Light, Blue Moon, and Corona. In

addition, Table 4-3 summarizes the results of multiple discriminant analysis. This table includes

unstandarized and standardized coefficients for both functions, Wilks' Lambda, F-ratio, p-value,

for each brand, centroids for both group functions, eigenvalues, and the Wilks' Lambda and Chi

Square for canonical discriminant functions.

In this analysis, two functions were produced. The first function produced a high degree of

separation among the second function based on the final Wilks' lambda (.48) and canonical

correlation (.64). Therefore, there is a moderate to strong correlation between the discriminant

function and the independent variables. In addition, 73.8% of variance was explained by the first

function. The first function was found to be statistically significant (p = .00, x2 = 63.90, df = 30).

The second function had a Wilks' lambda of .80 and a canonical correlation of .44. The second

function was not significant (p = .17).

Standardized canonical discriminant function coefficients for the two functions were

included in Table 4-3. A structure matrix, listed in Table 4-4, was used to discover the largest

absolute correlation between variables and discriminant functions. This structure matrix indicates

the greatest impact on the discriminatory impact on the predictor variables on beer band choice.

According to the structure matrix, in function one, risk-taking behavior, social, and age were the









highest three variables correlated with discriminant scores. In function two, environment, gender,

and price were the highest three variables correlated with discriminant scores.

The average discriminant scores for the function 1 centroidss) with respect to brands were

the following: Bud Light = -.89, Blue Moon = 1.02 and Corona = .38. This means that in the 1st

function, while significant will do a good job discriminating between all three beer brands. The

most discriminating power existed between Bud Light and Blue Moon. Blue Moon and Corona

were more closely related than Bud Light and the other two brands.

Table 4-6 illustrates the mean and standard deviation for brand choice variables and the

independent variables used in the discriminant analysis. Variables that had statistical significance

with Wilks' Lambda were the following: price (F-ratio = 6.64, p-value = .00), risk-taking

behavior (F-ratio = 4.08, p-value = .02), and environment (F-ratio =3.63, p-value = .03). The

mean scores price according to brand choice were the following: Bud Light = 5.23 (standard

deviation = .92), Blue Moon = 4.74 (standard deviation = .82), and Corona= 4.38 (standard

deviation = 1.13). Mean scores for risk-taking behaviors were the following: Bud Light = 3.70

(standard deviation = 1.26), Blue Moon = 4.58 (standard deviation = 1.32), and Corona = 3.98

(standard deviation = 1.13). Mean scores for environment were the following: Bud Light = 3.94

(standard deviation = .95), Blue Moon = 4.27 (standard deviation = .61), and Corona = 3.65

(standard deviation = .84). The most significant variable affecting Bud Light was price with a

mean of 4.23. Price was also the most significant variable with regards to the other two brands of

Blue Moon (mean = 4.74) and Corona (mean = 4.38).

The three most important variables in terms of individual variables for Wilks' Lambda are

price, risk-taking behavior, and environment. These three variables had the smallest Wilks'

Lambda values of .88 for price, .92 for risk-taking behavior, and .92 for environment. In









addition, they have the highest F-ratios among the other variables, which are statistically

significant (p-value <= .05). Based on the interaction of the independent variables, price is the

most important variable in the first function (standardized coefficient = -.89) in discriminating

brand choice in terms of the standardized coefficients. The second most important variable with

interaction for the first function is social (standardized coefficient = -.69). However, the social

benefit variable was not found to be statistically significant (p-value = .06). The third most

important variable (standardized coefficient = .49) based on interaction of variables was risk-

taking behavior which was found statistically significant (p-value = .02). Based on these results,

college students that like the price of their favorite beer are less likely to change beer brands. In

addition, if a college student exhibits risk-taking behaviors, then they are more likely to change

beer brands. Finally, when a college student is in favorite of the environmental features of a beer

brand, then the individual is more likely to change beer brand choice.

Predictive results of this analysis based on the multiple discriminant analysis are illustrated

in Table 4-5. According the results, 66.70% of original grouped cases were correctly classified.

According to these results, 83.33% of Bud Light, 57.69% of Blue Moon, and 50.00% of Corona

were correctly classified. The largest prediction errors occurred in the Corona brand (12 at

42.86%). The overall hit ratio had a t-ratio of 6.44 and thus the resultant hit rate was statistically

significant (8 = .05).

Additional analysis was used in order to determine the most significant contributions to

beer brand choice. A stepwise discriminant analysis was used in order to determine the greatest

linear combination of continuously measured independent variables that best classify cases in

one of several known groups. This analyze was used in order to limit the assumption that

predictor variables are independent or noncollinear. In addition, Table 4-7 summarizes the results









of stepwise discriminant analysis. This table includes unstandarized and standardized

coefficients for both functions, Wilks' Lambda, F-ratio, eigenvalues, p-value, for each brand,

centroids for both group functions, and the Wilks' Lambda and Chi Square for canonical

discriminant functions. In this analysis, two of the fifteen variables were selected when Wilks'

lambda was at the lowest level. The first function produced a high degree of separation among

the second function based on the final Wilks' lambda (.77) and canonical correlation (.44). The

second function had a Wilks' lambda of .96 and a canonical correlation of .21. Both functions

were found to be statistically significant (p < .05).

Based on group centroids for group one, Bud Light = .54, Blue Moon = -.44, and Corona =

-.41, the first function discriminates Bud Light from the other two brands: Blue Moon and Bud

Light. The second function discriminates between all three brands with the group two centroids

of Bud Light at .01, Blue Moon at .28, and Corona at -.27. However, the degree of separation

among the brands in function two does not discriminate as strongly as in function one. In Table

4-7, the two-plot function of group centroids demonstrates this separation among Bud Light, and

the two other brands, Blue Moon and Corona.

The canonical correlation (R) for the first function is .44. This means that between the

independent variables included in the analysis, there is a moderate correlation between the

discriminant function and the independent variables. The Wilks' Lambda for the canonical

discriminant function is .77. This means that 77% of discriminant scores for variance are not

explained by group differences. The Wilks' Lambda (.77) and Chi-Square test (23.76) were

found to be statistically significant. The Chi-Square test has a p-value of .00 with 4 degrees of

freedom. Of the variance explained by the two functions, the 1st explains 84.3%.









The canonical correlation (R) for the second function is .21. This means that between the

independent variables included in the analysis, there is a low correlation between the

discriminant function and the independent variables. The Wilks' Lambda for the canonical

discriminant function (2) is .96. This means that 96% of discriminant scores for variance are not

explained by group differences. The Wilks' Lambda (.96) and Chi-Square test (4.01) were found

to be statistically significant. The Chi-Square test has a p-value of .05 with 1 degree of freedom.

Of the predictor variables in the second function, explains 15.7% was explained.

Standardized canonical discriminant function coefficients for the two functions were

included in Table 4-7. The two most important variables in terms of individual variables for

Wilks' Lambda are price and risk-taking behaviors. These two variables had the smallest Wilks'

Lambda values of .88 for price and .92 for risk-taking behavior. This means that 87.5% of price

was not explained by group differences and that 91.9% of risk-taking behavior was not explained

by group differences as well. In addition, they have the highest F-ratios among the other

variables, which are statistically significant (p-value <= .05). Therefore, price and risk-taking

behaviors were analyzed in the canonical discriminant functions. Based on the interaction of the

variables price and risk-taking behavior, price is the most important variable in the first function

(standardized coefficient = .92) in discriminating the levels of the dependent variable in terms of

the standardized coefficients. The second most important variable with interaction for the first

function is risk-taking behavior (standardized coefficient = -.73). In the second function, risk-

taking behavior is the most important function (standardized coefficient = .74). The second most

important variable for this second function is price (standardized coefficient = .49). Based on the

results from this stepwise analysis, the first function indicates that college students that were









positively affected by pricing had a negative impact on risk-taking behavior. Therefore, as

pricing of a beer brand goes down, risk-taking behavior goes up.

Predictive results of this analysis based on the two predictor variables, illustrated in Table

4-8, indicated that 54.60% of original grouped cases were correctly classified. According to these

results, 78.57% of Bud Light, 37.04% of Blue Moon, and 35.71% of Corona were correctly

classified. The largest prediction errors occurred among Blue Moon (10 at 37.04%) and Corona

(13 at 46.43%) groups. There is some ambiguity in determining whet her Blue Moon drinker

should be predicted as a member of that group or of the Corona group. The overall hit ratio was

4.17. This was found to be statistically significant (8 = .05).









Table 4-3 Multiple Discriminant Analysis: Beer Brand Choice
Function 1 Function 2 Wilks' F-ratio p-value*
Unstandardized Standardized Unstandardized Standardized Lambda
Discriminant Coefficients Discriminant Coefficients
Coefficients Coefficients
(Constant) -5.59 2.92

Brand Quality .17 .18 .05 .05 .98 .97 .38
Benefits Price -.92 -.89 -.45 -.43 .88 6.64 .00
Social -.52 -.69 -.10 -.13 .94 2.92 .06
Emotion .03 .03 .48 .56 .99 .27 .76
Environment .50 .42 -.45 -.38 .93 3.63 .03
Health .09 .10 -.04 -.04 .98 1.11 .33
Exploratory Risking .39 .49 -.20 -.27 .92 4.08 .02
Shopping Taking
Behaviors Curiosity .45 .45 -.07 -.09 .95 2.68 .07
Seeking
Variety -.18 -.25 -.20 -.20 .99 .25 .78
Seeking
Consumer Interpersonal .33 .32 .38 .37 .99 .30 .74
Factors Influence
Product .11 .18 -.29 -.48 .97 1.50 .23
Category
Involvement
Gender .13 .05 1.67 .71 .94 2.88 .06
Employment .19 .10 .05 .02 .98 .82 .45
Greek .48 .22 .95 .44 .98 1.19 .31
Membership
Age .17 .37 -.06 -.14 .94 2.79 .07









Table 4-3 Continued
Function Eigenvalue % of Variance Cumulative % Canonical Correlation
1 .69 73.8 73.8 .64
2 .24 26.2 100.0 .44

After Function Wilks' Lambda Chi-square df Significance*
1 through 2 .48 63.90 30 .00
2 .80 18.80 14 .17

Beer Brand Choice Function 1 Function 2

Bud Light -.89 -.16
Blue Moon 1.02 -.52
Corona .38 .72
*Significance < .05
tF-ratio is bolded for significance (p-value<=.05)










Function 1 Function 2

Risk Taking .33* -.23
Social -.29* -.12
Age .28* -.14
Curiosity Seeking .27* -.18
Employment -.16* .04
Variety Seeking -.09* .04
Environment .11 -.54*
Gender -.06 .49*
Price -.36 -.48*
Health .00 -.31*
Quality .07 -.27*
Product Category .15 -.26*
Involvement
Greek Membership .16 .18*
Emotion .04 -.14*
Susceptibility to -.08 .09*
Interpersonal Influence
*Largest absolute correlation between the variable and any discriminant function.

Table 4-5 Group Classification for Beer Brand Choice: Multiple Discriminant Analysis
Brand Choice Predicted Group Membership Number of Cases
Bud Light Blue Moon Corona
Bud Light 35 (83.33%) 2 (4.76%) 5 (11.90%) 42
Blue Moon 4(15.38%) 15 (57.69%) 7(26.92%) 26
Corona 9 (32.14%) 5 (17.86%) 14 (50.00%) 28
*Percentage of group cases classified: 66.70%


Table 4-4


Stnlctllre Matrix for Beer Brand Ch oi ce









Table 4-6 Mean and Standard Deviation based on Brand Choice
Bud Light Blue Moon Corona
Mean Standard Mean Standard Mean Standard
Deviation Deviation Deviation
Brand Quality 5.42 .80 5.66 1.08 5.26 1.38
Benefits Index*
Price Index* 5.23 .92 4.74 .82 4.38 1.13
Social Index* 3.42 1.32 2.74 1.44 2.79 1.23
Emotion 4.47 1.18 4.62 1.11 4.39 1.15
Index*
Environment 3.94 .95 4.27 .61 3.65 .84
Index*
Health 2.37 1.07 2.50 1.26 2.07 .96
Index*
Exploratory Risking 3.70 1.26 4.58 1.32 3.96 1.13
Shopping Taking
Behaviors Index*
Curiosity 4.62 .96 5.19 1.04 4.80 1.00
Seeking
Index *
Variety 4.52 1.48 4.28 1.15 4.42 1.45
Seeking
Index*
Consumer Interpersonal 3.56 1.06 3.38 .88 3.54 .93
Factors Influence
Index*
Product 2.90 1.58 3.54 1.86 2.86 1.53
Involvement
Index*
GenderT .74 .45 .62 .50 .89 .31
Employment' .62 .49 .46 .51 .54 .51
Greek .62 .49 .73 .45 .79 .42
Membership
Age 20.14 1.75 21.46 2.89 20.68 2.21
*Measured on a 7-point Likert scale
tVariables were coded as bivariates (0 and 1)









Table 4-7 Stepwise Discriminant Analysis: Beer Brand Choice
Function 1 Function 2 Wilks' F-ratioT p-value*
Unstandardized Standardized Unstandardized Standardized Lambda
Discriminant Coefficients Discriminant Coefficients
Coefficients Coefficients
(Constant) -2.26 -4.88
Quality .98 .97 .38
Price .95 .92 .51 .49 .88 6.64 .00
Social .94 2.92 .06
Emotional .99 .27 .76
Environmental .93 3.63 .03
Health .98 1.11 .33
Risking Taking -.59 -.73 .60 .74 .92 4.08 .02
Curiosity .95 2.68 .07
Seeking
Variety .99 .25 .78
Seeking
Interpersonal .99 .30 .74
Influence
Product .97 1.50 .23
Category
Involvement"
Gender' .94 2.88 .06
Employment" .98 .82 .45
Greek .98 1.19 .31
Membership
Agea .94 2.79 .07

Function Eigenvalue % of Variance Cumulative Canonical Correlation
1 .24 84.3% 84.3% .44
2 .04 15.7% 100.0% .21












Table 4-7 Continued
After Function Wilks Lambda Chi-squared D.F. Significance*
1 through 2 .77 23.76 4 .00
2 .96 4.01 1 .05


Function 1 Function 2


Bud Light .54 .01
Blue Moon -.44 .28
Corona -.41 -.27


Canonical Discriminant Functions


1-

C

U-
-


NewBiandChoice
0 Bud Light
0 Blue Moon
Corona
* Gioup Centioid


-4 -3 -2 -1 0
Function 1


1 2 3


tF-ratio is bolded for significance (p-value<=.05); Test of Equality of Group Means: df for all variables are (2, 93), N = 96
aThese variables were not used in analysis for unstandardized and standardized coefficients.
*Significance < .05


0

Son
o 0 0


0
o o
Blue Mhn 0 o 0
cE o mud Ligl (3
0 Corona

o o o
0 0
0
0


'' '' ''









Table 4-8 Group Classification: Beer Brand Choice Stepwise Discriminant Analysis
Brand Choice Predicted Group Membership Number of Cases
Bud Light Blue Moon Corona
Bud Light 33 (78.57%) 4 (9.52%) 5 (11.90%) 42
Blue Moon 9(33.33%) 10 (37.04%) 8(29.63%) 27
Corona 12(42.86%) 6(21.43%) 10 (35.71%) 28
*Percentage of group cases classified: 54.6%

Variables Affecting Beer Consumption

A multiple discriminant analysis was used in determining linear relationship of the

independent variables on the dependent variable, beer consumption behavior. The independent

variables were all continuous variables, and the dependent variable was categorical (light beer

drinkers, moderate beer drinkers, and heavy beer drinkers). Table 4-9 illustrates the results of the

multiple discriminant analysis containing unstandardized and standardized coefficients, Wilks'

Lambda, F-ratio, eigenvalues, and p-value, for each consumption behavior, centroids for the

function, and the Wilks' Lambda and Chi-square for the canonical discriminant function.

There were two functions produced from the multiple discriminant analysis. In function

one, the Wilks' lambda was .60 and the canonical correlation was .58. Therefore, there is a

moderate to strong correlation between the discriminant function and the independent variables.

This function had 82.3% of variance explained by this function. The first function was found to

be statistically significant (p-value = .04, x2 = 44.47, df = 30). The second function had a Wilks'

lambda of .90 and canonical correlation of .31. This function was not found to be statistically

significant (p-value = .83).

The structure matrix for this analysis is listed in Table 4-10. This table illustrates the

largest absolute correlation between the independent variables and the discriminant functions.

For the first function, the top three variables correlated with beer consumption behavior were









emotion, product category involvement, and quality. For the second function, the top three

variables were risk-taking behaviors, variety-seeking behaviors, and curiosity-seeking behaviors.

The discriminant scores for the first function centroidss) with respect to beer consumption

levels were the following: light beer drinkers = -.89, moderate beer drinkers = .26 and heavy beer

drinkers = .80. This means that the function does a good job of discriminating between all three

different levels of consumption behaviors. The most discriminating power among the first

function was between light beer drinkers and heavy beer drinkers. Moderate and heavy beer

drinkers were more closely related than light beer drinkers.

The mean and standard deviations for independent and dependent variables are located on

Table 4-12. Independent variables that had statistical significance with Wilks' Lambda were the

following: quality (F-ratio = 4.20, p-value = .02), emotion (F-ratio = 11.80, p-value = .00), health

(F-ratio = 3.04, p-value = .05), and product category involvement (F-ratio = 9.15, p-value = .00).

The mean scores for quality according to consumption behaviors were the following: light beer

drinkers = 5.06 (standard deviation = 1.27), moderate beer drinkers = 5.52 (standard deviation =

.94), and heavy beer drinkers = 5.81 (standard deviation = .77). Mean scores for emotion

according to consumption behavior were the following: light beer drinkers = 3.82 (standard

deviation = 1.24), moderate beer drinkers = 4.75 (standard deviation = .96), and heavy beer

drinkers = 5.00 (standard deviation = .80). Mean scores for health with consumption behavior

were light beer drinkers = 1.97 (standard deviation = 1.03), moderate beer drinkers 2.44

(standard deviation = 1.18), and heavy beer drinkers = 2.61 (standard deviation = 1.00). The

mean scores for product category involvement for consumption behaviors were the following:

light beer drinkers = 2.26 (standard deviation = 1.35), moderate beer drinkers = 3.21 (standard









deviation = 1.69), and heavy beer drinkers = 3.89 (standard deviation = 1.54). The most

significant and highest variable affecting light, moderate, and heavy beer drinkers was quality.

The four most important variables in terms of individual variables for Wilks' Lambda

were emotion, product category involvement, quality, and health. These four variables had the

smallest Wilks' Lambda values of .80 for emotion, .84 for product category involvement, .92 for

quality, and .94 for health. In addition, these four variables have the highest F-ratios among all

other variables and are statistically significant (p-value <= .05). Therefore, emotion

(standardized coefficient = .56), product category involvement (standardized coefficient = .49),

quality (standardized coefficient = -.06), and health (standardized coefficient = .16) were the

most important factors in discriminating consumption behavior among all other independent

variables.

Based on these results, quality was an important factor in the consumption of beer all on

levels of consumption. This variable was negatively related to beer consumption behavior.

Therefore, as quality decreases, then beer consumption increases and vice versa. Emotion was

positively related with beer consumption behaviors. Therefore, the more passionate a respondent

was about their favorite brand of beer, the more beer they were likely to consume more. Product

category involvement also had a positive relationship meaning that as product category

involvement increases, then consumption behavior increases. Finally, health had a low,

positively related impact on consumption behaviors.

On Table 4-11, the classification matrix for beer consumption behavior and demonstrates

that there were 77.14% correctly classified cases for light beer drinkers. There were 51.52%

correctly classified cases for moderate beer drinkers and 57.14% correctly classified for heavy

beer drinkers. The largest prediction errors occurred in the moderate beer drinker group. There









were 16 cases incorrectly classified (48.48% errors). Therefore, 62.50% of the classification

matrix was correctly classified. The value of the hit rate was 5.29 (8 = .05) and was statistically

significant.

Further analysis was used in determining which independent variables have the greatest

impact on the dependent variable, beer consumption behavior. This analyze also followed up in

order to clear up problems involving independence and noncollinear predictor variables used

from a multiple discriminant analysis. Table 4-13 illustrates the results of the stepwise

discriminant analysis containing unstandardized and standardized coefficients, Wilks' Lambda,

F-ratio, eigenvalues, and p-value, for each consumption behavior, centroids for the function, and

the Wilks' Lambda and Chi-square for the canonical discriminant function. One variable out of

the fifteen variables was selected when the Wilks' lambda (.80) was the lowest and the canonical

correlation was .45. There was one function discriminated between all three beer drinking levels,

light, moderate, and heavy. The function was found to be statistically significant (p < .05).

The canonical correlation (R) for the function is .45. This means that between the

independent variables included in the analysis, there is a moderate correlation between the

discriminant function and the independent variables. The Wilks' Lambda for the canonical

discriminant function is .80. This means that 80% of the discriminant scores for variance are not

explained by group differences. The Wilks' Lambda (.80) and Chi-Square test (21.04) were

found to be statistically significant. The Chi-Square test has a p-value of .00 with 2 degrees of

freedom. This means that there can be inferences made from the data onto the general population

based on these results. The average discriminant scores for the function centroidss) with respect

to consumption levels were the following: light beer drinkers = -.64, moderate beer drinkers =









.26 and heavy beer drinkers = .50. This means that the function does a good job of discriminating

between all three different levels of consumption behaviors.

The two most important variables in terms of individual variables for Wilks' Lambda

were emotion and product category involvement. These two variables had the smallest Wilks'

Lambda values of .80 for emotion and .84 for product category involvement. This means that

79.8% of emotion was not explained by group differences and that 83.6% of product category

involvement was not explained by group differences. In addition, these two variables have the

highest F-ratios among all other variables and are statistically significant (p-value <= .05).

However, based on the stepwise discriminant analysis, all independent variables except emotion

were removed for analysis. In the discriminant function, emotion is the most important factor

(standardized coefficient = 1.00) and was the only independent variable analyzed. This

relationship was positive as in the multiple discriminant analysis. Therefore, the more a college

student feels emotionally toward a brand, then the more likely they to exhibit increased beer

consumption behavior.

On Table 4-14, the classification matrix demonstrates that there were 57.14% correctly

classified cases for light drinkers. There were 52.94% correctly classified cases for moderate

drinkers and 24.14% correctly classified for heavy drinkers. The largest prediction errors

occurred in the heavy drinker group. There were 53 cases incorrectly classified (54.1% errors).

Therefore, 45.9% of the classification matrix was correctly classified. The value of the hit rate

was 2.71 (8 = .05). This hit rate was statistically significant.









Table 4-9 Multiple Discriminant Analysis: Beer Consumption Behavior
Function 1 Function 2 Wilks' F-ratio p-value*
Unstandardized Standardized Unstandardized Standardized Lambda
Discriminant Coefficients Discriminant Coefficients
Coefficients Coefficients
Constant) -4.81 -5.85
Brand Quality -.06 -.06 -.21 -.22 .92 4.20 .02
Benefits Price .10 .10 -.27 -.27 .96 2.15 .12
Social -.02 -.03 .07 .09 .98 1.12 .33
Emotion .55 .56 .18 .18 .80 11.80 .00
Environment -.36 -.31 -.01 -.01 .97 1.23 .30
Health .15 .16 .04 .04 .94 3.04 .05
Exploratory Risking -.13 -.16 .35 .48 .97 1.57 .21
Shopping Taking
Behaviors Curiosity .30 .31 .37 .47 .97 1.50 .23
Seeking
Variety -.10 -.14 .33 .33 .97 1.64 .20
Seeking
Consumer Interpersonal .15 .15 .25 .25 .99 .56 .57
Factors Influence
Product .32 .49 -.10 -.15 .84 9.15 .00
Category
Involvement
Gender -.90 -.39 -.88 -.38 .98 .99 .38
Employment .18 .09 .86 .43 .98 .91 .41
Greek -.98 -.46 .18 .08 .99 .71 .49
Membership
Age .12 .28 .10 .23 .97 1.68 .19

Function Eigenvalue % of Variance Cumulative % Canonical Correlation
1 .51 82.3 82.3 .58
2 .11 17.7 100.0 .31









Table 4-9 Continued
After Function Wilks' Lambda Chi-square df Significance*
1 through 2 .60 44.47 30 .04
2 .90 8.96 14 .83

Beer Consumption Behavior Function 1 Function 2

Light Beer Drinkers -.89 -.13
Moderate Beer Drinkers .26 .43
Heavy Beer Drinkers .80 -.35
F-ratio is bolded for significance (p-value<=.05)
*Significance < .05









Structure Matrix for Beer Consumption Behavior


Function
1 2
Emotion .70* .18
Product Category Involvement .62* -.15
Quality .42* -.08
Health .36* .05
Price .30* -.08
Age .26* -.06
Gender -.20* .00
Greek -.17* -.05
Risk-Taking -.01 .56*
Variety-Seeking -.12 .50*
Curiosity-Seeking .14 .45*
Employment -.04 .41*
Social .14 .36*
Environment .19 .28
Susceptibility to Interpersonal .08 .28
Influence

Table 4-11 Group Classification: Beer Consumption Behavior Multiple Discriminant Analysis
Consumption Behavior Predicted Group Membership Number of
Light Beer Moderate Heavy Beer Cases
Drinker Beer Drinker Drinker
Light Beer Drinker 27 (77.14%) 7 (20.00%) 1 (2.86%) 35
Moderate Beer Drinker 12 (36.36%) 17(51.52%) 4 (12.12%) 33
Heavy Beer Drinker 3(10.71%) 9(32.14%) 16 (57.14%) 28
*Percentage of group cases classified: 62.5%


Table 4-10









Mean and Standard Devi r


Light Beer Drinker Moderate Beer Heavy Beer Drinker
Drinker
Mean Standard Mean Standard Mean Standard
Deviation Deviation Deviation
Brand Quality 5.06 1.27 5.52 .94 5.81 .77
Benefits Index*
Price Index* 4.59 1.23 4.89 .80 5.12 .91
Social Index* 2.82 1.24 3.31 1.49 3.04 1.31
Emotion 3.82 1.24 4.75 .96 5.00 .80
Index*
Environment 3.77 .88 4.09 .80 3.99 .90
Index*
Health 1.97 1.03 2.44 1.18 2.61 1.00
Index*
Exploratory Risking 3.93 1.32 4.32 1.32 3.76 1.16
Shopping Taking
Behaviors Index*
Curiosity 4.64 1.07 5.06 .98 4.79 .95
Seeking
Index *
Variety 4.49 1.42 4.68 1.40 4.05 1.28
Seeking
Index*
Consumer Interpersonal 3.40 .90 3.65 .96 3.48 1.08
Factors Influence
Index*
Product 2.26 1.35 3.21 1.69 3.89 1.54
Involvement
Index*
GenderT .83 .38 .73 .45 .68 .48
Employment' .54 .51 .64 .49 .46 .51
Greek .77 .43 .67 .48 .64 .49
Membership
Age 20.14 1.42 20.76 2.45 21.18 2.83
*Measured on a 7-point Likert scale
tVariables were coded as bivariates (0 and 1).


Table 4-12









Table 4-13 Stepwise Discriminant Analysis: Beer Consumption Behavior
Function 1 Wilks' F-ratio p-value*
Unstandardized Standardized Lambda
Discriminant Coefficients
Coefficients
(Constant) -4.35
Quality .92 4.20 .02
Prices .96 2.15 .12
Social .98 1.12 .33
Emotion .97 1.00 .80 11.80 .00
Environmental .97 1.23 .30
Health .94 3.04 .05
Risking Taking' .97 1.57 .21
Curiosity Seekinga .97 1.50 .23
Variety Seekinga .97 1.64 .20
Susceptibility to .99 .56 .57
Interpersonal
Influence
Product Category .84 9.15 .00
Involvement
Gender .98 .99 .38
Employment .98 .91 .41
Greek Membershipa .99 .71 .49
Agea .97 1.68 .19

Function Eigenvalue % of Variance Cumulative Canonical
Correlation
1 ..25 100.0% 100.0% .45

After Function Wilks Chi-squared D.F. Significance
Lambda
1 .80 21.04 2 .00

Beer Consumption Behaviors Function 1
Light Beer Drinkers -.64
Moderate Beer Drinkers .26
Heavy Beer Drinkers .50
F-ratio is bolded for significance (p-value<=.05); Test of Equality of Group Means: df for all
variables are (2, 93), N = 96
aThese variables were not used in analysis for unstandardized and standardized coefficients.
*Significance < .05









Table 4-14 Group Classification: Beer Consumption Behavior Stepwise Discriminant Analysis
Beer Brand Choice Predicted Group Membership Number
Light Drinker Moderate Drinker Heavy Drinker of Cases
Light Beer Drinker 20 (57.14%) 15 (42.86%) 0 (0%) 35
Moderate Beer Drinker 9 (26.47%) 18 (52.94%) 7 (20.59%) 34
Heavy Beer Drinker 6 (20.69%) 16 (55.17%) 7 (24.14%) 29
*Percentage of group cases classified: 45.9%









CHATPER 5
DISCUSSION

The three brands used in this study were excellent brands to analyze based on the

differences in their positioning and marketing strategies. Bud Light is positioned as a convenient,

light beer that focuses on fun and great taste. Bud Light is a brand extension from the popular

brand, Budweiser. Bud Light utilizes massive marketing efforts (television, radio, internet,

outdoor, sponsorships, print, etc.) by focusing humor and popular culture references used by

their consumers. This brand has the most market share among the three brands, and based on pre-

test and main study results, Bud Light was the most popular brand among all other brands

investigated.

The positioning for Blue Moon is a premium, craft beer. The marketing for Blue Moon

focuses on word of mouth and point of sale. Most consumers associate the brand with the slice of

orange that is traditionally placed on the side of the glass when served. This is a key

differentiation of the brand among other beers. Consumers know that a pint of beer with an

orange slide garnished on the side of the glass is Blue Moon. The brand is not highly marketed

and does not have a presence in most marketing media. This strategy appeals to their consumers

that love premium, craft beers.

The positioning for Corona is escapism. Corona has traditionally been known as an import

brand from Mexico, and the brand has moved into the idea of escaping to an exotic destination.

The brand tries to completely differentiate itself from all other beers. The brand competes with

premium beers based on its import status in the U.S. It has something similar with Blue Moon

based on its garnish, a sliced lime pushed through its clear bottle. The marketing efforts focus on

this idea of escaping to a relaxing place, whether it is a beach or just calm ocean waters. Corona









creates a mood with their branding. They focus heavily on television and print advertising for

their marketing efforts.

All three brands are completely different in their marketing and positioning strategies. The

prices of a six pack of twelve ounce bottles for all three brands from a local retailer were the

following: Bud Light = $4.54, Blue Moon = $6.97, and Corona = $6.97. Therefore, based on

using these three brands, there were different segments of consumers within the sample with

different marketing strategies for each. Therefore, these three brands provided variance in their

positioning and marketing strategy among the college market.

Results from this study demonstrated variables of significance for both beer brand choice

and beer consumption behavior. These variables were quality, price, emotion, environment,

health, risk-taking behavior, and product category involvement. However, other variables were

not found to be significant. Subjective norm variables such as social and interpersonal influence

may not have been deemed significant because this truly may not be a variable that influenced

this market. It could also be possible that this market has difficulty answering questions in this

area. College students did not seem to be aided in a beer consumption or beer brand choice

decision based on social/interpersonal factors. Previous research studies have found that there

were social and interpersonal factors influence drinking alcohol (Wall et al. 1998, July;

Trafimow, 1996; O'Callaghan et al. 1997, September). However, this study concluded that they

were not found to be influential in beer consumption behavior and brand choice. In addition,

retail marketing could potentially have very little effect on this market based on these results.

The exploratory shopping behaviors of curiosity-seeking and variety-seeking did not have

significant results as well. Variety-seeking consumers looked for alternatives that the consumer

was familiar with. College students appeared to either be loyal to a brand or be a risk-taker. They









did not look for alternatives to their favorite brands. However, they may be looking for

something new and different to become their new favorite. Curiosity-motivated behavior meant

that the consumer sought out information through shopping and interpersonal means. The fact

that this is not found to be significant reiterates the subjective norm results found from other

variables.

Study findings provided evidence that there was not a significant association between

brand choice of beer among college students and situational variation. Although, some of the

results had consistency with brand positioning (i.e. respondents drank Corona by the pool), there

was not significant data found situation and brand choice. College students appeared to either

remain brand loyal or strayed to other brands based on other factors.

Significant factors influencing beer brand choice were price, risk-taking behavior, and

environment. The main factor that influenced brand choice was price. This supports previous

research in terms of college students purchasing beer based on pricing for that particular product

category. As price goes up, then the brand choice for the brand declines. Inversely, as the price

decreases, then brand choice for the brand increases. Risk-taking behaviors were found to be

prevalent among college students with brand choice. These behaviors coincided with pricing

because college students were open to choice with the product category of beer. Therefore,

college students were willing to take risks when choosing a brand of beer, and pricing could be a

major factor influencing this risk-taking behavior. For example, a student could potentially

retract from buying their favorite beer and partake in risk-taking behavior in order to save some

money by purchasing a cheaper brand of beer that is not their favorite. Price and risk-taking

behavior had an inverse relationship among the two most important variables. Therefore, as the

price of a beer brand decreases, then the likelihood of a college student to exhibit risk-taking









behavior increases. College student are very price sensitive. Some loyalties and favorites exist.

However, most college students cannot turn down a bargain. Therefore, in terms of impacting

brand switching behaviors, the most influential factor is pricing strategies and promotions.

It was unclear if marketing in a particular situation would be effective because there was

not a relationship discovered that influences a beer brand choice with situation. However, the

situation does influence the amount of beer consumed. However, marketers must be very

cautious if they market to these heavy to moderate drinking behaviors and maintain a responsible

outlook when attempting to reach these high beer consumption consumers.

Beer consumption behaviors varied based on the situations that college students were

involved. College students consumed lighter amounts of beer in situations such as special events,

beach/pool areas, at a restaurant or relaxing at home, or in any situation that is not a party, bar, or

club setting. Heavy and moderate beer consumption occurred mostly in bars, clubs, or parties.

College students tend to exhibit consumption behavior based on the situation that they are in.

Variables that were significant with beer consumption behavior were quality, emotion, health,

and product category involvement. The most influential factor was emotion. Based on beer

consumption behaviors, a light beer drinker is more likely to not have a strong emotional

attachment to brands. However, a heavy to moderate drinker has more of a strong emotion for

brands of beer. In addition, the product category involvement and quality also seems to coincide

with these behaviors. If college students were highly involved with beer, then they consume

more and had a greater interest in the overall quality of the beer. College students had mild

concerns regarding health, and this could deal with light beer consumption. The health concerns

associated with drinking beer could potentially be curving the behavior with the consumption.









For example, if a college student is concerned with gaining weight due to carbohydrates, then the

student would be more likely to consume less beer.

College students are a vast market that contains different segments involving several

factors that influence their brand choice. It was apparent that there are divisions among their beer

consumption behaviors. Light beer drinkers appeared to be willing to try different brands and

base purchasing on pricing. Heavy to moderate beer consumption behaviors seemed to be more

passionate emotionally about brands and were more involved with the product category of beer.

This group could potentially exhibit brand loyalties. This ideal is similar in marketing to the

80/20 rule, where 20% of the market for a brand represents 80% of the sales for that particular

brand.

Among group differences in analysis in beer consumption behaviors, groups contrast

because moderate to heavy beer drinkers tend to be less likely to take a risk in their shopping

behaviors. The light beer drinkers were vast in numbers, but they are less likely to carry loyalties

to brands of beer. Marketers should recognize these factors in the college student population. A

very specific, targeted message involving emotion for a beer brand could affect buying behavior

for this heavy to moderate beer consumer. These college students want to be recognized on a

personal, emotional level and have a strong passion for beer. In marketing to the vast segment of

college students influenced by pricing and risk-taking behaviors, the message should involve

competitive pricing strategies and emphasize differentiation. Brand extensions and new beer

products would be of interest for this group.

With the Theory of Reasoned Action (TRA) model in mind, it could be potentially feasible

to present information from this study in order to seek certain behaviors from the college student

market. Advertising to this market provides the opportunity to influence brand attitudes,









intention, and ultimately attempt to result in a purchasing behavior for a particular brand of beer.

In addition, advertising may have some indirect impact on subjective norms by reaching opinion

leaders or individuals among a social group. The model also creates areas of investigation where

further research could be utilized in order to determine influential factors that were not explored

which have an impact beer consumption and beer brand choice.

This research study further validates the notion of using brand benefits instead of

measuring product attributes for brand choice research. In addition, insights from the college

market were discovered from this research study. This study has expanded upon previous brand

choice studies. In addition, this study adds to the research studies involved with the topic of

college students and consumption behavior. There has been evidence provided that influential

factors do influence beer brand choice and beer consumption behavior among college students.

However, this study was only exploratory in nature. This research creates a starting point for

further areas of investigation involving situation variation, the college student marketing, beer

brand choice, and beer consumption behavior.









CHAPTER 6
LIMITATIONS

The influential factors on beer brand choice with brand benefits among other variables

have not been explored previously. More research is needed in order to have a better

understanding of the college market. In this study, none of the demographic factors were found

to be significant. This could have been attributed to the sample. It is evident that there were vast

differences in the college market when it involves making a beer brand choice. In addition, the

population used in this study was not a true representation of the college market. Sampling was

limited to courses in the College of Journalism and Communications at the University of Florida.

Therefore, there were differences among gender, Greek membership, employment, and age could

have been affected by this.

In addition, two measures used in the study scored low reliability scores and were reduced

to two items. Health (reduced from four items to two) and curiosity-motivated behavior (reduced

from three items to two) scored reliability scores of .74 and .53 respectively. This reliability may

have hindered the results for these variables involved in this study. Health was found to be

statistically significant for beer consumption behaviors and curiosity-motivated behavior was not

found to be significant for neither brand choice or consumption behavior. These variables may

need further investigation and research. However, these measures were found to have an

acceptable reliability in previous studies (Orth, 2005).

Based on the broken down sample of 222 to 98, the sample could not be further narrowed

in order to achieve more reliable results based on multiple and stepwise discriminant analyses.

This study does not use previously recognized alcohol consumption levels used in previous

studies. Beer brand consumption behavior levels used in this study differed in contrast to other

studies involving both the general population and other studies using college students.









Situational variation and beer brand choice was not found to be significant. This could

have resulted from many possible problems. The sample size could have been too small in order

to create differentiated results, the construction of measuring variables may not have been the

best way to indicate situational variation, or there could have been a problem with recalling the

exact situation in which the respondent was last drinking their favorite beer. The fact that the

beer brand choice situation question asked college students where they last consumed there

favorite beer also creates some limitations as well. It is possible that a college student may drink

their favorite beer more routinely in other locations other than simply the last location they

consumed it. Therefore, in future research studies, the measure of this variable should be

carefully analyzed in order to gain significant and valid findings.

Multiple discriminant analyses were used in the main analysis for the study. However, a

stepwise discriminant analysis was used in order to limit multicollinearity. Stepwise discriminant

analysis has been criticized in other research studies. The issues that have been associated with

this analysis are incorrect degrees of freedom based on computer packages to analyze data,

sampling error capitalization, and failure to select the best subset of variables in a given size.

Problems with this research could be addressed and improved with more research on the topic,

recalculations, improved sampling, further analyses on data, and better reliability for construct

scales would aid in improving the current research (Thompson, 1995).

The Theory of Reasoned Action (TRA) model was used as a theoretical foundation for this

research study. However, the model does involve some limitations. One limitation of the TRA

model is based on ignoring internal or external conditions influencing or hindering the

individual's behavior. Extraneous variables such as alcohol tolerance, opportunity, and money

could influence behavior (Netemeyer et al. 1991). Also, the model does not recognize more than









one behavior at a time. It is possible for an individual to have simultaneous intentions leading to

multiple behaviors. The model uses attitudes and subjective norms to predict behavior. However,

based on this model, there is limited amount of learning that can be obtained (Ajzen and

Fishbein, 1980; Schlegel et al. 1977; Hays, 1985).









APPENDIX A
PRETEST









Hello,


My name is Dave Ritter, and I am a graduate student in advertising at the University of Florida. This is a research study questionnaire
for my thesis project. My topic involves the study of attitudes, perceptions, and decisions involving brand choice for the product
category of beer. Please take a few moments to read over the questions carefully. There are no right or wrong answers, so please
answers the questions honestly. Your results will remain confidential, and you will be anonymous upon the completion of this survey.
I would just like to thank you very much for your participation in this study.

Returning the survey indicates your consent for use of the answers you supply. If you have any questions, you may contact Dave
Ritter at 309-531-3509. You may also reach me at dritter@ufl.edu.

Please fill out the follow information in order to receive credit for participating in this study. This section will be separated from the
survey upon completion to protect your anonymity.

Name:

UF E-Mail:


UF ID:









Construct: Brand Choice


1. Think about the brands of beer you have consumed in the past month. What brands were they?
apply.


Please check as many or as few as


Amber Bock
Amstel Light
Beck's
Blue Moon
Bud Extra
Bud Light
Bud Select
Budweiser
Busch
Coors Light
Corona


Dos Equis
Fosters
Guinness
Heineken
Keystone Light
Killians
Michelob
Michelob Ultra

Miller High Life
Miller Lite

Natural Light


Old Milwaukee's Best
Pabst Blue Ribbon
Red Stripe
Rolling Rock
Sam Adams
Schlitz
Steel Reserve
Stella
Yuengling

Other, Please Specify
None of the Above


Construct: Consumption Behavior

2. On average how many days a week do you drink beer (place a circle around the appropriate number)?


0 1 2 3 4 5 6 7









Construct: Situational Factors


3. Please indicate by circling how likely you are to drink beer in the following situations (place a circle around a number for each
situation).


In a bar or club with a date
In a bar or club with friends
At a sporting event
Family events
At a restaurant with a date
At a restaurant with friends

Party with friends at home
Alone at home


Not Very Likely
1 2
1 2
1 2
1 2
1 2
1 2
1 2
1 2


Very Likely
6 7
6 7
6 7
6 7
6 7
6 7
6 7
6 7


Construct: Situational Factors

4. In addition to these situations, what are other occasions or situations that you drink beer? Please indicate your response in the
blanks provided below.









Construct: Consumption Behavior


5. How much beer on average do you drink (one drink = 12 oz. bottle and/or can) in one occasion?
(Please do not include mixed drinks or other drinks that are not considered beer for your answer)



The following questions ask about your perceptions and attitudes regarding Budweiser. Please answer them to the best of our
knowledge.

Construct: Quality/Performance Benefits

6. Please indicate the extend of your agreement with the following statements about the quality of Budweiser beer (place only one
circle for each statement).


Strongly Disagree


Budweiser has consistent quality.
o
W Budweiser is well made.

Budweiser has an acceptable standard of quality.
Budweiser has poor craftsmanship.
Budweiser is a brand that would last a long time among other brands of beer.
Budweiser would perform consistently.
Construct: Health Benefits


1 2 3
1 2 3
1 2 3
1 2 3
1 2 3
1 2 3


Strongly Agree
6 7
6 7
6 7
6 7
6 7
6 7


7. Please indicate in the blanks below all the health benefits from Budweiser that you can think of.









Construct: Price/Value for Money Benefits


8. Please indicate the extend of your agreement with the following statements about the
circle for each statement).


pricing of Budweiser beer (place only one


Budweiser is reasonably priced.
Budweiser offers value for money.
Budweiser is a good product for the price.
Budweiser is economical.
Construct: Normative/Personal (Social) Benefits


Strongly Disagree
1 2
1 2
1 2
1 2


Strongly Agree
6 7
6 7
6 7
6 7


9. Please indicate the extend of your agreement with the following statements about the
one circle for each statement).
Strongly Disagree


Budweiser helps me feel acceptable.
Budweiser improves the way I am perceived by others.
Budweiser makes a good impression on other people.
Budweiser gives its owner social approval.


social impact of Budweiser beer (place only


1 2 3
1 2 3
1 2 3
1 2 3


Strongly Agree
6 7
6 7
6 7
6 7


Construct: Emotion Benefits

10. Please indicate the extend of your agreement with the following statements about your attitudes and enjoyment of Budweiser
beer (place only one circle for each statement).


Budweiser is a product that I would enjoy.
Things about Budweiser make me want to use it.
Budweiser makes me feel relaxed.


Strongly Disagree
1 2
1 2
1 2


Strongly Agree
3 4 5 6 7
3 4 5 6 7
3 4 5 6 7









Construct: Emotion Benefits


Budweiser makes me feel good.
Budweiser would give me pleasure.
Budweiser evokes thoughts of happiness.
Budweiser soothes me.
Budweiser eliminates all fear.
Budweiser eliminates all anger.
Budweiser makes me anxious.
Budweiser makes me want to use it.


1 2
1 2
1 2
1 2
1 2
1 2
1 2
1 2


Construct: Environment (Stewardship) Benefits

11. Please indicate the extend of your agreement with the following statements about the environmental impact of Budweiser beer
(place only one circle for each statement).


Strongly Disagree


Budweiser is produced in an environmentally friendly way.
Budweiser is made without polluting the environment.
Budweiser is made by people who care for the environment.
Construct: Health Benefits


1 2 3
1 2 3
1 2 3


Strongly Agree
6 7


6 7
6 7


12. Please indicate the extend of your agreement with the following statements about the health implications of Budweiser beer
(place only one circle for each statement).


Strongly Disagree
Budweiser comes with lots of health benefits. 1 2
Budweiser promotes one's health when enjoyed in moderation. 1 2


Strongly Agree
3 4 5 6 7
3 4 5 6 7









Construct: Brand Choice


13. Think about the brands of beer that you plan to consume in the next month. Please rank order the following brands of beer by
placing a '1' next to the brand of beer that you will most likely consume in the next month, a '2' next to the brand you feel the next
likely to be consumed, through '32' (including Other) which would represent the brand that you are least likely to consume in the next
month. If you do not plan to consume any of the following brands, then please check None of the Above.


Amber Bock
Amstel Light
Beck's
Blue Moon
Bud Extra
Bud Light
Bud Select
Budweiser


Busch


Coors Light
Corona
Dos Equis
Fosters


Heineken


Steel Reserve


Keystone Light
Killians


Michelob


Michelob Ultra

Miller High Life
Miller Lite


Stella


Yuengling
Other Specify
None of the Above


Natural Light
Old Milwaukee's Best
Pabst Blue Ribbon


Red Stripe
Rolling Rock
Sam Adams


Guinness


Schlitz









Please take a few moments to fill to me a little about yourself by circling the appropriate responses. Please keep in mind that your
information will remain completely confidential and anonymous.

14. What is your gender?

1) Male
2) Female
15. What is your current age (write in number in years)?


16. What is your current year as a student?

1) Freshman
2) Sophomore
3) Junior
4) Senior
5) Master's Graduate Student


6) PHD Student


7) Other


Please Specify


17. What is your current employment status?

1) Full Time
2) Part Time
3) Unemployed









18. What is your current major?


1) Public Relations
2) Advertising
3) Telecommunications
4) Journalism
5) Other Please Specify


19. Please circle your current living situation?

1) Live with Parents
2) Live alone (Apartment or House) on campus
3) Live alone (Apartment or House) off campus
o4) Live with roommates) (Apartment or House) on campus

5) Live with roommate(s) (Apartment or House) off campus
6) Fraternity or Sorority House
7) Live in a Dorm Alone
8) Live in a Dorm with roommate(s)
9) Other Please Specify

20. Are you a member of a fraternity or sorority?
1) Yes
2) No









APPENDIX B
MAIN STUDY









Hello,


My name is Dave Ritter, and I am a graduate student in advertising at the University of Florida. This is a research study questionnaire
for my thesis project. My topic involves the study of attitudes, perceptions, and decisions involving brand choice for the product
category of beer. Please take a few moments to read over the questions carefully. There are no right or wrong answers, so please
answers the questions honestly. Your results will remain confidential, and you will be anonymous upon the completion of this survey.
I would just like to thank you very much for your participation in this study.

Returning the survey indicates your consent for use of the answers you supply. If you have any questions, you may contact Dave
Ritter at 309-531-3509. You may also reach me at dritter@ufl.edu.

Please fill out the follow information in order to receive credit for participating in this study. This section will be separated from the
survey upon completion to protect your anonymity.

Name:

E-Mail:


UF ID:









Please take a moment to think about the last time you consumed your favorite beer.

Construct: Brand Choice

1. What brand of beer is it (Please write only one beer brand in the blank provided below)?



Construct: Situational Factors

2. For the brand you mentioned in question 1, which of the following situations did you last consume this beer (Please circle only
one choice for your answer)?

1) In a bar or club with friends 9) At home alone (not a party) 16) Party (not at home) with friends

2) In a bar or club with a date 10) Sporting event/Concert with friends 17) Party (not at home) with date

3) In a bar or club alone 11) Sporting event/Concert with date 18) Party (not at home) alone

4) In a restaurant with friends 12) Sporting event/Concert alone 19) Party at your home

5) In a restaurant with a date 13) Beach/Pool with friends (not a 20) Other; Please Specify

6) In a restaurant alone party)

7) At home with friends (not a party) 14) Beach/Pool with date (not a party)

8) At home with a date (not a party) 15) Beach/Pool alone (not a party)









3. For the brand you mentioned in question 1, when did you most recently consume this beer (Please circle only one choice for your
answer)?
1) A week ago (within the last 7 days) 4) More than one month but less than 3 months (30-90 days)

2) More than one week but less than two weeks (8-14 days) 5) More than 3 months (90+ days)

3) More than two weeks but less than a month (15-29 days) 6) Does not apply

4. For the brand mentioned in question 1, if you purchased this beer, then what was the cost per beer (Please estimate cost for one
beer, where one beer = 12 ounces)? If you did not purchase, then please indicate by writing "N/A" in the blank provided.




Construct: Quality/Performance Benefits

3 5. Please indicate the extent of your agreement with the following statements about your favorite beer (as mentioned in question
one). Please place only one circle for each statement.
Strongly Disagree Strongly Agree
It has consistent quality. 1 2 3 4 5 6 7
It is well made. 1 2 3 4 5 6 7
It has an acceptable standard of quality. 1 2 3 4 5 6 7
It has poor craftsmanship. 1 2 3 4 5 6 7
It would last a long time among other brands of beer. 1 2 3 4 5 6 7
It would perform consistently. 1 2 3 4 5 6 7









Construct: Price/Value for Money Benefits


6. Please indicate the extent of your agreement with the following statements about your favorite beer (as mentioned in question
one). Please place only one circle for each statement.


It is reasonably priced.
It offers value for money.
It is a good product for the price.
It is economical.


Strongly Disagree
1 2
1 2
1 2
1 2


Strongly Agree
6 7


6 7
6 7
6 7


Construct: Normative/Personal (Social) Benefits

7. Please indicate the extent of your agreement with the following statements about your favorite beer (as mentioned in question
one). Please place only one circle for each statement.


It helps you feel acceptable.
It improves the way you are perceived by others.
It makes a good impression on other people.
It gives you social approval.


Strongly Disagree
1 2
1 2
1 2
1 2


Strongly Agree
3 4 5 6 7
3 4 5 6 7
3 4 5 6 7
3 4 5 6 7









Construct: Emotion Benefits


8. Please indicate the extent of your agreement with the following statements about your favorite beer (as mentioned in question


one). Please place only one circle for each statement.


It is a product that you enjoy.
Things about it make you want to use it.
It makes you feel relaxed.
It makes you feel good.
It gives you pleasure.
It evokes thoughts of happiness.
It soothes you.
It eliminates all fear.
It eliminates all anger.
It makes you anxious.
It makes you want to use it.


Strongly Disagree
1 2 3
1 2 3
1 2 3
1 2 3
1 2 3
1 2 3
1 2 3
1 2 3
1 2 3
1 2 3
1 2 3


Strongly Agree
6 7
6 7
6 7
6 7
6 7
6 7
6 7
6 7
6 7
6 7
6 7


Construct: Environment (Stewardship) Benefits

9. Please indicate the extent of your agreement with the following statements about your favorite beer (as mentioned in question
one). Please place only one circle for each statement.


It is produced in an environmentally friendly way.
It is made without polluting the environment.
It is made by people who care for the environment.


Strongly Disagree
1 2 3
1 2 3


Strongly Agree
6 7
6 7


1 2 3 4 5 6 7









Construct: Health Benefits


10. Please indicate the extent of your agreement with the following statements about your favorite beer (as mentioned in question
one). Please place only one circle for each statement.


It comes with lots of health benefits.
It promotes one's health when enjoyed in moderation.
It is a good way to relieve your stress.
It glorifies unhealthy drinking behaviors.


Strongly Disagree
1 2 3
1 2 3
1 2 3
1 2 3


Strongly Agree
6 7
6 7
6 7
6 7


Construct: Exploratory Shopping Behaviors

11. Please indicate the extent of your agreement with the following statements (place only one circle for each statement).


Strongly Disagree


I have little interest in fads and fashions.
I like to shop around and look at displays.
I shop around a lot for my clothes just to find out more about the latest styles.
I hate window shopping.


1 2 3
1 2 3
1 2 3
1 2 3


Strongly Agree
6 7
6 7
6 7
6 7


12. Please indicate the extent of your agreement with the following statements (place only one circle for each statement).

When I eat out, I like to try the most unusual items the restaurant serves, even if I am not sure I would like them.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree
When I go to a restaurant, I feel it is safer to order dishes I am familiar with.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree









I am very cautious in trying new/different products.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree
I would rather wait for others to try a new store or restaurant than try it myself.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree
I never buy something I don't know about at the risk of making a mistake.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree
13. Please indicate the extent of your agreement with the following statements (place only one circle for each statement).

A new store or restaurant is not something I would be eager to find out about.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree
Investigating new brands of grocery and other similar products is generally a waste of time.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree
When I hear about a new store or restaurant, I take advantage of the first opportunity to find out more about it.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree


Construct: Susceptibility to Interpersonal Influence

14. Please indicate the extent of your agreement with the following statements (place only one circle for each statement).

I often consult other people to help choose the best alternative available from a product class.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree
If I want to be like someone, I often try to buy the same brands that they buy.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree
It is important that others like the products and brands I buy.









Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree
To make sure I buy the right product or brand, I often observe what others are buying and using.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree
I rarely purchase the latest fashion styles until I am sure my friends approve of them.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree
I often identify with other people by purchasing the same products and brands they purchase.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree
If I have little experience with a product, I often ask my friends about the product.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree
When buying products, I generally purchase those brands that I think others will approve of.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree
I like to know what brands and products make good impressions on others.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree
I frequently gather information from friends or family about a product before I buy.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree
If other people can see me using a product, I often purchase the brand they expect me to buy.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree
I achieve a sense of belonging by purchasing the same products and brands that others purchase.
Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree









Construct: Consumption Behavior


15. On average how many days a week do you drink beer (place a circle around the appropriate number)?

0 1 2 3 4 5 6 7

16. How much beer on average do you drink (one drink = 12 oz. bottle and/or can) in a one occasion?
(Please do not include mixed drinks or other drinks that are not considered beer for your answer)

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17+

Construct: Product Category Involvement

17. Please indicate the extent of your agreement with the following statements (place only one circle for each statement).


Strongly Disagree
Generally, I am someone who finds it important what beer he or she buys. 1 2
Generally, I am someone who is interested in the kind of beer he or she buys. 1 2
Generally, I am someone whom it means a lot what beer he or she buys. 1 2


Strongly Agree
3 4 5 6 7
3 4 5 6 7
3 4 5 6 7


Please take a few moments to tell us about yourself by circling the appropriate responses. Please keep in mind that your information
will remain completely confidential and anonymous.

18. What is your gender?


1) Male


2) Female


19. What is your current age (write in number in years)?









20. What is your current year as a student?


1) Freshman
6) PHD Student


2) Sophomore
7) Other


3) Junior
Please Specify


4) Senior 5) Master's Graduate Student


21. What is your current employment status?


1) Full Time


2) Part Time


3) Unemployed


22. What is your current major?


1) Public Relations 2) Advertising
5) Other Please Specify


23. Please circle your current living situation?

1) Live with Parents
2) Live alone (Apt/Home) on campus
3) Live alone (Apt/Home) off campus
4) Live with roommate(s) (Apt/Home) on campus
5) Live with roommate(s) (Apt/Home) off campus


24. Are you a member of a fraternity or sorority?
1) Yes 2) No


3) Telecommunications


4) Journalism


6) Fraternity or Sorority House
7) Live in a Dorm Alone
8) Live in a Dorm with roommate(s)
9) Other Please Specify


Thank you for your participation in this study and remember if you do drink, then please drink responsibly!!!









REFERENCES


Aaker, D. A., Batra, R. and Myers, J. G. (1992), Advertising Management, 4th ed, Prentice Hall,
Englewood Cliffs, NJ.

Ajzen, I. and Fishbein, M. (1980), Understanding Attitudes and Predicting Social Behavior,
Englewood Cliffs, NJ: Prentice-Hall, Inc.

Alcohol makers tread tricky path in marketing to college students. (2005) USA
Today, Money, pp. 0lb.

Alvarez, B. A. and Casielles, R. V. (2005), Consumer evaluations of sales promotion: the effect
on brand choice, European Journal ofMarketing, Vol. 39, No. 12, pp. 54-70.

Andorfer, J. B. (2005), Got Beer? A-B plots health pitch, Advertising Age, Vol. 76,
Issue 42.

Auger, P., Burke, P., Devinney, T. M., and Louviere, J. J. (2003), What will consumers pay for
social product features?, Journal of Business Ethics, Vol. 42, No. 3, pp. 281-304.

Babor, T. F., and Mendelson, J. H., Greenberg, I., and Kuehnle, J. (1978), Experimental analysis
of the 'happy hour: effects of purchase price on alcohol consumption',
Psychopharmocology, Vol. 58, pp. 35-41.

Bagozzi, R. P. (1986), Principles of Marketing Management, Science Research Associates,
Chicago, IL.

Bagozzi, R. P., Baumgartner, H., and Yi, Y. (1992), State versus action orientation and the
theory of reasoned action: application to coupon usage, Journal of Consumer Research,
Vol. 18, March, pp. 505-518.

Ballantyne, R., Warren, A., and Nobbs, K. (2006), The evolution of brand choice,
Brand Management, Vol. 13, No. 4/5, pp. 339-352.

Bandura, A. (1977). SocialLearning Theory. New York: Prentice Hall.

Basler, R. P. (1953), Collected works of Abraham Lincoln, Vol. 1, New Brunswick, New Jersey:
Rutgers University Press.

Batra, R., Homer, M.P., and Kahle, L. R. (2001), Values, susceptibility to normative influence,
and attribute importance Weights: A Nomological Analysis, Journal of Consumer
Psychology, Vol. 11, No. 2, pp. 115-128.

Baumgartner, B. (2003), Measuring changes in brand choice behavior, Schmalenbach
Business Review. Vol. 55, pp. 242-256.









Bearden, W. 0. and Etzel, M. J. (1982), Reference group influence on product and brand
purchase decisions, Journal of Consumer Research, Vol. 9, No. 2, pp. 183-194.

Bearden, W., Netemeyer, R. G., and Tell, J. E. (1989), Measurement of consumer susceptibility
to interpersonal influence, Journal of Consumer Research, Vol. 15, No. 4, pp. 473-481.

Beer has an image crisis; wine and spirits gain. (2005) USA Today, 1AP.

Beer's Bad Rap for Carbs Unjustified. (2006), USA Today Magazine, Vol. 134, Issue
2731.

Belch, G. E. and Belch, M. A. (1995), Introduction to Advertising and Promotion: An Integrated
Marketing Communications Perspective, 3rd ed., Boston, MA: Irvin.

Belk, R. (1974), An exploratory assessment of situational effects in buyer behavior,
Journal of Marketing Research, Vol. 11, pp. 156-163.

Bentz, Y. and Merunka, D. (2000), Neural networks and the multinomial logit for brand choice
modelling: a hybrid approach, Journal ofForecasting, Vol. 19, pp. 177-200.

Berden, T. P. J., Brombacher, A. C., and Sander, P. C. (2000), The building bricks of product
quality: An overview of some basic concepts and principles, International Journal of
Production Economics, Vol. 67, pp. 3-15.

Bern, C., Contifias, M., Elorz, M., and Mugica, J. M. (2004), The use of retail store
database for brand choice analysis, Int. Rev. ofRetail, Distribution and Consumer
Research, Vol. 14, No. 1, pp. 19-29.

Boulard, G. (2005), Drinking too much, too: trying to find an answer to the persistent
habit or binge drinking among young people vexes the nation's policymakers. State
Legislatures, pp. 12-15.

Bowlby, J. (1979), The Making and Breaking ofAffectional Brands, London: Tavistock.

Bozinoff, L. and Victor, R. J. (1984), Recall and recognition memory for product attributes and
benefits, Advances in Consumer Research, Vol. 11, No. 1, pp. 348-352.

Braglia, M. and Petroni, A. (2000), Stakeholders' influence and internal championing of product
stewardship in the Italian food packaging industry, Journal ofIndustrial Ecology, Vol. 4,
No. 1, pp. 75-92.

Bruen, J. (2002), Product liability: The role of the product steward, Risk Management, Vol. 49,
No. 2, pp. 34-35.

Calantone, R. and Knight, G. (2000), The critical role of product quality in the international
performance of industrial firms, Industrial Marketing Management, Vol. 29, pp. 493-506.









Charlton, P. and Ehrenberg, A. S. C. (1973), McConnell's experimental brand choice
data, Journal ofMarketing Research, Vol. X, pp. 302-307.

Chib, S., Seetharaman, P. B., and Strijnev, A. (2004), Model of brand choice with a no
purchase option calibrated to scanner-panel data, Journal ofMarketing Research, Vol.
XLI, pp. 184-196.

Choi, C. S. and Coughlan, A. T. (2006), Private label positioning: Quality versus feature
differentiation from the national brand, Journal ofRetailing, Vol. 82, No, 2, pp. 79-93.

Cioletti, J. (2006), Spread the brews, Beverage World, Vol. 125, No. 1.

Christie, J., Fisher, D., Kozup, J. C., Smith, S., Burton, S., and Creyer, E. H. (2001), The effects
of bar-sponsored alcohol beverage promotions across binge and nonbinge drinkers,
Journal ofPublic Policy and Marketing, Vol. 20, No. 2, pp. 240-253.

Coate, D. and Grossman, M. (1988), Effects of alcoholic beverage prices and legal
drinking ages on youth alcohol use, Journal ofLaw andEconometric Evidence. pp. 145-
171.

Collins, R. L., Schell, T., Ellickson, P. L., and McCaffrey, D. (2003), Predictors of beer
advertising awareness among eighth graders, Addiction, Vol. 98, pp. 1297-1306.

Coxe, D. (2004), Sex is out, carbs are in, Maclean 's, Vol. 117, No. 5.

Day, J. C. and Jamieson A. (2005) School enrollment: 2000, U.S. Census Bureau. Washington,
DC: U.S. Government Printing Office. Retrieved April 12, 2006 from
http://www.census.gov/prod/2003pubs/c2kbr-26.pdf

Domestic ad spending by category (2006), Advertising Age, from TNS Media
Intelligence. Retrieved September 27, 2006, from www.adage.com

Eagly, A. H. and Shelly, C. (1993), The Psychology ofAttitudes. Orlando, FL: Harcourt Brace
Jovanovich, Inc.

Ennew, C. T., Reed, G. V., and Binks, M. R. (1993), Importance-performance analysis and the
measurement of service quality, European Journal ofMarketing, Vol. 27, No. 2, pp. 59-
70.

Erdem, T. and Swait, J. (2004), Brand credibility, brand consideration, and choice, Journal
of Consumer Research, Vol. 31, pp. 191-198.

Faden, Vivian B. (2006), Trends in initiation of alcohol use in the United States 1975 to 2003,
Alcoholism: Clinical and Experimental Research, Vol. 30, No. 6, pp. 1011-1022, June,
from NHSDA 1998.









Fall 2005 Product Beverages. (2006), MRI+ Mediamark Internet Reporter database 2006,
accessed April 12, 2006

Farquar, P. H. (1989), Managing brand equity, Marketing Research, Vol. 1, pp. 24-
33.

Feltham, T. S. (1998), Leaving home: Brand purchase influences on young adults, Journal of
Consumer Marketing, Vol. 15, No. 4, pp. 372-385.

Fomell, C., Johnson, M. D., Anderson, E. W., Cha, J. and Bryant, B. E. (1996), The
American customer satisfaction index: Nature, purpose, and findings, Journal of
Marketing, Vol. 60, pp. 7-18.

Fry, J. N. (1971), Personality variables and cigarette brand choice, Journal ofMarketing
Research, Vol. III, pp. 298-304.

Green, C. A., Perrin, N. A., and Polen, M. R. (2004), Gender differences in the relationships
between multiple measures of alcohol consumption and physical and mental health,
Alcohol Clin Exp Res, Vol. 28, pp. 754-764.

Grossman, M., Coate, D., and Arluck, G. M. (1987), Price sensitivity of alcoholic beverages in
the United States, Control Issues in AlcoholAbuse Prevention: Strategies for States and
Communities, Harold H. Holder ed., JAI Press, Inc., Greenwich, Conn., pp. 169-198.

Gruenewald, P., Madden, P., and Janes, K. (1992), Alcohol availability and formal power and
resources of state alcohol beverage control agencies, Alcohol Clin Exp Res, Vol. 16, pp.
591-597.

Gruenewald, P., Miller, A., and Treno, A. (1993), Alcohol availability and the ecology of
drinking behavior, Alcohol Health Res World, Vol. 17, pp. 39-45.

Haley, R. I. (1968), Benefit segmentation: A decision-oriented research tool, Journal of
Marketing, Vol. 32, pp. 30-35.

Hartford, T. C., Weschsler, H., and Rohman, M. (1983), 'The structural context of college
drinking', Journal of Studies on Alcohol, Vol. 44, pp. 722-732.

Havan, C. and Shaver, P. R. (1994), Attachment as an organizational framework for research on
close relationships, Psychological Inquiry, Vol. 5, pp. 1-22.

Hays, R. (1985), An integrated value-expectancy theory of alcohol and other drug use, British
Journal ofAddiction, Vol. 80, pp. 379-384.

Helliker, K. and Ellison, S. (2005), Anheuser wants world to know beer is healthy,
Wall Street Journal (Eastern edition), New York, N.Y., pp. B1.









Hoyer, W. D. and MacInnis, D. J. (2004), Consumer Behavior, 3rd Edition, Boston: Houghton
Mifflin Company.

Hickle, G. T. and Stitzhal, D. (2003), Apportioning the responsibilities for product stewardship:
A case for a new federal role, Environmental Quality Management, Vol. 12, No. 3, pp.1-
16.

Hruschka, H. (2002), Market share analysis using semi-parametric attraction models, European
Journal of Operational Research, Vol. 138, pp. 112-225.

Snider, M. (2006), IPods knock over beer mugs, USA Today; Section: Life, pp. 09d.
taken from Spring 2006 Lifestyle and Media Study, Student Monitor.

Is the glass half-empty or half-full? (2004), Tufts University Health and Nutrition
Letter, Vol. 22, No. 9.

Jessor, R. (1981), The perceived treatment environment in psychological theory and Research. In
D. Magnusson (ed.), Toward a Psychology of Situations: An Interactionist Perspective,
Hillsdale, NJ: Lawrence Erlbaum Associates.

Johnston, L. D., O'Malley, P.M., Bachman, J. G., and Schulenberg, J. E. (2004), Monitoring the
Future National Survey Results on Drug Use, 1975-2003. Vol. II: College Students and
Adults Ages 19-45, Bethesda, MD: National Institute on Drug Abuse, NIH Pubilication
No. 04-5508.

Jones-Webb, R., Toomey, T., Short, B., Murray, D., Wagenaar, A., and Wolfson, M. (1997),
Relationship among alcohol availability, drinking location, alcohol consumption, and
drinking problem in adolescents, Subst Use Misuse, Vol. 32, pp. 219-228.

Kandel, D. B. (1980), Drug and drinking behavior among youth, In A. Inkeles, N.J. Smelser and
R. Turner (eds.), AnnualReview of Sociology, Vol. 6, pp. 235-285, Palo Alto, CA:
Annual Reviews, Inc.

Keller, K. L. (1993), Conceptualizing, measuring, and managing customer-based brand equity,
Journal ofMarketing, Vol. 57, No. 1, pp. 1-22.

Klasky, A. L. (2006), Drink to your health?, Scientific American Special Edition, Vol. 16, No. 4.

Knibbe, R. A., Oostveen, T., and Van de Goor, I. (1991), Young people's alcohol consumption
in public places: Reasoned behavior or related to the situation? British Journal of
Addiction, Vol. 86, pp. 1425-1433.

Kondo, K. (2004), Beer and health: Preventative effects of beer components on lifestyle related
diseases, BioFactors, Vol. 22, pp. 303-310.









Kuo, M., Wechsler, H., Greenberg, P., and Lee, H. (2003), The marketing of alcohol to college
students: The role of low prices and special promotions, American Journal of Preventive
Medicine, Vol. 25, No. 3, pp. 204-211.

Lai, A. (1991), Consumption situation and product knowledge in the adoption of a new product,
European Journal ofMarketing, Vol. 25, No. 10, pp. 55-67.

Lambert, Z. V. (1972), Price and choice behavior, Journal ofMarketing Research,
Vol. IX, pp. 35-40.

Lancaster, K. J. (1971), Variety, Equity and Efficiency, New York: Columbia Press.

Laurent, G. and Kapferer, J. (1985), Measuring consumer involvement profiles, Journal of
Marketing Research, Vol. 22, No. 1, pp. 41-53.

Leger, J. and Scholz, D. (2004), The fickle beer consumer, Marketing Magazine, Vol. 109, No.
17.

Levy, S. and Sheflin, N. (1983), New evidence on controlling alcohol use through price, Journal
of Studies on Alcohol, Vol. 44, pp. 929-937.

Levy, S. and Sheflin, N. (1985), The demand for alcoholic beverages: An aggregate time series
analysis, Journal ofPublic Policy and Marketing, Vol. 4, No. 1, pp. 47-54.

Loftus, P. (2005), Study brews on health benefits of hops, Wall Street Journal
(Eastern edition), pp. B4C.

Mayhew, G. E. and Winer, R. S. (1992), An empirical analysis of internal and external reference
prices using scanner data, Journal of Consumer Research, Vol. 19, June, pp. 62-70.

McGuire, W. J. (1968), Personality and susceptibility to social influence. In E. F. Borgetta and
W. W. Lambert (eds), Handbook ofPersonality Theory and Research, Chicago: Rand
McNally, pp. 1130-1187.

Miller, K. E. and Ginter, J. L. (1979), An investigation of situational variation in brand
choice behavior and attitude, Journal ofMarketing Research, Vol. XVI, pp. 111-123.

Netemeyer, R. G., Burton, S., and Johnston, M. (1991), Volitional and goal-directed
behaviors: A confirmatory analysis approach, SocialPsychology Quarterly, Vol. 54, No.
2, pp. 87-100.

O'Callaghan, F. V., Chant, D. C., Callan, V. J., and Baglioni, A. (1997), Models of
alcohol use by young adults: An examination of various attitude-behavior theories,
Journal of Studies on Alcohol, Vol. 58, No. 5, pp. 502-507.









Orth, U. R. (2005), Consumer personality and other factors in situational brand choice variation,
Brand Management, Vol. 13, No. 2, pp. 115-133.

Orth, U. R., McDaniel, M. R., Shellhammer, T., and Lopetcharat, K. (2004), Promoting brand
benefits: The role of consumer psychographics and lifestyle, Journal of Consumer
Marketing, Vol. 21, No. 4, pp. 31-47.

Orth, U. R., McGarry Wolf, M., and Dodds, T. (2005), Dimensions of wine region equity and
their impact on consumers preferences, Journal of Product and Brand Management, Vol.
14, No. 4, pp. 477-489.

Osterberg, E. (1995), Do alcohol prices affect consumption and related problems? In: H. D.
Holder and G. Edwards (eds.), Alcohol and Public Policy: Evidence and Issues. New
York: Oxford University Press, pp. 145-163.

Papatla, P. and Krishnamurthi, L. (1996), Measuring the dynamic effects of
promotions on brand choice, Journal ofMarketing Research, Vol. XXXIII, pp. 20-35.

Park, C. W., Jaworski, B. J., and Maclnnis, D. J. (1986), Strategic brand concept-image
management, Journal ofMarketing, Vol. 50, pp. 621-635.

Park, C. W. and Lessig, V. P. (1977), Students and housewives: Differences in susceptibility to
reference group influence, Journal of Consumer Research, Vol. 4, No. 2, pp. 102-110.

Park, C. W. and Mittal, B. (1985), A theory of involvement in consumer behavior: Problems and
issues, Research in Consumer Behavior, Vol. 1, pp. 201-232.

Puth, G., Mostert, P., and Ewing, M. (1999), Consumer perceptions of mentioned product and
brand attributes in magazine advertising, Journal of Product and Brand Management,
Vol. 8, No. 1, pp. 38-49.

Quester, P. G. and Smart, J. (1998), The influence of consumption situation and product
involvement over consumers' use of product attributes, Journal of Consumer Marketing,
Vol. 15, No. 3, pp. 220-238.

Raju, P. S. (1980), Optimum stimulation level: Its relationship to personality, demographics, and
exploratory behavior, Journal of Consumer Research, Vol. 7, No. 3, pp. 272-282.

Ram, S. and Jung, H. (1989), The link between involvement, use innovativeness and product
usage, Advances in Consumer Research, Vol. 16, pp. 160-166.

Redman, S., Sanson-Fisher, R. W., Wilkinson, C. Fahey, P., and Gibberd, R. W. (1987),
Agreement between two measures of alcohol consumption, Journal of Studies on
Alcohol, Vol. 48, pp. 104-108.









Reeves, C. A. and Bednar, D. A. (1994), Defining quality: Alternatives and implications,
Academy ofManagement Review, Vol. 18, No. 3, pp. 419-445.

Rhodes, P. (2005), 3 cheers for beer, Natural Health, Vol. 35, Vol. 9.

Richins, M. L. and Bloch, P. H. (1986), After the new wears off: The temporal context of
product involvement, Journal of Consumer Research, Vol. 13, No. 2, pp. 280-285.

Romaniuk, J. (2003), Brand attributes 'distribution outlets' in the mind, Journal ofMarketing
Communications, Vol. 9, pp. 73-92.

Romaniuk, J. and Sharp, B. (2003), Measuring brand perceptions: Testing quantity and quality,
Journal of Targeting, Measurement and Analysis for Marketing, Vol. 11, No. 3, pp. 218-
229.

Schlegel, R. P., Crawford, C. A., and Sanburn, W. D. (1977), Correspondence and mediational
properties of the Fishbein model: An application to adolescent alcohol use, Journal of
Experimental Social Psychology, Vol. 13, pp. 421-430.

Schwartz, R. H., Farrow, J. A., Banks, B., and Giesel, A. E. (1998), Use of false ID cards and
other deceptive methods to purchase alcoholic beverages during high school, Journal of
Addictive Diseases, Vol. 17, No. 3, pp. 25-33.

Simonson, I., Carmon, Z., and O'Curry, S. (1994), Experimental evidence on the negative effect
of product features and sales promotions on brand choice, Marketing Science, Vol. 13,
No. 1, Winter, pp. 23-40.

Singh, V. P., Hansen, K. T., and Sachin, G. (2005), Modeling preferences for common
attributes in multicategory brand choice, Journal ofMarketing Research, Vol. XLII, pp.
195-209.

Smith, W. (1956), Product differentiation and market segmentation as alternative marketing
strategies, Journal ofMarketing, Vol. 21, pp. 3-8.

Stafford, J. E. and Cocanougher, B. A. (1977), Reference group theory. SelectedAspects of
Consumer Behavior, Washington, DC: Superintendent of Documents, U.S. Government
Printing Office, pp. 361-380.

Sweeney, J. C. and Soutar, G. N. (2001), Consumer perceived value: The development of a
multiple item scale. Journal of retailing, Vol. 77, No. 2, pp. 203-220.

Thompson, B. (1995), Stepwise regression and stepwise discriminant analysis need not apply
here: A guidelines editioral, Educational and Psychological Measurement, Vol. 55, No.
4, pp. 525-535.









Thomson, M., MacInnis, D. J., and Park, C. W. (2005), The ties that bind: Measuring the
strength of consumers' emotional attachments to brands, Journal of Consumer
Psychology, Vol. 15, No. 1, pp. 77-91.

Trafimow, D. (1996), The importance of attitudes in the prediction of college students' intentions
to drink, Journal ofApplied Social Psychology, Vol. 26, No. 24, pp. 2167-2188.

Trafimow, D. and Fishbein, M. (2001), The moderating effect of behavior type on the subjective
norm-behavior relationship, The Journal ofSocialPsychology, Vol. 134, No. 6, pp. 755-
763.

Uncles, M. D. and Ehrenberg, A. S. (1990), Brand choice among older consumers, Journal of
Advertising Research, August/September, Vol. 30, No. 4, pp. 19-22.

U.S. market share leaders. (2007), Advertising Age, from TNS Media Intelligence.
Retrieved August 11, 2007, from www.adage.com

Van Trijp, H. M. C. (1994), Product-related determinants of variety-seeking behavior for foods.
Appetite, Vol. 22, pp. 1-10.

Van Trijp, H. C. M., Hoyer, W. D., and Inman, J. (1996), Why switch? Product category level
explanations for true variety-seeking behavior, Journal ofMarketing Research, Vol. 33,
No. 3, pp. 281-292.

Vazquez, R., Belen del Rio, R., and Iglesias, V. (2002), Consumer-based brand equity:
Development and validation of a measurement instrument, Journal ofMarketing
Management, Vol. 18, No. /2, pp. 27-48.

Viral can't cure what ails 'beast'. (2006), Advertising Age, Vol. 77, No. 42.

Wagner, U. and Taudes, A. (1986), A multivariate polya model of brand choice and purchase
incidence, Marketing Science, Vol. 5, No. 3, pp. 219-244.

Wahlers, R. G., Dunn, M. G., and Etzel, M. J. (1986), The congruence of alternative OSL
measures with consumer exploratory behavior tendencies, Advances in Consumer
Research, Vol. 13, No. 1, pp. 398-402.

Walker, R. (2004), The idea of a beer with a 'health benefit' is debatable, but that's
clearly what's resonating, New York Times Magazine, Vol. 153, No. 52746, pp. 18.

Wall, A., Hinson, R. E., and McKee, S. A. (1998), Alcohol outcome expectancies, attitudes
toward drinking and the theory of planned behavior, Journal of Studies on Alcohol, Vol.
59, No. 4, pp. 409-419.

Walsh, G. and Mitchell, V. (2005), Demographic characteristics to consumers who find it
difficult to decide, Marketing Intelligence andPlanning, Vol. 23, No. 2/3, pp. 281-295.









White, T. and Pomponi, R. (2003), Gain a competitive edge by preventing recalls, Quality
Progress, Vol. 36, No. 8, pp. 41-49.

Wilkie, W. L. (1986), Consumer Behavior, New York: John Wiley and Sons.

Wilks, J. and Callan, V. J. (1990), A diary approach to analyzing young adults' drinking events
and motivations, Australian Drug and Alcohol Review, Vol. 5, pp. 3-7.

Witt, R. E. (1969), Informal social group influence on consumer brand choice,
Journal ofMarketing Research, Vol. VI, pp. 473-476.

Wood, L. M. (2004), Dimensions of brand purchasing behavior: Consumers in the 18-24 age
group, Journal of Consumer Behavior, Vol. 4, No. 1, pp. 9-24.

Woods, P. (2005), Good news about brews, Health, Vol. 19, No. 8.

Woodside, A. G. and Fleck Jr., R. A. (1979), The case approach to understanding brand choice,
Journal ofAdvertising Research, Vol. 19, No. 2, pp. 23-30.

Xiaohua, L. and Germain, R. (2003), Product quality orientation and its performance
implications in Chinese state-owned enterprises, Journal ofInternational Marketing, Vol.
11, No. 2.

Yang, S., Allenby, M. G., and Fennell, G (2002), Modeling variation in brand preference: The
roles of objective environment and motivating conditions, Marketing Science, Vol. 21,
No. 1, pp. 14-21.

Zaichkowsky, J. L. (1985), Familiarity: Product use, involvement or expertise? In E. C.
Hirschman and M. B. Holbrook (eds), Advances in Consumer Research, Vol. 12, Provo,
UT: Association for Consumer Research.









BIOGRAPHICAL SKETCH

Dave Ritter (davidritter 99@hotmail.com) is a graduate student majoring in advertising

at the College of Journalism and Communications at the University of Florida. His research

focuses on brand choice, marketing, and consumer behavior. Dave completed his Associate of

Science in business administration from Lakeland College. In addition, Dave completed his

Bachelor of Science in marketing from the University of Illinois.





PAGE 1

1 INFLUENTAL FACTORS ON BRAND CHOICE AND CONSUMPTION BEHAVIORS: AN EXPLORATORY STUDY ON CO LLEGE STUDENTS AND BEER By DAVE RITTER A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULLFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ADVERTISING UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Dave Ritter

PAGE 3

3 To my family, friends, educators, and collea gues, cheers! Without your love, support, and guidance none of this would be possible.

PAGE 4

4 ACKNOWLEDGEMENTS I would like to thank my entire family fo r supporting my education and research at the University of Florida. I would like to thank Lake land College, the University of Illinois, and the University of Florida for providing me with the skills and education possible to complete this document. I also would like to express my gratitude to all of my committee members for their help, guidance, and comments throughout this proce ss. In addition, I would like to give a special thank-you to my thesis chair, H yojin Kim. Finally, I would like to thank God for giving me faith, ability, and opportunity to persevere.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGEMENTS.............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES.........................................................................................................................8 ABSTRACT.....................................................................................................................................9 CHAP TER 1 INTRODUCTION..................................................................................................................10 2 LITERATURE REVIEW.......................................................................................................13 Brand Choice..........................................................................................................................13 Desired Brand Benefits...........................................................................................................19 Performance/Quality........................................................................................................20 Price/Value for Money.................................................................................................... 22 Emotions....................................................................................................................... ...23 Normative/Personal (Social)........................................................................................... 23 Environment (Stewardship)............................................................................................. 24 Health......................................................................................................................... .....25 Situational Factors............................................................................................................ ......27 Consumer Factors............................................................................................................... ....29 Exploratory Shopping Behaviors.................................................................................... 30 Interpersonal Influence....................................................................................................30 Consumption Behaviors.................................................................................................. 31 Product Category Involvement........................................................................................ 32 Demographics..................................................................................................................32 Theory of Reasoned Action....................................................................................................33 Research Questions............................................................................................................. ....36 3 METHOD......................................................................................................................... ......37 Pretest.....................................................................................................................................37 Subjects............................................................................................................................37 Design and Procedure...................................................................................................... 38 Measures.................................................................................................................................38 Brand Choice...................................................................................................................38 Situational Variation........................................................................................................ 39 Desired Brand Benefits....................................................................................................39 Consumption Behavior.................................................................................................... 39 Demographics..................................................................................................................40 Main Study..............................................................................................................................40

PAGE 6

6 Subjects............................................................................................................................40 Design and Procedure...................................................................................................... 42 Measures.................................................................................................................................43 Brand Choice...................................................................................................................43 Situational Variation........................................................................................................ 43 Desired Brand Benefits....................................................................................................46 Exploratory Shopping Behaviors.................................................................................... 46 Interpersonal Influence....................................................................................................47 Consumption Behavior.................................................................................................... 47 Product Category Involvement........................................................................................ 49 Demographics..................................................................................................................50 4 RESULTS...............................................................................................................................51 Brand Choice among Situational Variation............................................................................ 51 Consumption Behavior among Situation Variation................................................................ 58 Variables Affecting Beer Brand Choice................................................................................. 67 Variables Affecting Beer Consumption.................................................................................. 79 5 DISCUSSION.........................................................................................................................90 6 LIMITATIONS.......................................................................................................................96 APPENDIX A PRETEST........................................................................................................................ .......99 B MAIN STUDY.....................................................................................................................109 REFERENCES............................................................................................................................120 BIOGRAPHICAL SKETCH.......................................................................................................130

PAGE 7

7 LIST OF TABLES Table page 2-1 Brand Choice Studies .........................................................................................................16 3-1 Beer Brand Choice ............................................................................................................. 41 3-2 Demographic Frequency Distribution................................................................................42 3-3 Situation Frequency Distribution .......................................................................................45 3-4 Average Beers Consumed and Average Occasio ns Drinking Beer................................... 48 3-5 Consum ption Behavior Frequency.................................................................................... 49 4-1 Situational Groups and Brand Choice................................................................................ 53 4-2 Situational Groups and Consum ption Behavior................................................................ 61 4-3 Multip le Discriminant Analysis: Beer Brand Choice........................................................ 73 4-4 Structure Matrix for Beer Brand Choice ............................................................................ 75 4-5 Group Classification for Beer Brand Choi ce: Multiple Discrim inant Analysis................ 75 4-6 Mean and Standard Deviation based on Brand Choice ..................................................... 76 4-7 Stepwise Discrim inant Anal ysis: Beer Brand Choice....................................................... 77 4-8 Group Classification: Beer Brand Choice Stepwise Discrim inant Analysis..................... 79 4-9 Multip le Discriminant Analysis: Beer Consumption Behavior......................................... 84 4-10 Structure Matrix for Beer Consum ption Behavior............................................................. 86 4-11 Group Classification: Beer Consum ption Behavior Multiple Discriminant Analysis....... 86 4-12 Mean and Standard Deviation: Beer Consum ption Behavior............................................ 87 4-13 Stepwise Discrim inant Analysis : Beer Consumption Behavior........................................ 88 4-14 Group Classification: Beer Consum ption Behavior Stepwise Discriminant Analysis...... 89

PAGE 8

8 LIST OF FIGURES Figure page 2-1 Theory of Reasoned Action: Bra nd Choice/Consum ption of Beer................................... 35

PAGE 9

9 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Advertising INFLUENTIAL FACTORS ON BRAND CHOI CE AND CONSUMPTI ON BEHAVIORS: AN EXPLORATORY STUDY OF CO LLEGE STUDENTS AND BEER By Dave Ritter May 2008 Chair: Hyojin Kim Major: Advertising This study examined factors influencing beer brand choice and beer consumption behavior among college students. It was determined after using six situational groupings that situational variation did not ha ve a significant impact on bran d choice of beer for college students. Relationships among bra nd benefits, exploratory shopping behaviors, susceptibility to interpersonal influence, product category involveme nt, and demographics were studied in order to determine what factors had the most significant impact on br and choice. It was determined that price and risk-taking behavi ors had a significant impact on beer brand choice. Consumption behaviors among college students were influenced by emotions. In addition, an association between situation and consumption behavior existed among college students consuming their favorite beer. Bars, clubs, or parties demonstrated heavy to moderate drinking behavior. Situations where college students are relaxing such as at the poo l/beach, at home with friends, and not a party demonstrated li ght beer drinking behaviors.

PAGE 10

10 CHAPTER 1 INTRODUCTION Studies have shown that there are vast num bers of heavy users for consumption of alcohol among college students in the United States. Acco rding to a study, 31.5% of 17 to 25 year olds reported heavy usage once in the past thirty da ys (Faden, 2006). Over 80% of college students drink alcohol (Johnston et al. 2004), and more than 43% of these students have been reported as exhibiting heavy episodic drin king behaviors (Wechsler et al 2003). Demographically, college students that have a greater likelihood of being a heavy user are white, around 23 years old or younger, and are in a fraternity or sorority. Among members of Greek organizations, 75.1% of fraternity members and 62.4% of sorority members are heavy users. Out of all heavy users, 91% of women and 78% of men do not consider themselv es such. Instead, they classify themselves as moderate or light dr inkers (Boulard 2005). Beer is the majority of overall consump tion for alcoholic beverages involving heavy consumption with this consumer segment (Coa te and Grossman, 1988). According to a Fall 2005 report from MRI data, nearly ha lf (43.9%) of consumers aging fr om 18 to 24 purchase beer/ale products. This group is 50% more likely (index of 150) to have purchased beer/ale in the past six months in comparison to all other age groups. In addition, this age group is 21% more likely (index of 121) to consume at least 5 beers in the la st 7 days, which is also the most for any other age group (Fall, 2006). There are factors that co ntribute to the opportunity for beer to be a marketable product in a college market. First, according to a Student Monitors Lifestyle and Media study, college students consider drinking beer one of the top in things on their campuses. This study found drinking beer rated as the s econd in thing on campuses at 71%, tied with the college networking website Facebook.com (Snider, 2006). Second, there is a relationship among the

PAGE 11

11 availability and consumption of al coholic beverages. Therefore, as availability increases, then the consumption of alcoholic beverages increases (G ruenewald et al. 1992; Gruenewald et al. 1993; Jones-Webb et al. 1997). Third, the college market has a large, disposable income. In addition, as beer availability often exposes this market to beer brands for the first time, consumers begin trying brands and formulating their consideratio n set for this product ca tegory. Brand preferences for beer often begin in college, and college students have plenty of income to spend on it with $231 billion annually (Beer, 2005; Ballantyne et al, 2006). In efforts to reach the college market, mark eters have used many different strategies (Beer, 2005). First, pricing strategies have been used based on the assumption that as the price decreases, then consumption increases (Levy a nd Sheflin, 1985). Second, marketers have utilized advertisements at the point of purchase and ha ppy hours, offering discounts on drinks in bars during certain allocated periods of time (Kuo et al. 2003; Christie et al. 2001). In addition, some companies use bar promotions such as theme nights or brand sales representatives in the establishment of purchase to promote their br ands. Third, marketers have utilized college sporting events for reaching college students with event sponsorships (Alcohol, 2005). Finally, companies have been more aggressive using innovative techniques such as viral marketing campaigns on popular websites such as www.youtube.com and targeting areas just off cam pus for promotions and advertising ( Viral, 2006; Domestic, 2006). There are some problems involved with the beer market and college students. First of all, despite high advertising expend itures (advertising spending of $478,000,000 for the first half of the year), the beer product category has been losing market share (U.S., 2007). There has been a decrease in the consumption of beer based on increased usage of mixed drinks and other alcoholic beverages. Second, ther e are many health concerns asso ciated with beer on college

PAGE 12

12 campuses. Companies have a responsibility to pr omote safe drinking practices based on heavy usage (Boulard, 2005). As the level of awareness for beer on college campuses increases, this could potentially lead to nega tive consequences for sponsorships on or off campuses (Kuo et al. 2003). Third, marketers are curren tly battling pricing and promo tional wars with competitors based on heavy product proliferation. These promotions have little consistency among a branding strategy. Therefore, strategies that do no t contribute to a brand co uld lead to increased brand switching for consumers (Beer, 2005; Kuo et al. 2003). If marketed effectively, there is a treme ndous opportunity in the college segment. According to the 2000 U.S. Census Bureau, th is market is compri sed of 17,472,000 people (Day and Jamieson, 2005). It has been said that this group exhibits low brand loyalties. However, according to Wood (2004), To suggest that young consumers have low loyalty would be to miss the richness of their complexity (p. 13). This stu dy is aimed at helping marketers have a better understanding of the current beer and college market. It also aids in finding what factors influence brand choice and consum ption behavior for this product category. Information from this study can help marketers position beer brands and help them segment the market for college students based on the benefit needs of this group (Smith, 1956; Bagozzi, 1986). Marketers can use information from this study to develop the mo st cost effective adve rtising and promotional messages for this market (Vazquez et al. 2002). From a research perspective, there are few studies involving brand choice, th e college market, and the produc t category of beer. Therefore, this study could lead to further re search studies in the future. This study tests the validation of the desired brand benefits and choice model as a predictor of brand choice (Orth, 2005). Finally, this literature adds to existing studies on brand choice.

PAGE 13

13 CHAPTER 2 LITERATURE REVIEW Brand Choice Understanding and predicting brand choice deci sions by consum ers has been a topic of interest to both marketers a nd researchers. Brand choice inve stigation involves understanding consumer behaviors in their se lection of brands among various product categories (Bentz and Merunka, 2000). In the past, brands have been pe rceived as products with different attributes; however, brands are now viewed as personalities, identities, and have special meanings intrinsic to consumers (Ballantyne et al. 2006). Brand ch oice research has been investigated for many years and has intensified as product categories have become more proliferated. For example, 30 years ago there were only a handf ul of beer brands in grocery stores. Now, there are several brands of beer with brand exte nsions featuring light beers, im ports, ice beers, as well as many others. Consumers have more options and many different brands to choose from (Lger and Scholz, 2004). Much of brand choice research has been thr ough probability models to test the impact of marketing mix variables as a pr edictor of brand choice (Wagner and Taudes, 1986; Chib et al. 2004; Bentz and Merunka, 2000). These variables (referred in most re search studies as the 4 Ps) are elements such as product features, displays (i.e. advertising, sales pr omotions), availability (stock of inventory), and pr ice (Chib et al. 2004, May; Bent z and Merunka, 2000; Wager and Taudes, 1986). When used in probability modeli ng, marketing mix variab les are considered nonstationary and heterogeneous among th e population (Wagner and Taudes, 1986). There are other areas that have been research ed with brand choice as well. Researchers have examined the casual effects of brand re lated variables on brand choice. These variables include situational factors, c onsumer personality, social bene fits, emotions, quality, brand

PAGE 14

14 credibility, product attributes, se asonality, and trends. The studies used within brand choice researches have involved experiments and surveys of key marketing variables to measure impact on brand choice (Charlton and Ehrenberg, 1973; Simonson et al. 1994; Erdem and Swait, 2004; Wagner and Taudes, 1986; Orth, 2005). Table 2-1 demonstrates these brand choice studies. Among specific marketing mix variables, pricin g appears to have the most consistent impact in studies. Promotions such as sales pr omotions have shown influence on brand choice which ultimately effect bottom-line prices for co nsumers. For example, pricing promotions could involve coupons or simply a reduction of pri ce within the pr oduct category (Singh et al. 2005; Papatla and Krishnamurthi, 1996; Wagner and Taudes, 1986; Orth, 2005). In probability modeling studies, it has been shown that displays and features have some impact on brand choice, but this evidence is not as overwhelming or as consiste nt as other factors among brand choice research studies (Chib et al. 2004; Papatla and Krishnamurt hi, 1996; Alvarez and Casielles, 2005). Product attributes have hi gh importance on discovering what areas of the product can be altered in order to make their br and more appealing to the consumer. According to current research, it has been found that the greater the number of brand attributes for a product, then the more likely the consumer is to make that particular band choice (Greenwald et al. 1986; Romaniuk, 2003). Pr oduct attributes are important to ma rketers in order to differentiate products from their competitors (Aaker et al. 1992; Belch and Belch, 1995). Non-marketing mix variables have been resear ched in order to disc over external factors that impact brand choice. Seasonality and tre nds have been researched with brand choice. However, their outcomes depend upon the produ ct category. For example, a product such as laundry detergent will most likely have better sales figures in the summertime when the weather is more favorable and people are outside more (Wagner and Taudes, 1986). Personality factors

PAGE 15

15 have shown an impact based on what brands consumers buy. Brand cr edibility has shown significance in determining br and choice as well (Erdem a nd Swait, 2004; Fry, 1971). Other areas such as purchase time, purchase order, and product name have been researched but have not been deemed to be main factors in dete rmining a brand choice decision (Charlton and Ehrenberg, 1973). These studies allow marketer s to understand consumer switching behaviors and allow for market share penetration, which gi ve marketers a better understanding of what elements effect a particular brand or produc t category (Chib et al. 2004; Wagner and Taudes, 1986). Several product categories have been used in order to study brand choice. The majority of product categories include low consumer involve ment retail products. Some examples of products studied in the past with brand choice are laundry detergent, s oda, athletic shoes, ketchup, coffee, snack foods, and bar soaps. Tabl e 2-1 provides a listing of the various product categories used in previous brand choice resear ches (Wagner and Taudes, 1986; Chib et al. 2004; Erdem and Swait, 2004; Baumgartner, 2003; Pa patla and Krishnamurt hi, 1996; Alvarez and Casielles, 2005; Bern et al. 2004; Si ngh et al. 2005; Auger et al. 2003). Among previous brand choice literature, there have been very few studies involving the product category of beer. Woodside and Fleck Jr. (1979) conducted a qualit ative study regarding brand choice of beer drinkers. The methodology for this study consisted of two in-depth personal interviews with two beer drin kers. The researchers conclude d that involvement, normative, situational, and product attributes all influenced bra nd choice in the study. Charlton and Ehrenberg (1973) conducted an experiment with the product category of beer where variables manipulated were price, purchase time, purchas e order, product name, and brand name. More recently a study was conducted (Orth et al. 2004) wh ich examined craft beer preference and the

PAGE 16

16 relationship of brand benefits w ith consumer demographics. Brand benefits were considered to be significant drivers of consumer preferences in this product category. Brand benefits were shown to be an effective predictor in the product category of beer for brand choice. Table 2-1 Brand Choice Studies Author Independent Variables Dependent Variables Product Categories Studied Methods Orth (2005) Situations (Host, Gift, Self) Quality* Social Benefits* Price* Emotional* Health Environment Brand Choice Wine Electronic Survey Wagner and Taudes (1986) Marketing Mix (Advertising, Price)* Seasonality* Trends* Purchase Rate Brand Choice Probability Laundry Detergent Testing of Multivariate Polya Process Model using Consumer Panel Purchase Data Chib et al. (2004) Marketing Mix (Price, Feature, Display) Brand Choice Soda (Beverage) Testing of Model of Brand Choice with Scanner-Panel Data Erdem and Swait (2004) Brand Credibility* (Expertise, Trustworthiness, Perceived Quality, Perceived Risk, Information Cost Saved) Brand Consideration, Brand Choice Athletic Shoes, Cellular Providers, Headache Medication, Personal Computer, Shampoo Survey Papatla and Krishnamurthi (1996) Price* Sales Promotion (Display*, Feature*) Brand Choice Laundry Detergent Testing the Utility Model Using Household Scanner Data Miller and Ginter (1979) Situation* Attributes* Brand Choice Fast Food Restaurants Survey from Mail Panel

PAGE 17

17 Table 2-1 Continued Author Independent Variables Dependent Variables Product Categories Studied Methods Simonson et al. (1994) Quality Rating Brand Name Price Premiums Product Features (Brand Choice) Quality Promotion Consumer Need Brownie Mix 35 mm Film CD Player Experiment (Three Studies) Romaniuk (2003) Product Attributes* Benefit Attributes* Situation-Based Attributes* Brand Choice Fast Food Market Survey Fry (1971) Personality Variables* (Sex, Social Class, Self-Confidence, etc.) Brand Preference Cigarettes Experiment with a Field Study Panel Alvarez and Casielles (2005) Sales Promotions (Price*, Reference Price, Losses and Gains, Sales Promotion Techniques) Brand Choice Soda (Beverage) Testing of Brand Choice Models using Logit Models from Consumer Panel Data Bern et al. (2004) Price Brand Coffee Type (Blend, Natural, Special) Promotional Discount Consumer Type (Regular or Occasional Shopper) Brand Choice Ground Coffee Testing of Brand choice Logit Models using Consumer Panel Data

PAGE 18

18 Table 2-1 Continued Author Independent Variables Dependent Variables Product Categories Studied Methods Baumgartner (2003) Price Promotion Goodwill Brand Choice Brand Loyalty Ketchup Coffee Testing the Multinomial Logit Model for Time Variations with Brand Choice Using Panel Data. Auger et al. (2003) Basic Product Features (i.e. Weight, Ankle Support, Price) Ethical Features (Tested on Animals, Child Labor, Biodegradable) Consumer Personality Demographics Brand Preference Bar Soaps Athletic Shoes Experiment Singh et al. (2005) Product Attributes (Price*, Feature, Display, Flavor, No Salt/Light*, Pack Sizes*, Brand Names*) Brand Choice Pretzels Potato Chips Tortilla Chips Mayonnaise Sliced Cheese Testing of Multicategory Brand Choice Model using Household Panel Data Charlton and Ehrenberg (1973) Price Purchase Time Purchase Order Product Name Brand Name Brand Choice Beer Experiment

PAGE 19

19 Table 2-1 Continued Author Independent Variables Dependent Variables Product Categories Studied Methods Orth et al. (2004) Brand Name Functional Benefits Price/Value Social Benefit Positive Emotional Benefit Negative Emotional Benefit Consumer Preferences Craft Beer Online Survey from Consumer Panel Data Bentz and Merunka (2000) Marketing Mix Variables (Price per Quantity, Promotional Price Cut as a Percentage of Normal Price) Product Characteristics HouseholdSpecific Variables (Brand and Size Loyalties) Brand Choice Instant Coffee Store Purchases Chocolate Testing of the Multinomial Logit Model in Combination with Neutral Net work Model using Panel Scanner Data Indicates variables were found to be significantly associated with brand choice. Indicates these studies involve d interaction effects among depende nt variables but had no main effects on individual dependent variables. Desired Brand Benefits Researchers do not always account f or separation of effects for brand name with product attributes. Keller (1993) suggests th at the brand name creates adde d benefits separate from the

PAGE 20

20 product for consumers. Benefits are personal valu es that consumers associate with a product or service. It is what the consumer believes the product or service can do for them (Park et al. 1986). Brand benefits create a value by the brand name (i.e. logo, design), which transcends the functional value of the product. Brand benefits fo cus on the needs that the product fulfills for the consumer (Lancaster, 1971; Haley, 1968; Fa rquar, 1989, p. 24; Orth et al. 2004). Some researchers believe that consum ers make purchases based on the product benefits and not the attributes offered. However, consumers evaluate purchase decisions base d on product attributes with the promise of benefits received from pr oduct attributes (Haley, 1968; Aaker et al. 1992; Belch and Belch, 1995; Puth et al. 1999). A r ecent study has shown that consumers do not always seek both attributes and benefits in products. Consumers tend to seek benefits when involved with simple products: low involvement (i .e. food products). Consumers seek attributes when dealing with a more technical product: high involvement (i.e. television, automobile) (Bozinoff and Roth, 1984). More recently research has focused on both bran d name and brand bene fits that led to a brand choice by consumers. Brand benefits have been analyzed in terms of dimensions that impact brand choice. Findings have been discover ed by researching brand benefits that brands outperform and are more actionable than previo us research studies in product attributes. Dimensions that have been researched in the past include performance/quality, value-for-money, emotion, health, social, and environmental bene fits (Orth 2005; Orth et al. 2005). Six brand benefit measures were shown to be significant in measuring brand choice in a previous study (Orth 2005). Performance/Quality Quality refers to th e degree of excellence in a product or service (Xianhua and Germain, 2003). Therefore, quality is one of the most impo rtant factors influencing customer satisfaction

PAGE 21

21 (Fornell et al. 1996) and is considered the ability of a product or service to perform its specific task (Ennew et al. 1993). The success of a brand in customer satisfaction is quality. Companies conform to requirements set by consumers (Berde n et al. 2000). Quality is significant on the performance of a product (Calantone and Knight 2000). The interaction of a product meeting or exceeding consumer expectations based on its perf ormance is how quality is evaluated (Fornell et al. 1996; Reeves and Bednar, 1994). Performance specifications generally define how quality is judged for products (Ennew et al. 1993). Findings from research indicate that marketing strategies, differentiation, cost leadership, and focus are driv ers of quality (Calantone and Knight, 2000). Product quality adds many benefits for a company. Product quality allows companies to charge higher prices to consumers. In a ddition, having a higher product quality gives a competitive advantage which leads to gains in profit margins and market share. However, research has shown that quality may not equate to success without the proper marketing techniques in order to reach and communicate with consumers (Calantone and Knight, 2000; Choi and Coughlan, 2006). Quality is not defined as a situation of spending money to make money. Often times a products quality can be improved by reducing wast e, fewer dissatisfied consumers, and being more efficient in the production of the product. There has been research to support the theory that companies do not have to incur costs to make their product superior in order to have superior quality. Instead, attention to qual ity as a differentiating approach in dealing with competitors often can make a larger overall impact on quality (Calantone and Kni ght, 2000; Berden et al. 2000). Quality is important for impacting brand c hoice because it is the portion of personal risk

PAGE 22

22 that a consumer takes on the decision making processing in evaluating th e purchase of a product (Berden et al. 2000; Hoyer and MacInnis, 2004). Price/Value for Money In retail m arkets, consumers are value driv en, where value is cons idered a tradeoff among price and value. Price can serve as an indicator of quality for consumers. The higher the price of a product, the more perceived risk a consumer incurs (Quester and Smart, 1998). In general, consumers often associate a high-priced retail product with higher quality than those of lower pricing (Lambert, 1972). However, some resear chers believe that this quality and price relationship is too simplistic (Sw eeney and Soutar, 2001). Prices are used by marketers in retail stores in order to appeal to different consumers on different levels. The consumer uses comparative judgments in order to evaluate a potential purchasing decision. The consumer utilizes reference prices in order to make these comparisons (Alvarez and Casielles, 2005). Reference pricing is a subjective price level that is used by the consumers to determine if the product is at an acceptable price for purchase (Mayhew and Winer, 1992). Brands in most product categories have a wide range of different prices. These pric es vary for a vast number of reasons (advertising, lower economies of scale, premium brand positioning, generics, and several other factors). These prices demonstrate information perceived in many different ways by consumers. A consumer might perceive a lower priced product to be considered cheap or having low quality, whereas a different consumer could potentially see the low cost as a good value (Hruschka, 2002; Lambert, 1972). Therefore, price is a major fact or in determining brand choice First, several studies have been conducted in order to determine the effect of price on alcohol c onsumption. Studies have found an inverse relationship for sales and pricin g. For example, as price of alcohol beverages increase, then sales for these products decr ease and vice versa ( sterberg 1995; Levy and

PAGE 23

23 Sheflin, 1983). Second, the consumer wants the best product at the best pric e. Therefore, a higher priced item will have more economic risk, but hi gher priced goods are more visible to others socially. For example, some consumers choose to never purchase generic products because they believe the quality of the product to be inferior. In addition, they ha ve a social fear that others will perceive that they are not economi cally well off (Hoyer and MacInnis, 2004). Emotions Consum ers can develop emotional feelings for products, specifically brands. These emotions toward brands can have a major influe nce based on brand choice. Research has shown that emotions lead to an interaction with th e product on a personal level (Bowlby, 1979; Hazan and Shaver, 1994). These emotions can lead to brand loyalty, paying premiums, and influencing others to purchase the brand. Therefore, a consum ers emotional attachment to a brand may be able to predict their commitment and willingness to make sacrifices to obtain it. Some basic ideals that are associated with this emotional involvement fo r brands are a positive brand attitude, high involvement in the product categor y, brand loyalty (willingness to pay a premium), affection, passion, connection, and the overall satisfacti on associated from the brand (Thompson et al. 2005). Normative/Personal (Social) Social influences consist of influential factors determ ined by family and friends. College students have more of a propensit y to drink the brands that their parents and friends consume on a regular basis. When children leave their parent s home to join the workforce or go off to college, then a majority of them are taking thei r parents purchasing beha viors with them. These behaviors may diminish over time as the young a dult is separated from their family, but the influence is still apparent (Feltham, 1998). In ad dition, adolescents are exposed to peer-pressure and group-think mentalities, which lead them to consuming brands that their friends and peers

PAGE 24

24 consume (Collins et al. 2003). Th is social influence stems from persuasion by attitudes and behaviors of fellow peers (Jessor, 1981; Kande l, 1980). Therefore, normative influences can have an affect on brand choice for the beer product category. Throughout research on social behavior, other individuals be haviors may serve as cues which could increase the potential for behavior. In addition, the behavior of others might remind the i ndividual that altern atives to their own behavior are available (Bandura, 1977). Social influence has an affect on brands that consumers choose. There is a social risk associated with every purchase decision a consumer makes. Opinion leaders, family/friend influence, reference groups, social class, cultur e, and subculture can a ffect the brands that a consumer purchases. This social risk is often associated with what the consumer believes are acceptable brands based on the brand perceptions in the individuals social group. For example, a consumer may purchase a higher pr iced, upscale brand in order to identify and be accepted by a higher social class (Hoy er and MacInnis, 2004). Environment (Stewardship) Stewardship involves a com p any or brand taking an act ive responsibility for the environmental impact of their product. This can come from the de sign, manufacturing, usage from consumers, disposal of the product, and literature of the produc t to stay within the boundaries of government law, i ndustry standards, and consum er standards (Bruen, 2002). In addition, product stewardship involve s the environmental concerns i nvolving health and safety in all phases of the brands life cycl e (Braglia and Petroni, 2000; Hickle and Stitzhal, 2003; White and Pomponi, 2003). Environmental management has become a ma jor issue in today s business world. New green strategies are making companies reth ink the production and wa ste practices involved with company operations (Braglia and Petroni, 2 000; Hickle and Stitzhal, 2003). Brands that do

PAGE 25

25 not exhibit stewardship miss out on green marketing opportunities. Results from a study analyzing the effects of stewar dship concluded, firms that ar e more environmentally aware derive more benefits from their green activities (Braglia and Petroni, 2000). Consumer benefits that come from impl ementing stewardship are enhancing brand reputation and image. In addition, stewardship can gain access to new markets, in some cases reduction of costs (not paying fines and other environmental liabilities), and comply with regulations. Firms that do not practice stewardshi p might have cost and regulatory benefits over firms that do (Braglia and Petroni, 2000). An example of stewardship involving the product category of beer would be if the can or bottle is recyclable or was recycled itself before consumption. Health Historically, there has always been debate ba sed on whether alcohol has been considered a good thing or a bad thing for health purposes. Abraham Lincoln suggested in reference to alcohol, Many were greatly injure d by it, but none seemed to think the injury arose from the use of a bad thing but from the abuse of a very good thing (Basler, 1953, p. 275). Today, consumers are more health consci ous than they have been in previous generations. This trend has made marketers c onform to these health concerns by catering products and brands to meet consumer criteria. Americans obsession with obesity has reached the beer market. Low carbohydrate diets are becoming more and more popular, informing consumers to cut carbohydrates from their diet. Beer marketers have recognized this need and have begun to follow suit by marketing light beers aggressively (Coxe, 2004; Walker, 2004). Research studies have shown that drinking alcohol in moderate amounts has consumer benefits with cardiovascular health through bloo d thinning properties in alcohol. Also, alcohol has been proven to ward off conditions such as b one health, cancer, hot flashes, heart attacks and

PAGE 26

26 ischemic strokes. In addition, some studies actua lly determined that it sh owed protection against dementia (Woods, 2005; Rhodes, 2005; Klatsk y, 2006; Kondo, 2004). In addition, past studies have shown that alcohol consumption can lowe r the risk of developi ng diabetes by improving insulin sensitivity (Is, 2004). Th e main medical health benefit of alcohol is a lowered risk of coronary heart disease (Kondo, 2006; Klatsky, 2006). With these recent consumer benefits, ther e are also risks associated with heavy consumption of alcoholic beverages. There has b een some evidence that drinking beer and other alcoholic beverages in moderati on (one drink a day) can increas e rates of breast cancer in postmenopausal women by 30% (Helliker and Ellison, 2005; Andorfer, 2005). Drinking beer does add calories to a diet, and if these calorie s are not burned off, then the cardiovascular benefits are void (Tufts University Health and Nutrition Lette r, 2004). The beer market has started a shift toward lower ca rbohydrate beers in order to try and meet the needs of these consumers (Beers, 2006). Hops, an ingredient in beer products, pr ovide cancer prevention and treatment of osteoporosis. However, to get these benefits, an individual would have to drink large quantities of beer, which add separate health risks (Lof tus, 2005). Heavy drinking has many consequences associated with it based on the alcohol content in beer. Heavy consumption can lead to liver disease, high blood-pressure, increase the ri sk of cancer, and pancreas damage (Kondo, 2006). Some beer companies are beginning to mention health benefits through marketing efforts. However, it is something that is handled very de licately for liability purpos es. There is research that is supporting health benefits from dri nking in moderation (Cioletti, 2006). There is a perception that wine is healthie r than beer. The University of Western Ontario found that one beer has the same antioxidants as a glass of red wine. Therefore, beer companies are making

PAGE 27

27 efforts to investigate and change these perceptions (Is, 2004). Indicating health benefits to consumers, such as promotion of drinking in moderation as a health benefit and low carbohydrate beers, influences a brand choice de cision for health conscious consumers (Coxe, 2004; Walker, 2004; Woods, 2005; K ondo, 2002). Many of the research studies involving health benefits of beer are still in the exploratory st ages and have not produced consistent results in some areas. Therefore, companies and researchers are not suggesting for consumers to begin beer consumption for health benefits. Instead th ey are merely pointing out that there are some health benefits for those that curre ntly consume beer (Klatsky, 2006). Situational Factors Benefits sought out by consum ers can differ base d on the situation that the consumer is in (Yang et al. 2002). According to Belk (1974), S ituations may be defined as those factors particular to a time and place of observation whic h have a demonstrable and systemic effect on behavior (p. 157). Consumers evaluate brands in different manners based on the situation (Vazquez et al. 2002). It is suggested from previous research th at situational factors are a better predictor for consumer behavior than measures involving consumer attitudes. Research has indicated that consumer preferences change accord ing to their environment (Quester and Smart, 1998; Lai, 1991, Belk, 1974). According to Lai (1991), there ar e three types of situations that are used in marketing strategy among situational fact ors: communication situation, purchase situation, and consumption situation. Situational drivers shoul d have a frequent number of customers per situation. In addition, each situati on must be clearly different than the other in order to account for variance measures. Therefore, effects from environmental factors are not homogenous but rather heterogeneous (Miller and Ginter, 1979; Yang et al. 2002).

PAGE 28

28 A consumer might choose a brand based on being in different situations and will therefore, be motivated to drink a certain brand (Yang et al. 2002). Acco rding to drinking studies, around 80% of young peoples total alcohol consumption o ccurs at a public place (Knibbe et al. 1991). The greatest occurrences of drinking are in the home or in ba rs (Wilks and Callan, 1990). In addition, heavy and light drinkers tend to drink twice as much during happy hours in bars than they do during times that are not involved in such promotions. Theref ore, there are some interaction effects of brand bene fits based on situational factor s (Babor et al. 1978; Orth, 2005). Consumers may face similar environments, but there are several motivating conditions that play a role on brand choice depending on the cons umer (Yang et al. 2002). Several studies have shown this idea of situational infl uences proving that individuals pr efer to drink different brands based on different occasions (Bearden and Etzel, 1982). For example, Quester and Smart (1998) used the purchase of a bottle of red wine for a dr ink during the week (alone or with ones family) over dinner, for a dinner party at a friends house on a weekend (with 5 to 6 close friends), and as a gift for an employer or respected friend. Orth (2005) evaluated three different situations based on drinking red wine with the same scale from Quester and Smart. Miller and Ginter (1979) explored situational impacts on brand choice with respect to fast food rest aurants. The situation variations analyzed were lunc h on a weekday, snack during a s hopping trip, evening meal when rushed for time, and evening meal with the family when not rushed for time. All of the studies involving situational factors demonstrated significance based on impacting brand choice (Orth, 2005; Miller and Ginter, 1979). Areas that have been studied with situati onal drivers include product involvement, brand choice, and product attributes. High product involvement was consider ed a factor that influences behaviors with the interaction of situational drivers. Product fact ors have different levels of

PAGE 29

29 importance to consumers based on situation. Brand choice has been found to be impacted significantly by situational fact ors (Orth, 2005; Quester and Smar t, 1998; Miller and Ginter, 1979; Yang et al. 2002). It is important for marketers to understand wher e brands are effective in given situations. This gives marketers insights as to where the brands are being effectively communicated, purchased, and consumed (Mille r and Ginter, 1979; Quester a nd Smart, 1998). However, one study has argued with these notions Results from a research study using a probability models to determine preferences indicated th at marketers do not have to ma ke their brands congruent to consumers or their environment. It is suggested that the source of bra nd preferences must be understood in order to have an impact on situationa l factors that influence brand choice (Yang, et al. 2002). Situation variation depends on the product catego ry used for research (Belk, 1974). Beer is an important category to use because it is a narrowly defined product category in accordance with researching situatio nal drivers (Miller and Ginter, 1979). Drinking beer is considered an activity that may occur in distinct situations Therefore, there should be a clear variance according to their changing environment (Yang et al. 2002). Consumer Factors Marketers and researchers have studied consum er factors in order to have an understanding of what characteristics and traits (such as consumer demographics susceptibility to interpersonal influence, product category involvement) impact purchasing decisions. In addition, consumer factors are utilized to target and segment popul ations (Park and Lessig, 1977; Bearden et al. 1989; Quester and Smart, 1998). Consumer behavi ors (such as explorat ory behavior, product usage, and frequency of purchase) have been re searched in the past in order to have an understanding of choice (Raju, 1980, Redman et al. 1987; Uncles and Ehrenberg, 1990). These

PAGE 30

30 variables have been linked histor ically in research as potential drivers of situational variation based on brand choice with brand benef its (Orth, 2005; Orth et al. 2004) Exploratory Shopping Behaviors It is im portant to understand brand switching an d exploratory behaviors in consumer brand choice decisions. Previous rese arch has indicated a link betw een situational drivers and behaviors based on personality tr aits of a consumer. According to research by Raju (1980), consumers take a risk, seek variety, and have curiosity in purchase be haviors such as brand switching. Risk taking involves the consumers n eed for innovation and a lternatives in which they are not familiar with. There is more risk pe rceived with this behavior. When a consumer is variety-seeking, they are looking for alternatives that they are familiar with. The final consumer exploratory tendency is curiosity -motivated behavior, which the consumer seeks out information about a product or service through shopping or interpersonal comm unications (Raju, 1980; Wahlers et al. 1986). Product Categories such as beer are a good fit for exploratory behaviors based on the desire for variety and brand-related factors. C onsumers begin to have boredom with a brand and seek alternatives in product categories similar to beer. In extreme cases where consumers have a high level of involvement, these exploratory behaviors may not exist. For example, if the consumer has a level of brand loyalty and habitu ally purchases the same brand irrelevant of factors influencing choice, then this consumer will most likely not seek alternative brands (Van Trijp, 1994; Van Trijp et al. 1996). Interpersonal Influence Social influence is m ajor driver of brand choi ce. However, it is important to understand an individual consumers susceptibility to social influence in order to have an understanding of how much social influence impacts their purchase d ecisions. If there is a high degree of social

PAGE 31

31 influence, then the consumer could potentially change perceptions and purchasing behaviors (Batra et al. 2001). Interpersonal in fluence is the individual traits or characteristics that impact social influence within an indi vidual (McGuire, 1968). Consumer susceptibility to interpersonal influence has an impact on brand choice behaviors (Witt, 1969; Stafford and Cocanougher, 1977). Previous research indicate s that interpersonal influence has a direct effect on personal behaviors (McGuire, 1968). There ar e varying degrees of social im pact within each individual consumer. These levels indicate how much social influence an individual is susceptible to in contrast with other individua ls (Bearden and Etzel, 1982). Consumption Behaviors There are several b ehavioral factors that pl ay a role in determining brand choice for consumers. Product usage is among one of these f actors and plays a major influential role in impacting consumer behavior (Ram and Jung, 1989) Product usage consists of two dimensions: usage variety and usage frequency (Zaichkows ky, 1985). Variety usage is how the product is used and depends upon the product category and situation. Market share for product brands could increase based on an event or situation. For example, sales and volume for specific brands of beer could fluctuate before and after the super bowl (Ram and Jung, 1989). Therefore, brand choice measurements should take into account fo r temporal changes in brand choice behavior (Wilkie, 1986). There have been studies that ha ve compared the differences in drinking consumption of males and females. These studies included usage measures consisting of quantity per occasion, average volume, and frequency of drinking (Green et al. 2004). According to previous studies, there are three categories of drinkers in order to classify users: heavy, moderate, and light users (Redman et al. 1987). Frequency of purchases, the second dimension of product usage, deals with the amount of a single item purchased during a given time perio d. According to a study, frequency of purchases

PAGE 32

32 can provide insight on brand choice (Ram and Jung, 1989; Uncles and Ehrenberg, 1990). In addition, expenditures on the product category its elf also have some insights based on how consumers select a brand based on a product category (Orth, 2005). Product Category Involvement Park and Mittal define involvem ent as a state of mental readiness that impacts cognitive resources for an action, object, or decision with consumer consumption (1985). Product category involvement has a major effect on consumer decision making. If a consumer feels strongly positive about the product category, then they are mo re prone to seek increased value, pay more attention, and try to find the most product be nefits among the product category (Quester and Smart, 1998; Richins and Bloch, 1986). Product category involvement has been shown to demonstrate considerable influence over consumer decision processing (Laurent and Kapferer, 1985). Demographics De mographic variables have been proven to be indicators for brand c hoice. Factors such as age and gender play a role in how consumers eval uate and ultimately purch ase brands in several different product categories (Walsh and Mitchell, 2005). Based on studies involving demographics and drinking behaviors, males tend to drink in larger quantities in same sex groups, whereas women drink with mixed crowds or with a male (Hartford et al. 1983). Age is also a variable to be explored for college student s because there are those of legal age and others that are obtaining beer illegally. There are a numbe r of these college students that purchase beers illegally via a false ID or by having an older pe er purchase it for them (Schwartz et al. 1998). In addition, there is very little known about demographi c issues such as gender, age, and education (year in college) with particular subject matter as it relates to this segment and brand choice.

PAGE 33

33 Theory of Reasoned Action Ajzen and Fishbeins (1980) Theory of Reasoned Action (TRA) is one of the most researched models that describe s the psychological pro cesses of decision making. It is comprised of three main components in order to predic t behavior. The three components are attitude, subjective norms, and intention. This model has been applied to many different areas of study such as alcohol, marijuana, and purchasing c onsumer products (Eagly and Chaiken, 1993). In this model, attitude involves the positive or negative associations an individual has on specific behavior. Subjective norms deal with the normative and social influences that impact an individuals behavior. Social influence on an individual a nd susceptibility to interpersonal influence are factors that measure subjective norms. In a given population, there may be cases that lean more towards attitude providing more influence in terms of behavior. However, in other cases, subjective norms might potentially lead to a different behavior (T rafimow and Fishbein, 2001). Other influential factors could be intrinsic and extrinsic. They result from situational and/or interpersonal factors (Chatzisarantis and Biddle, 199 8; Bagozzi et al. 1992). The two main factors involved in TRA, atti tude and subjective norms, lead to intention. Intention is the likelihood of completing a ce rtain behavior, and the relative importance of normative influence and attitudi nal considerations. Intention is utilized for understanding judgment based on how a final decision is made (Ajzen and Fishbein, 1980). Consumer factors such as demographics and consumption beha viors provide an unders tanding of intention. Intention can give marketers an idea of how a co nsumer will behave toward particular brands (Bagozzi et al. 1992). According to previous research studies, other variables aside from attitude and subjective norm can have an overall impact on behavior (T rafimow and Fishbein, 20 01). Susceptibility to interpersonal influence and social influence lead to subject norms in an individual. Quality, price,

PAGE 34

34 emotion, environment, health benefits, and product category involvement deal with the individuals attitude toward th e brand. In addition, importance of subjective norms and attitudes can vary depending upon the situation (Bagozzi et al. 1992). All of these components, either weighing more heavily on subjective norm or attitude, lead to intention. This intention results in an individual beer brand choice and beer consumption behavior. There have been several studi es involving the TRA model a nd alcohol research studies. These studies involved predicting alcohol consumption behavior (OC allaghan et al. 1997; Trafimow, 1996; Wall et al. 1998). Most of these particular studi es were used in efforts to understand and predict drinking be haviors prevalent among college students. These studies were ultimately used in order to curve drinking behaviors. The TRA model has been utilized in this study to conceptualize research questions involving beer brand choice and beer consumption behaviors. Figure 2-1 illustrates the TRA model used for this study. The model has been extended in order to demonstrate all measures involved in this study. This modified model is exploratory in nature in order to gather a theoretical understandin g among variables used in the study. The model lists the organization of variables as they relate to the con cepts and relationshi ps in the model.

PAGE 35

35 Figure 2-1 Theory of Reasoned Action: Brand Choice/Consumption of Beer. [Reprinted with permission from Ajzen and Fishbein (1980). Theory of Reasoned Action. Master thesis (Page 37, Figure 2-1). University of Florida, Gainesville, Florida Susceptibility to Interpersonal Influence Social Brand Benefits Subjective Norms Intention Attitude Price, Quality, Emotion, Environment, Health, and Product Category Involvement Behavior Beer Brand Choice Beer Consumption Consumer Factors (Demographics, Exploratory Shopping Behaviors and Consumption Behavior) Situational Variation

PAGE 36

36 Research Questions Based on the previous sections, research ques tions were developed in order to have a theore tical understanding of the relationship between brand c hoice, brand benefits, interpersonal influence, consumption behaviors, situati on, product category invol vement, exploratory behaviors, and demographics. RQ1: How does beer brand choice vary by situation? RQ2: How does beer consumpti on behavior vary by situation? RQ3: Which factors (percei ved brand benefits, exploratory shopping behaviors, susceptibility to interpersonal influe nce, product category involvement, and demographics) have the most signifi cant impact on beer brand choice? RQ4: Which factors (percei ved brand benefits, exploratory shopping behaviors, susceptibility to interpersonal influe nce, product category involvement, and demographics) have the most significant impact on beer consumption behaviors?

PAGE 37

37 CHAPTER 3 METHOD Pretest A pretest was administered in order to improve accuracy of measures and to test reliability of pre-existing measures (Appendix A). Brand choice, consideration set formulation, situation, and brand benefits were all aske d to respondents with a predetermined brand, Budweiser. This brand was chos en based on its awareness, popul arity, and market share in the beer product category. Beer cons umption questions and demographics were asked in order to gain knowledge in these areas be fore developing the main study. Subjects Participants for the pretest were selected from two Junior/Senior-level courses. These courses were an advertising course and a health communications course at the University of Florida. College students in these courses were offered extra credit in exchange for their participation (N =46). The pretes t results provided a profile of college student participants. The average age of respondents was 21.61 (standard deviation = .95) The minimum age was 20, and the maximum age was 25 years old. Out of the sample of 46, 30 were female (65.2%) and 16 were male (34.8%). Among students, 27 were employed (58.70%) and 19 were unemployed (41.30%). The respondents had little Greek membership from this sample. There were 38 students that were not Greek members (82.6%) and 7 students were Greek members (15.2%). Based on past month beer consumption, the top five brands in the pretest were Bud Light, Blue Moon, Corona, Coors Light, and Miller Lite. Participants also answered questions regarding beer brands that they were most likely to consume in the next month. These brands were ranked, and the top brands were the exact same as the past beer consumption rankings. These beer brand choice questions created some confusion among respondents. Therefore, it was

PAGE 38

38 determined to make beer brand choice an openended question for the main survey in order to account for individual considerati on sets. Based on an opened-ended situational question, there were 20 situation items resulting from this que stion. Brand benefits scales had satisfactory reliability and were utilized for the main study. However, two health item questions were added to the health benefits scale based on explorator y data from an open-ende d question. These health items added were It is a good way to relieve your stress and It glor ifies unhealthy drinking behaviors. These questions we re evaluated based on agreement or disagreement on a 7-point Likert scale. Beer consumption behavior que stions provided accurate and clear results to measure the construct. Design and Procedure A paper questionnaire w as used in order to co llect data. The survey was administered and collected during class time. Subjects completed separate forms in order to keep their anonymity. Subjects were given 15 minutes to complete th e survey. Students were asked regarding the constructs of brand choice, cons umption behavior, demographics, si tuational variation, and brand benefits. Measures Brand Choice Brand choice questions (Appendix A: Question 1 and 13) were asked with com prehensive but not exhaustive lists of 31 b eer brands. In addition to thes e choices, the respondent had the option of choosing an Other category and if so, then to specify th e brand. A None of the Above option was used to account for non-beer dri nkers. In the first question, respondents were asked to check the brands of beer they have c onsumed in the past month. This question was used in order to get a perspective of consideration set for participants. In the second brand choice

PAGE 39

39 question (13), respondents were asked to rank order the brands of beer they were most likely to consume. Situational Variation Situations (Appendix A: Question 3) were asked on a 7-point Likert scale based on where a respondent would consume Budweiser beer (1 being Not Very Likely and 7 being Very Likely). Eight factors were used, exploratory in nature, to get an idea of discovering strength of situations among college students. In conjunction with this information, an open-ended question was asked (Appendix A: Question 4) regarding other occasions or situ ations that were not listed. Desired Brand Benefits Previous brand benefit scales (Appendix A: Questions 6, 8-12) used by Orth (2005) were tested for their reliability. Measurement for desi red brand benefits were based on six dimensions: quality/performance, price/value for money, social/normative, emotion, environment, and health benefits (Orth 2005; Vasquez et al. 2002; Sweeney and Soutar, 2001; Orth et al. 2004). Quality/performance consisted of 6 items. Price/value for money consisted of 4 items. Emotions were measured with 4 items. Social/normative sc ales consisted of 11 total items. Environmental brand benefits were measured with 3 items. Health benefits consisted of 2 items. All of these items were used on a 7-point Likert scale involvi ng the agreement of disagreement of statements involving each benefit (1 = Str ongly Disagree and 7 = Strongly Ag ree). Scale items were tested for reliability and confirmed to be used in the main study based on their results. An open-ended health benefit question was asked in order to potentially add more items to the health scale in order to increase reliabilit y (Appendix A: Question 7). Consumption Behavior Two questions were used to have an unders tanding of college student beer consumption behavior (Appendix A: Question 2 and 5). These two questions were inquiries regarding number

PAGE 40

40 of days drinking (per week) and quantity of drinks per occasion. These two values provided product usage information. Demographics Demographic questions (Appendix A: Questions 14-20) were asked in order to have an understanding of the respondent profile befo re the main study is conducted. Age, sex, employment, Greek membership, living situation, acad emic year as a student, and college major were asked to gain insight into college student respondents. Main Study Subjects Participants for this study were selected am ong college students in Public Relations and Advertising courses at the Universi ty of Florida. College students that participated in the study were rewarded extra credit for their completion of survey materials. A sample of 222 participants was used to complete the main survey. The total number of brands chosen for this study was 38 different brands of beer (Table 3-1). Among these brands, it was determined that with a 95% confidence interval (Equation 3-1) only th ree brands would be used for this study: 95% CI = +/1.96 ( /N) (3-1) The three brands extracted from the main study were the following: Blue Moon, Bud Light, and Corona. Based on participants only ch oosing the three brands, in conjunction with subjects that did not have cons istent beer consumption behavior (have not consumed a beer in excess of the last two weeks), the sample size was reduced to 98 participants. Based on these 98 participants, the average age was 20.69 (M inimum = 18, Maximum = 31, and Standard Deviation = 2.28). A freque ncy table demonstrates the results from these demographic questions (Table 3-2). The gender distri bution was 75.5% females and 24.5% males. There were 55.1% of respondents that were not employed and 44.9% were employed. There were 30.6% of Greek

PAGE 41

41 membership and 69.4% that were not Greek memb ers. Most of the college students (63.3%) lived with a roommate or roommates in an apartment or house off campus. Table 3-1 Beer Brand Choice (N = 222) Beer Brand Frequency Percent Sam Adams 1 .5% Busch Light 1 .5% Bud Light 43 19.4% Michelob Ultra 11 5.0% Stella 3 1.4% Coors Light 7 3.2% Heineken 12 5.4% Budweiser 9 4.1% Corona 28 12.6% Miller Lite 13 5.9% Blue Moon 27 12.2% Yuengling 8 3.6% Heineken Light 1 .5% Natural Light 8 3.6% Amber Boch 4 1.8% Guinness 3 1.4% Bud Select 2 .9% Killians 2 .9% Sapporo 1 .5% Hardcore 1 .5% Shipyard Blueberry Ale 1 .5% Labatt Blue 1 .5% New Castle 2 .9% Red Stripe 8 3.6% Presidente 2 .9% Smirnoff 2 .9% Khalik Gold 1 .5% Pabst Blue Ribbon 1 .5% Michelob Light 1 .5% Milwaukees Best 1 .5% Hornsby 1 .5% Miller High Life 1 .5% Grolsch Light 2 .9% Goldwesser 1 .5% Shiner Bock 1 .5% Corona Light 1 .5% Molson 1 .5% Coopers 1 .5% Total 214 96.4%

PAGE 42

42 Table 3-2 Demographic Frequency Distribution Frequency Percentage Gender Male Female 24 74 24.5% 75.5% Age 18 19 20 21 22 23 24 27 29 30 31 5 24 29 19 10 2 5 1 1 1 1 5.1% 24.5% 29.6% 19.4% 10.2% 2.0% 1.0% 1.0% 1.0% 1.0% 1.0% Employed Yes No 44 54 44.9% 55.1% Greek Membership Yes No 30 68 30.6% 69.4% Academic Year Freshman Sophomore Junior Senior Masters Graduate Student 10 39 28 20 1 10.2% 39.8% 28.6% 20.4% 1.0% Academic Major Public Relations Advertising Journalism Other 25 27 2 44 25.5% 27.6% 2.0% 44.9% Living Situation Live with Parents Live Alone (apt/home) on Campus Live Alone (apt/home) off Campus Live with Roommate(s) (apt/home) on Campus Live with Roommate(s) (apt/home) off Campus Fraternity or Sorority House Live in Dorm Alone Live in Dorm with Roommate(s) Other 1 1 9 8 62 6 2 8 1 1.0% 1.0% 9.2% 8.2% 63.3% 6.1% 2.0% 8.2% 1.0% N = 98 Design and Procedure A paper questionnaire w as used in order to coll ect data for the main study. The survey was administered and collected during class time. Subjects completed separate forms in order to keep their anonymity. Subjects were given 30 minutes to complete the study. Students were asked

PAGE 43

43 brand choice, consumption behavior, demographic, susceptibility to interpersonal influence, product category involvement, exploratory shopping behaviors, situational variation, and brand benefit questions. Measures Brand Choice To measure brand choice, respondents were asked a fill in a blank question (Appendix B: Question 1) asking them to write in their favorite brand of beer. This information allows the respondent to select their favor ite brand among that particular individuals consideration set when it comes to the pr oduct category of beer. Situational Variation Based on pretest results, it was discovered that 20 situations existed (Appendix B: Question 2) in which college students consum e their favorite beer. M odeling for situational variation was developed from Orths study (2005) Consumption situations were varied by the company of the individual dri nking. Situations were varied w ith three possible scenarios: drinking beer with a group of friends, drinki ng with a date, and drinking alone. Among the 20 situations possible, 13 situations were chosen by respondents in the main study (Table 3-3). Six groupings were created with multiple categories in an attempt to limit the disparity among situation data. Each grouping d ealt with a respondent answering what situation they last consumed their favorite beer. Groupings were asse ssed in order to find a significant relationship based on similarities of the situation, frequencies of situations selected, and social interaction based on situations. These situational groupings can be seen in left column on Table 4-1. Group 1 consisted of 5 categories: in a bar or club with friends, at home with friends not a party, a house/apartment party not in your home with friends, partying at home or pool/beach, and other. At the University of Florida, most housing for college students has a pool. Therefore,

PAGE 44

44 party at home was combined with pool/beach. This combination was utilized in order to distribute the data more evenly among the ca tegories. The other category contains eight situations to contrast cate gory size among these categories. Group 2 consisted of 4 categories: in a bar or cl ub with friends, at home with friends not a party, a house/apartment party no t at home with friends, and ot her. The first three categories remained unchanged. However, the other category cons ists of ten situations combined in order to contrast with the high numbers select ed among the first three situations. Group 3 was made up of in a bar or club, a restaurant or at home not partying, and partying/special events. This first category consis ts of an environment when college students go out to a bar or club and have the social interactions a nd situations that go along with the bar/club scene on a college campus. In addition, the category of restaurant and at home not partying deals with the college student that want s to relax and have a few drinks in a comfortable situation. The final category deals with a special event such as a party, beach/pool, or sp orting/concert. College students participate in activities throughout the school year such as tailgating at football games, music concerts, pool parties, and the occasional house/apartment party. In these environments, students are attending with the intent of having a good time and enjoying these special events. Group 4 contains two categories. The first category deals with co llege students drinking in a bar, club or party situation. These student s are looking to have a good time and social interaction. The second category invo lves students not in bars, clubs, or parties. These students are in a more relaxed, comfortable environmen t where there may not be as much social interaction or pressure to consume large quantities of beer. Group 5 has four categories consisting of soci al, date, alone, and relax with friends. The social category contains situati ons where there is high social interaction. The date situation is

PAGE 45

45 when the respondent was drinking beer with a da te. The alone category dealt with situations when the individual drank beer by themselves. Relax with friends was used in order to demonstrates college students that were just drinking their favorite beer to relax and not necessarily drinking to have a good time or socially interact. Group 6 involved the same social, date, and alone categories as lis ted in group 5. Data from relax with friends was omitted from this group. The three respondents that chose other on the main survey were also omitted from groups 5 and 6 based on a lack of frequency for this selection. Table 3-3 Situation Frequency Distribution Situation Frequency In a bar or club with friends 35 In a bar or club with a date 1 In a bar or club alone 1 In restaurant with friends 4 At home with friends (not a party) 17 At home with a date (not a party) 2 At home alone (not a party) 2 Sporting event/concert with friends 1 Beach/pool with friends (not a party) 8 Party (not at home) with friends 20 Party (not at home) with a date 2 Party at your home 3 Other 3 N = 97

PAGE 46

46 Desired Brand Benefits Measurement for desired brand benefits (A ppendix B: Questions 5-10) was consistent with items in the pretest. However, respondents were answering statements for their favorite beer and not a forced choice (Budweiser as used in the pretest). Brand bene fit index scores were calculated by using averages for each of the si x brand benefit dimensions. Scale items were tested for reliability. Quality/performance c onsisted of 6 items (Chronbachs alpha = 0.89). Price/value for money consisted of 4 items (Chr onbachs alpha = 0.85). Emotions were measured with 4 items (Chronbachs alpha = 0.94). Social/normative scales consisted of 11 total items (Chronbachs alpha = .90). Environmental brand benefits were measured with 3 items (Chronbachs alpha = 0.86). Health benefits co nsisted of 4 items (Chronbachs alpha = .50). When one item was removed, It glorifies unhe althy drinking behaviors, then reliability improved (Chronbachs alpha = .66 ). Finally, another item was re moved (It is a good way to relieve your stress) in order to improve re liability scores (Chr onbachs alpha = 0.74). Exploratory Shopping Behaviors Exploratory shopping behavior s (Appendix B: Questions 1113) are a measure that was adopted from many previous measures (Raju, 1980; Wahlers et al. 1986). The constructs used in this study for exploratory shopping behavior were risk-taking, variety-seeking, and curiositymotivated (Orth, 2005). Items were measured on a 7-point Likert scale based on agree with statements (1 = Strongly Disagr ee and 7 = Strongly Ag ree). Index scores we re calculated for each of the three exploratory shopping behaviors by using averages. Among these measures, risk taking behavior consisted of 5 items (Chronbachs alpha = 0.83). Variety-seeking behavior was made up of 4 items (Chronbachs alpha = 0.79). Curiosity-motivated behavior was a measure made up of 3 items (Chronbachs alpha = .47). Ho wever, one item was removed (When I hear

PAGE 47

47 about a new store or restaurant, I take advantage of the first o pportunity to find out more about it) in order to improve reliabil ity (Chronbachs alpha = 0.53). Interpersonal Influence Interpersonal influence measures were 12 items (Appendix B: Ques tion 14) used from previous research in order to measure the leve l of social influence within each individual (Bearden et al. 1989). Items were measured on a 7-point Likert scale based on agree with statements (1 = Strongly Disagr ee and 7 = Strongly Agree). A su sceptibility to interpersonal influence index score was calculated by using an average of the 12 items. These 12 items were found to be reliable (Chronbachs alpha = 0.89). Consumption Behavior Beer consumption behavior consisted of th ree questions. The firs t question (Appendix B: Question 3) was a question inquiring about basic beer consumption. This question was utilized to discover respondents that were not ordinarily beer drinkers and remove them from the sample. Beer consumption behavior was measured by cr eating a consumption index from two separate measures (Appendix B: Questions 15 and 16). The first measure was the amount of beer on average that an individual drinks in one occasion. The second m easure consisted of the average number of days a week in which an individual drinks. These two measures have been illustrated in a cross tabulation on Table 3-4. These measures were multiplied in order to cr eate an index score. Index scores were evaluated and grouped into light, moderate, and h eavy beer drinkers based on their consumption behaviors. These groups were di vided equally into 3rds base d on the frequency distribution illustrated on Table 3-5. Light beer drinkers ha d a drinking index of 0 2. This means that a college student that had a consumption index of 2 drank either 2 twelve ounce beers in one night on average in a week, or they drank one beer on average for two nights a week. Moderate beer

PAGE 48

48 drinkers had consumption indexes from 3 8, and heavy beer drinkers had consumption indexes of 9 54. It is important to note that these beer consumption behavi or levels differ in contrast to the general population. Due to high beer consumpti on levels of this par ticular college market, these measures should not be used for direct comparisons involving consumption behavior for other beer markets. Table 3-4 Average Beers Consumed and Average Occasions Drinking Beer Average Number of Occasions per Week Drinking Beer Average Number of Beers Drank per Occasion 0 1 2 2.5 3 4 5 6 Total 0 7 0 0 0 0 0 0 0 7 1 6 4 0 0 0 0 0 0 10 2 4 7 2 0 1 0 0 0 14 3 3 5 10 0 2 0 0 0 20 4 1 7 3 0 5 1 0 0 17 5 0 4 3 1 1 0 1 0 10 6 1 0 3 0 2 1 0 0 7 7 0 1 0 0 3 0 0 0 4 8 1 1 0 0 0 1 0 0 3 9 1 1 0 0 0 0 0 1 3 10 0 0 0 0 0 1 0 0 1 14 0 0 0 0 1 0 0 0 1 17+ 0 1 0 0 0 0 0 0 1 Total 24 31 21 1 15 4 1 1 98

PAGE 49

49 Table 3-5 Consumption Behavior Frequency Consumption Behavi or Frequency Percent (%) Light Beer Drinkers 0 1 2 24 4 7 24.5 4.2 7.1 Total 35 35.7 Moderate Beer Drinker 3 4 5 6 7 8 5 9 4 11 1 4 5.1 9.2 4.1 11.2 1 4.1 Total 34 34.7 Heavy Beer Drinker 9 10 12 12.5 15 16 17 18 21 24 25 32 40 42 54 3 3 8 1 1 1 1 2 3 1 1 1 1 1 1 3.1 3.1 8.2 1 1 1 1 2 3.1 1 1 1 1 1 1 Total 29 29.6% N = 98 Product Category Involvement Three items (Appendix B: Questi on 17) used from previous research were used in order to measure product category involvement (DeW olf, 2001). Items were measured on a 7-point Likert scale based on agree with statements (1 = Strongly Disagree and 7 = Strongly Agree). Index scores of the three items resulted in satisfactory reliab ility (Chronbachs alpha = 0.94).

PAGE 50

50 Demographics Demographic questions used from the pretest were also utilized in the main study in order to retrieve information in order to profile a nd identify respondents (Appendix B: Questions 1824).

PAGE 51

51 CHAPTER 4 RESULTS Brand Choice among Situational Variation Chi-square analysis was used in order to explore the possible association between beer band choice and the situation in which college st udents consumed their favorite beer. All data was tested at level of p .05. Beer brand choice was a categoric al variable consisting of three brands: Bud Light, Blue Moon, and Corona. Situation was a categoric al variable consisting of six groups with categories for each s ituation. Table 4-1 illustrates the cross tabulation of these results for the chi-square value, frequencies among groups with brand choice, and degrees of freedom. In grouping one with five situations, 45.71% that chose drank in a bar or club with friends chose (16 out of 35). Bud Light was also chosen more when at home with friends not at a party (41.18%, 7 out of 17), party not at home with friends (55%, 11 out of 20), and other (50%, 7 out of 14). When college students drank beer partyi ng at home or pool/beach area, then Corona was the highest chosen (63.64%, 7 out of 11). These results were found to not be statistically significant. Therefore, there is no significant re lationship between categories in group one with brand choice. Group two with four situations yielded some what similar results as group one. College students drinking in a bar or club chose Bud Light (45.71, 16 out of 35). Bud Light was most chosen in all categories: bar or club with friends (45.71, 16 out of 35), at home with friends not a party (42.18%, 7 out of 17), party not at home w ith friends (55%, 11 out of 20), and other (36%, 9 out of 25). Results were found to not be stat istically significant and no relationship for these four categories and brand choice exists.

PAGE 52

52 In grouping three with three situ ations, results were not statis tically significant. College students that drank in a bar or club preferred Bud Light (45.95%, 17 out of 38). College students that drank in a restaurant/home not partying pref erred Blue Moon (39.29%, 11 out of 28) slightly over Bud Light (35.71%, 10 out of 28). In a partyi ng/special event, college students chose Bud Light (50%, 16 out of 32) closer ove r Corona (37.5%, 12 out of 32). In grouping four with two situations, college students in a pa rty, bar, or club were more prone to choose Bud Light (50%, 30 out of 60). College students not in a party, bar, or club drank Bud Light (35.14%, 13 out of 37), Bl ue Moon (32.43%, 12 out of 37), and Corona (32.43%, 12 out of 37). Results for grouping 4 wi th two situations indicated no statistical significance. In grouping five with four situat ions, college students in soci al situations drank Bud Light more frequently (46.15%, 30 out of 65). College students that drank thei r favorite beer with a date preferred Bud Light (80%, 4 out of 5). College students that drank their favorite alone were equally distributed among brand choice at 33.33% (1 out of 3) for each brand. College students that drank their favorite beer to relax with friends chose Blue Moon (42.86%, 9 out of 21). These results did not provide st atistical significance. In grouping six with three situations, results we re exactly the same as group five other than omitting the relax with friends category. Ther efore in group six, Bud Light was the most dominantly chosen brand for these situations, but as in group five, results were also found to not be statistically significant. The chi-square analysis indicated that si tuation groups one (x = 12.17, p > .05), two (x = 2.55, p > .05), three (x = 6.15, p > .05), four (x = 2.11, p > .05), five (x = 6.89, p > .05), and

PAGE 53

53 six (x = 3.06, p > .05) were not associated with beer brand choice. Results for all situation groups were found to be not statistica lly significant with brand choice. Table 4-1 Situational Gr oups and Brand Choice Situational Grouping Items Included in Grouping Bud Light Blue Moon Corona Total Bar or club with friends 16 (45.71%) 10 (28.57%) 9 (25.71%) 35 At home with friends (not a party) 7 (41.18%) 5 (29.41%) 5 (29.41%) 17 Party (not at home) with friends 11 (55%) 3 (15%) 6 (30%) 20 At home alone (not a party) Beach/Pool with friends (not a party) Party at your home 2 (18.18%) 2 (18.18%) 7 (38.89%) 11 1 In a bar or club with a date In a restaurant with friends Other In a bar or club alone At home with a date (not a party) Sporting event/concert with friends Party (not at home) with date 7 (50%) 6 (42.86%) 1 (7.14%) 14

PAGE 54

54 Table 4-1 Continued Situational Grouping Items Included in Grouping Bud Light Blue Moon Corona Total Bar or club with friends 16 (45.71%) 10 (28.57%) 9 (25.71%) 35 At home with friends (not a party) 7 (41.18%) 5 (29.41%) 5 (29.41%) 17 Party (not at home) with friends 11 (55%) 3 (15%) 6 (30%) 20 2 At home alone (not a party) Beach/Pool with friends (not a party) Party at your home In a bar or club with a date In a restaurant with friends Other In a bar or club alone At home with a date (not a party) Sporting event/concert with friends Party (not at home) with date 9 (36%) 8 (32%) 8 (32%) 25

PAGE 55

55 Table 4-1 Continued Situational Grouping Items Included in Grouping Bud Light Blue Moon Corona Total In a bar or club with friends In a bar or club with a date In a bar or club alone 17 (45.95%) 11 (29.73%) 9 (24.32%) 37 In a restaurant with friends At home with friends (not party) At home with a date (not a party) At home alone (not a party) Other 10 (35.71%) 11 (39.29%) 7 (25%) 28 3 Sporting event/concert with friends Beach/pool with friends (not a party) Party (not at home) with friends Party (not at home) with a date Party at your home 16 (50%) 4 (12.5%) 12 (37.5%) 32

PAGE 56

56 Table 4-1 Continued Situational Grouping Items Included in Grouping Bud Light Blue Moon Corona Total In a bar or club with friends In a bar or club with a date In a bar or club alone Party (not at home) with friends Party (not at home) with a date Party at your home 30 (50%) 14 (23.33%) 16 (26.67%) 60 4 In a restaurant with friends At home with friends (not party) At home with a date (not a party) At home alone (not a party) Sporting event/concert with friends Beach/pool with friends (not a party) Other 13 (35.14%) 12 (32.43%) 12 (32.43%) 37

PAGE 57

57 Table 4-1 Continued Situational Grouping Items Included in Grouping Bud Light Blue Moon Corona Total In a bar or club with friends Sporting event/concert with friends Beach/pool with friends (not a party) Party (not at home) with friends Party at your home 30 (46.15%) 14 (21.54%) 21 (32.31%) 65 In a bar or club with a date At home with a date (not a party) Party (not at home) with a date 4 (80%) 1 (20%) 0 (0%) 5 At a bar or club alone At home alone (not a party) 1 (33.33%) 1 (33.33%) 1 (33.33%) 3 5 In a restaurant with friends At home with friends (not a party) 7 (33.33%) 9 (42.86) 5 (23.81%) 21

PAGE 58

58 Table 4-1 Continued Situational Grouping Items Included in Grouping Bud Light Blue Moon Corona Total In a bar or club with friends Sporting event/concert with friends Beach/pool with friends (not a party) Party (not at home) with friends Party at your home 30 (46.15%) 14 (21.54%) 21 (32.31%) 65 In a bar or club with a date At home with a date (not a party) Party (not at home) with a date 4 (80%) 1 (20%) 0 (0%) 5 6 At a bar or club alone At home alone (not a party) 1 (33.33%) 1 (33.33%) 1 (33.33%) 3 1: x = 12.17, df = 8, N = 97, p-value =.14 2: x = 2.55, df = 6, N = 97, p-value = .86 3: x = 6.15, df = 4, N= 97, p-value = .19 4: x = 2.11, df = 2, N=97, p-value = .35 5: x = 6.89, df = 6, N = 94, p-value = .33 6: x = 3.06, df = 4, N = 73, p-value = .55 *p-value .05 (%) Percentage of situation for a specific beer brand choice Consumption Behavior among Situation Variation An additional chi-square test was used in order to examine the relationship between consumption behavior and situati on variation. The same six situati onal groupings were used as in the previous analysis. Beer consumption behavior was a categorical variable (beer consumption

PAGE 59

59 levels of light, moderate, and heavy). Table 4-2 i llustrates the results for the chi-square value, frequencies among groups with bran d choice, and degrees of freedom. For group one consisting of five situations, college students th at drank in a bar/club with friends consisted of 40% moderate drinkers (14 out of 35), and 42.67% heavy drinkers (15 out of 35) drank in a bar/club with friends. Responde nts that drank at home with friends were dominantly light drinkers at 52.94% (9 out of 17). College students that drank their favorite beer at a party not in their home with friends were light drinke rs (35%, 7 out of 20) and moderate drinkers (40%, 8 out of 20). Respondents drinking beer partying at home or pool/beach area consisted of light drinkers at 72.73% (8 out of 11). Other situati ons were mostly comprised of moderate drinkers (42.86%, 6 out of 14) that co nsumed their favorite beer in these situations. These results were found to be statistically significant (p .05). In group two with four situations, the first three situations (b ar/club with friends, at home with friends not a party, and at a party not at home with friends) are the exact same as in group one. However, the other sections were added with the 4th and 5th situation from group one. Therefore, in other situations light drinkers at 48% (12 out of 25) consumed their favorite beer. However, unlike group the first grouping, these re sults were not statis tically significant. Therefore, the situations cannot be associated with consumptions behaviors. Group three contains three situations, and college students drinking a bar or club were mostly heavy drinkers (42.24%, 16 out of 37) a nd moderate drinkers (40.54%, 15 out of 37). Respondents that drank their favorite beer in a restaurant/home not part ying were 50% light drinkers (14 out of 28). At a party/special event the light drinke rs were comprised of 43.75% (14 out of 32). These results were statistically significant (p .05), and the situations have associations with the drin king behaviors during them.

PAGE 60

60 Group four has two situations. In a party, bar, or club moderate drinkers at 40% (24 out of 60) and heavy drinkers at 36.67% (22 out of 60) drank their preferred beer. The other group, drinking beer but not in a party, bar, or club consisted mainly of light drinkers at 54.05% (20 out of 37). This data was consid ered to be significant (p .05). Therefore, there is an association with these drinking behavior s and the two situations. Group five contains four situa tions, college students drinking in a social environment were distributed moderately even among consumption be haviors: light drinkers at 29.69% (19 out of 65), moderate drinkers at 38.46% (25 out of 65), and heavy drinkers at 32.31% (21 out of 65). College students that drank with a date were mostly heavy drinkers with 60% (3 out of 5). Students that drank alone were mostly light drin kers 66.67% (2 out of 3) Respondents that drank their favorite beer relaxing with friends were light drinkers (47.62%, 10 out of 21). Group five data was not statistically significan t so no associations can be made. Group six had three situations. These situations come directly from group five. However, the relax with friends situation was omitted form this grouping. Situations in group six were as follows: social, date, and alone. This informati on had the same results as in group five. In addition, this group was also not statistically significant. Chi-square test results for al l situational groupings were as follows: group one (x = 15.50, p < .05), two (x = 10.22, p > .05), three (x = 10.44, p < .05), four (x = 9.67, p < .05), five (x = 6.28, p > .05), and six (x = 4.07, p > .05). Groups one three, and five were significant (p .05). Therefore, associations can be made from these situations and consumptions behaviors. College students that drank in bars, clubs, and part y situation had heavy and moderate drinking behaviors. Students that just st ayed with friends not in a part y situation were light drinkers. Situations including special events and restaurant s exhibited light drinking behaviors as well in

PAGE 61

61 college students. When respondents tended to go to parties not in their home, college students drank in light to modera te drinking behaviors. Table 4-2 Situational Groups and Consumption Behavior Situational Grouping Items Included in Grouping Light Drinker Moderate Drinker Heavy Drinker Total Bar or club with friends 6 (17.14%) 14 (40%) 15 (42.86%) 35 At home with friends (not a party) 9 (52.94%) 4 (23.53%) 4 (23.53%) 17 Party (not at home) with friends 7 (35%) 8 (40%) 5 (25%) 20 At home alone (not a party) Beach/Pool with friends (not a party) Party at your home 8 (72.73%) 2 (18.18%) 1 (9.09%) 11 1* In a bar or club with a date In a restaurant with friends Other In a bar or club alone At home with a date (not a party) Sporting event/concert with friends Party (not at home) with date 4 (28.57%) 6 (42.86%) 4 (28.57%) 14

PAGE 62

62 Table 4-2 Continued Situational Grouping Items Included in Grouping Light Drinker Moderate Drinker Heavy Drinker Total Bar or club with friends 6 (17.14%) 14 (40%) 15 (42.86%) 35 At home with friends (not a party) 9 (52.94%) 4 (23.53%) 4 (23.53%) 17 Party (not at home) with friends 7 (35%) 8 (40%) 5 (25%) 20 2 At home alone (not a party) Beach/Pool with friends (not a party) Party at your home In a bar or club with a date In a restaurant with friends Other In a bar or club alone At home with a date (not a party) Sporting event/concert with friends Party (not at home) with date 12 (48%) 8 (32%) 5 (20%) 25

PAGE 63

63 Table 4-2 Continued Situational Grouping Items Included in Grouping Light Drinker Moderate Drinker Heavy Drinker Total In a bar or club with friends In a bar or club with a date In a bar or club alone 6 (16.22%) 15 (40.54%) 16 (43.24%) 37 In a restaurant with friends At home with friends (not party) At home with a date (not a party) At home alone (not a party) Other 14 (50%) 8 (28.57%) 6 (21.43%) 28 3* Sporting event/concert with friends Beach/pool with friends (not a party) Party (not at home) with friends Party (not at home) with a date Party at your home 14 (43.75%) 11 (34.38%) 7 (21.88%) 32

PAGE 64

64 Table 4-2 Continued Situational Grouping Items Included in Grouping Light Drinker Moderate Drinker Heavy Drinker Total In a bar or club with friends In a bar or club with a date In a bar or club alone Party (not at home) with friends Party (not at home) with a date Party at your home 14 (23.33%) 24 (40%) 22 (36.67%) 60 4* In a restaurant with friends At home with friends (not party) At home with a date (not a party) At home alone (not a party) Sporting event/concert with friends Beach/pool with friends (not a party) Other 20 (54.05%) 10 (27.03%) 7 (18.92%) 37

PAGE 65

65 Table 4-2 Continued Situational Grouping Items Included in Grouping Light Drinker Moderate Drinker Heavy Drinker Total In a bar or club with friends Sporting event/concert with friends Beach/pool with friends (not a party) Party (not at home) with friends Party at your home 19 (29.23%) 25 (38.46%) 21 (32.31%) 65 In a bar or club with a date At home with a date (not a party) Party (not at home) with a date 1 (20%) 1 (20%) 3 (60%) 5 At a bar or club alone At home alone (not a party) 2 (66.67%) 1 (33.33%) 0 (0%) 3 5 In a restaurant with friends At home with friends (not a party) 10 (47.62%) 6 (28.57%) 5 (23.81%) 21

PAGE 66

66 Table 4-2 Continued Situational Grouping Items Included in Grouping Light Drinker Moderate Drinker Heavy Drinker Total In a bar or club with friends Sporting event/concert with friends Beach/pool with friends (not a party) Party (not at home) with friends Party at your home 19 (29.23%) 25 (38.46%) 21 (32.31%) 65 In a bar or club with a date At home with a date (not a party) Party (not at home) with a date 1 (20%) 1 (20%) 3 (60%) 5 6 At a bar or club alone At home alone (not a party) 2 (66.67%) 1 (33.33%) 0 (0%) 3 1: x = 15.50, df = 8, N = 97, p-value =.05* 2: x = 10.22, df = 6, N = 97, p-value = .12 3: x = 10.44, df = 4, N = 97, p-value = .03* 4: x = 9.67, df = 2, N = 97, p-value = .01* 5: x = 6.28, df = 6, N = 94, p-value = .39 6: x = 4.07, df = 4, N = 73, p-value = .40 *p-value .05 (%) Percentage of situation for a specific drinking behavior

PAGE 67

67 Variables Affecting Beer Brand Choice A m ultiple discriminant analysis was used in order to determine a linear combination of continuously measured IVs that best classify cases in one of several known groups. The continuous independent variables us ed in this analysis were br and benefits (quality, price, emotion, social, environment, and health), expl oratory shopping behavior s (risk-taking, varietyseeking, and curiosity-seeking), susceptibility to interpersona l influence, product category involvement, gender, employment, Greek member ship, and age. The categorical dependent variable was brand choice among the three bran ds of Bud Light, Blue Moon, and Corona. In addition, Table 4-3 summarizes the results of multip le discriminant analysis. This table includes unstandarized and standardized coe fficients for both functions, Wilk s Lambda, F-ratio, p-value, for each brand, centroids for both group functions, eigenvalues, and the Wilks Lambda and Chi Square for canonical discriminant functions. In this analysis, two functions were produce d. The first function produ ced a high degree of separation among the second function based on the final Wilks lambda (.48) and canonical correlation (.64). Therefore, there is a moderate to strong correlation between the discriminant function and the independent variables. In addi tion, 73.8% of variance was explained by the first function. The first function was found to be stat istically significant (p = .00, x = 63.90, df = 30). The second function had a Wilks lambda of .80 and a canonical correla tion of .44. The second function was not signif icant (p = .17). Standardized canonical discriminant function coefficients for the two functions were included in Table 4-3. A structure matrix, listed in Table 4-4, wa s used to discover the largest absolute correlation between variables and discrimi nant functions. This structure matrix indicates the greatest impact on the discriminatory impact on the predictor variab les on beer band choice. According to the structure matri x, in function one, risk -taking behavior, social, and age were the

PAGE 68

68 highest three variables correlated with discriminant scores. In f unction two, environment, gender, and price were the highest three variables correlated with discriminant scores. The average discriminant scores for the functi on 1 (centroids) with respect to brands were the following: Bud Light = -.89, Blue Moon = 1.02 and Corona = .38. This means that in the 1st function, while significant will do a good job discri minating between all three beer brands. The most discriminating power existed between B ud Light and Blue Moon. Blue Moon and Corona were more closely related than B ud Light and the other two brands. Table 4-6 illustrates the mean and standard deviation for brand c hoice variables and the independent variables used in the discriminant an alysis. Variables that had statistical significance with Wilks Lambda were the following: pri ce (F-ratio = 6.64, p-valu e = .00), ri sk-taking behavior (F-ratio = 4.08, p-valu e = .02), and environment (F-r atio =3.63, p-value = .03). The mean scores price according to brand choice were the following: Bud Light = 5.23 (standard deviation = .92), Blue Moon = 4.74 (standard deviation = .82), and Corona = 4.38 (standard deviation = 1.13). Mean scores for risk-taking behaviors were the following: Bud Light = 3.70 (standard deviation = 1.26), Bl ue Moon = 4.58 (standard deviation = 1.32), and Corona = 3.98 (standard deviation = 1.13). Mean scores for environment were the following: Bud Light = 3.94 (standard deviation = .95), Blue Moon = 4.27 (standard deviation = .61), and Corona = 3.65 (standard deviation = .84). The most significant variable affec ting Bud Light was price with a mean of 4.23. Price was also the mo st significant variable with rega rds to the other two brands of Blue Moon (mean = 4.74) and Corona (mean = 4.38). The three most important variables in terms of individual variables for Wilks Lambda are price, risk-taking behavior, and environment. These three variables had the smallest Wilks Lambda values of .88 for price, .92 for risk-t aking behavior, and .92 for environment. In

PAGE 69

69 addition, they have the highest F-ratios among the other variables, wh ich are statistically significant (p-value <= .05). Based on the interaction of the indepe ndent variables, price is the most important variable in the first function (standardized coefficient = -.89) in discriminating brand choice in terms of the standardized coeffi cients. The second most im portant variable with interaction for the first function is social (standardized coeffici ent = -.69). However, the social benefit variable was not found to be statistica lly significant (p-value = .06). The third most important variable (standardized coefficient = .49) based on interaction of variables was risktaking behavior which was found statis tically significant (p-value = .02). Based on these results, college students that like the pr ice of their favorite beer are less likely to change beer brands. In addition, if a college student exhib its risk-taking behaviors, then th ey are more likely to change beer brands. Finally, when a college student is in favorite of the e nvironmental features of a beer brand, then the individual is more lik ely to change beer brand choice. Predictive results of this analysis based on the multiple discriminant analysis are illustrated in Table 4-5. According the resu lts, 66.70% of original grouped cas es were correctly classified. According to these results, 83.33% of Bud Li ght, 57.69% of Blue Moon, and 50.00% of Corona were correctly classified. The largest prediction errors occurre d in the Corona brand (12 at 42.86%). The overall hit ratio had a tratio of 6.44 and thus the result ant hit rate was statistically significant ( = .05). Additional analysis was used in order to de termine the most significant contributions to beer brand choice. A stepwise discriminant analys is was used in order to determine the greatest linear combination of continuously measured indepe ndent variables that be st classify cases in one of several known groups. This analyze was us ed in order to limit the assumption that predictor variables are independe nt or noncollinear. In addition, Ta ble 4-7 summarizes the results

PAGE 70

70 of stepwise discriminant analysis. This ta ble includes unstandarized and standardized coefficients for both functions, Wilks Lambda, F-ratio, eigenvalues, p-value, for each brand, centroids for both group functions, and the W ilks Lambda and Chi Square for canonical discriminant functions. In this analysis, two of the fifteen variables were selected when Wilks lambda was at the lowest level. The first f unction produced a high degree of separation among the second function based on the final Wilks la mbda (.77) and canonica l correlation (.44). The second function had a Wilks lambda of .96 a nd a canonical correlation of .21. Both functions were found to be statistically significant (p .05). Based on group centroids for group one, Bud Light = .54, Blue Moon = -.44, and Corona = -.41, the first function discriminates Bud Light from the other two brands: Blue Moon and Bud Light. The second function discriminates between all three brands with the group two centroids of Bud Light at .01, Blue Moon at .28, and Cor ona at -.27. However, the degree of separation among the brands in function two does not discrimi nate as strongly as in function one. In Table 4-7, the two-plot function of group centroids demonstrates this separation among Bud Light, and the two other brands, Blue Moon and Corona. The canonical correlation (R) for the first function is .44. This means that between the independent variables included in the analysis, there is a mode rate correlation between the discriminant function and the independent vari ables. The Wilks Lambda for the canonical discriminant function is .77. This means that 77 % of discriminant scores for variance are not explained by group differences. The Wilks Lambda (.77) and Chi-Square test (23.76) were found to be statistically significan t. The Chi-Square test has a pvalue of .00 with 4 degrees of freedom. Of the variance explained by the two functions, the 1st explains 84.3%.

PAGE 71

71 The canonical correlation (R) for the second f unction is .21. This means that between the independent variables included in the analysis, there is a low correlation between the discriminant function and the independent vari ables. The Wilks Lambda for the canonical discriminant function (2) is .96. This means that 96% of discriminant scores for variance are not explained by group differences. Th e Wilks Lambda (.96) and ChiSquare test (4.01) were found to be statistically significant. The Chi-Square te st has a p-value of .05 with 1 degree of freedom. Of the predictor variables in the second function, explains 15.7% was explained. Standardized canonical discriminant function coefficients for the two functions were included in Table 4-7. The two most important va riables in terms of individual variables for Wilks Lambda are price and risk-taking behavior s. These two variables had the smallest Wilks Lambda values of .88 for price and .92 for risk-t aking behavior. This means that 87.5% of price was not explained by group differences and that 91.9% of risk-taking beha vior was not explained by group differences as well. In addition, they have the highest F -ratios among the other variables, which are statistically significant (p-value <= .05). Th erefore, price and risk-taking behaviors were analyzed in the canonical discri minant functions. Based on the interaction of the variables price and risk-taking beha vior, price is the most importan t variable in the first function (standardized coefficient = .92) in discriminating the levels of the dependent variable in terms of the standardized coefficients. The second most im portant variable with in teraction for the first function is risk-taking behavior (standardized coefficient = -.73 ). In the second function, risktaking behavior is the most important function (standardized coefficient = .74). The second most important variable for this second function is pr ice (standardized coefficient = .49). Based on the results from this stepwise analysis, the first function indicates that college students that were

PAGE 72

72 positively affected by pricing had a negative impact on risk-taking behavior. Therefore, as pricing of a beer brand goes down, risk-taking behavior goes up. Predictive results of this analysis based on th e two predictor variables, illustrated in Table 4-8, indicated that 54.60% of original grouped cases were correctly classifi ed. According to these results, 78.57% of Bud Light 37.04% of Blue Moon, and 35.71% of Corona were correctly classified. The largest prediction errors occu rred among Blue Moon (10 at 37.04%) and Corona (13 at 46.43%) groups. There is some ambiguity in determining whet her Blue Moon drinker should be predicted as a member of that group or of the Corona group. The overall hit ratio was 4.17. This was found to be statistically significant ( = .05).

PAGE 73

73Table 4-3 Multiple Discriminant Analysis: Beer Brand Choice Function 1 Function 2 Unstandardized Discriminant Coefficients Standardized Coefficients Unstandardized Discriminant Coefficients Standardized Coefficients Wilks Lambda F-ratio p-value* (Constant) -5.59 2.92 Quality .17 .18 .05 .05 .98 .97 .38 Price -.92 -.89 -.45 -.43 .88 6.64 .00 Social -.52 -.69 -.10 -.13 .94 2.92 .06 Emotion .03 .03 .48 .56 .99 .27 .76 Environment .50 .42 -.45 -.38 .93 3.63 .03 Brand Benefits Health .09 .10 -.04 -.04 .98 1.11 .33 Risking Taking .39 .49 -.20 -.27 .92 4.08 .02 Curiosity Seeking .45 .45 -.07 -.09 .95 2.68 .07 Exploratory Shopping Behaviors Variety Seeking -.18 -.25 -.20 -.20 .99 .25 .78 Interpersonal Influence .33 .32 .38 .37 .99 .30 .74 Product Category Involvement .11 .18 -.29 -.48 .97 1.50 .23 Gender .13 .05 1.67 .71 .94 2.88 .06 Employment .19 .10 .05 .02 .98 .82 .45 Greek Membership .48 .22 .95 .44 .98 1.19 .31 Consumer Factors Age .17 .37 -.06 -.14 .94 2.79 .07

PAGE 74

74Table 4-3 Continued Function Eigenvalue % of Variance Cumulative % Canonical Correlation 1 .69 73.8 73.8 .64 2 .24 26.2 100.0 .44 After Function Wilks Lambda Chi-square df Significance* 1 through 2 .48 63.90 30 .00 2 .80 18.80 14 .17 Beer Brand Choice Function 1 Function 2 Bud Light -.89 -.16 Blue Moon 1.02 -.52 Corona .38 .72 *Significance .05 F-ratio is bolded for significance (p-value<=.05)

PAGE 75

75 Table 4-4 Structure Matrix for Beer Brand Choice Function 1 Function 2 Risk Taking .33* -.23 Social -.29* -.12 Age .28* -.14 Curiosity Seeking .27* -.18 Employment -.16* .04 Variety Seeking -.09* .04 Environment .11 -.54* Gender -.06 .49* Price -.36 -.48* Health .00 -.31* Quality .07 -.27* Product Category Involvement .15 -.26* Greek Membership .16 .18* Emotion .04 -.14* Susceptibility to Interpersonal Influence -.08 .09* *Largest absolute correlati on between the variable and any discriminant function. Table 4-5 Group Classification fo r Beer Brand Choice: Multip le Discriminant Analysis Predicted Group Membership Brand Choice Bud Light Blue Moon Corona Number of Cases Bud Light Blue Moon Corona 35 (83.33%) 4 (15.38%) 9 (32.14%) 2 (4.76%) 15 (57.69%) 5 (17.86%) 5 (11.90%) 7 (26.92%) 14 (50.00%) 42 26 28 *Percentage of group cases classified: 66.70%

PAGE 76

76 Table 4-6 Mean and Standard Deviation based on Brand Choice Bud Light Blue Moon Corona Mean Standard Deviation Mean Standard Deviation Mean Standard Deviation Quality Index* 5.42 .80 5.66 1.08 5.26 1.38 Price Index* 5.23 .92 4.74 .82 4.38 1.13 Social Index* 3.42 1.32 2.74 1.44 2.79 1.23 Emotion Index* 4.47 1.18 4.62 1.11 4.39 1.15 Environment Index* 3.94 .95 4.27 .61 3.65 .84 Brand Benefits Health Index* 2.37 1.07 2.50 1.26 2.07 .96 Risking Taking Index* 3.70 1.26 4.58 1.32 3.96 1.13 Curiosity Seeking Index 4.62 .96 5.19 1.04 4.80 1.00 Exploratory Shopping Behaviors Variety Seeking Index* 4.52 1.48 4.28 1.15 4.42 1.45 Interpersonal Influence Index* 3.56 1.06 3.38 .88 3.54 .93 Product Involvement Index* 2.90 1.58 3.54 1.86 2.86 1.53 Gender .74 .45 .62 .50 .89 .31 Employment .62 .49 .46 .51 .54 .51 Greek Membership .62 .49 .73 .45 .79 .42 Consumer Factors Age 20.14 1.75 21.46 2.89 20.68 2.21 *Measured on a 7-poi nt Likert scale Variables were coded as bivariates (0 and 1)

PAGE 77

77Table 4-7 Stepwise Discriminant Analysis: Beer Brand Choice Function 1 Function 2 Unstandardized Discriminant Coefficients Standardized Coefficients Unstandardized Discriminant Coefficients Standardized Coefficients Wilks Lambda F-ratio p-value* (Constant) -2.26 -4.88 Quality .98 .97 .38 Price .95 .92 .51 .49 .88 6.64 .00 Social .94 2.92 .06 Emotion .99 .27 .76 Environment .93 3.63 .03 Health .98 1.11 .33 Risking Taking -.59 -.73 .60 .74 .92 4.08 .02 Curiosity Seeking .95 2.68 .07 Variety Seeking .99 .25 .78 Interpersonal Influence .99 .30 .74 Product Category Involvement .97 1.50 .23 Gender .94 2.88 .06 Employment .98 .82 .45 Greek Membership .98 1.19 .31 Age .94 2.79 .07 Function Eigenvalue % of Varian ce Cumulative Canonical Correlation 1 .24 84.3% 84.3% .44 2 .04 15.7% 100.0% .21

PAGE 78

78Table 4-7 Continued After Function Wilks Lambda Chi-squared D.F. Significance* 1 through 2 .77 23.76 4 .00 2 .96 4.01 1 .05 Function 1 Function 2 Bud Light .54 .01 Blue Moon -.44 .28 Corona -.41 -.27 F-ratio is bolded for significance (p-value <=.05); Test of Equality of Group Means: df for all variable s are (2, 93), N = 96 These variables were not used in analysis fo r unstandardized and standardized coefficients. *Significance .05

PAGE 79

79 Table 4-8 Group Classification: Beer Brand Choice Stepwise Discriminant Analysis Predicted Group Membership Brand Choice Bud Light Blue Moon Corona Number of Cases Bud Light Blue Moon Corona 33 (78.57%) 9 (33.33%) 12 (42.86%) 4 (9.52%) 10 (37.04%) 6 (21.43%) 5 (11.90%) 8 (29.63%) 10 (35.71%) 42 27 28 *Percentage of group cases classified: 54.6% Variables Affecting Beer Consumption A m ultiple discriminant analysis was used in determining linear relationship of the independent variables on the dependent variable beer consumption behavior. The independent variables were all contin uous variables, and the dependent vari able was categorical (light beer drinkers, moderate beer drinkers, and heavy beer drinkers). Table 49 illustrates the results of the multiple discriminant analysis containing unstandardized and standardized coefficients, Wilks Lambda, F-ratio, eigenvalues, and p-value, for each consumption behavior, centroids for the function, and the Wilks Lambda and Chi-square for the canonical discriminant function. There were two functions produced from the mu ltiple discriminant analysis. In function one, the Wilks lambda was .60 and the canonical correlation was .58. Therefore, there is a moderate to strong correlation be tween the discriminant function and the independent variables. This function had 82.3% of variance explained by this function. The first function was found to be statistically signifi cant (p-value = .04, x = 44.47, df = 30) The second function had a Wilks lambda of .90 and canonical corre lation of .31. This function was not found to be statistically significant (p-value = .83). The structure matrix for this analysis is li sted in Table 4-10. This table illustrates the largest absolute correlation between the independ ent variables and the discriminant functions. For the first function, the top th ree variables co rrelated with beer consumption behavior were

PAGE 80

80 emotion, product category involvement, and qua lity. For the second function, the top three variables were risk-takin g behaviors, variety-seeking behavior s, and curiosity-seeking behaviors. The discriminant scores for the first function (centroids) with respec t to beer consumption levels were the following: light beer drinkers = -.89, moderate beer drinkers = .26 and heavy beer drinkers = .80. This means that the function doe s a good job of discrimina ting between all three different levels of consumption behaviors. The most discriminating power among the first function was between light beer drinkers and h eavy beer drinkers. Mode rate and heavy beer drinkers were more closely rela ted than light beer drinkers. The mean and standard deviations for indepe ndent and dependent variables are located on Table 4-12. Independent variables th at had statistical significance with Wilks Lambda were the following: quality (F-ratio = 4.20, p-value = .02), emotion (F-ratio = 11.80, p-value = .00), health (F-ratio = 3.04, p-value = .05), and product catego ry involvement (F-rati o = 9.15, p-value = .00). The mean scores for quality according to consumption behaviors were th e following: light beer drinkers = 5.06 (standard deviat ion = 1.27), moderate beer drinke rs = 5.52 (standard deviation = .94), and heavy beer drinkers = 5.81 (standard deviation = .77). Mean scores for emotion according to consumption behavior were the fo llowing: light beer drinkers = 3.82 (standard deviation = 1.24), moderate beer drinkers = 4.75 (standard deviation = .96), and heavy beer drinkers = 5.00 (standard deviat ion = .80). Mean scores for health with consumption behavior were light beer drinkers = 1.97 (standard de viation = 1.03), moderate beer drinkers 2.44 (standard deviation = 1.18), a nd heavy beer drinkers = 2.61 (standard deviation = 1.00). The mean scores for product category involvement fo r consumption behaviors were the following: light beer drinkers = 2.26 (sta ndard deviation = 1.35), moderate beer drinkers = 3.21 (standard

PAGE 81

81 deviation = 1.69), and heavy beer drinkers = 3.89 (standard deviation = 1.54). The most significant and highest variable a ffecting light, moderate, and hea vy beer drinkers was quality. The four most important variables in terms of individual variables for Wilks Lambda were emotion, product category involvement, quali ty, and health. These four variables had the smallest Wilks Lambda values of .80 for em otion, .84 for product category involvement, .92 for quality, and .94 for health. In a ddition, these four vari ables have the high est F-ratios among all other variables and are statis tically significant (p-value <= .05). Therefore, emotion (standardized coefficient = .56), product category involvement (standardized coefficient = .49), quality (standardized coefficient = -.06), and he alth (standardized coefficient = .16) were the most important factors in discriminating cons umption behavior among all other independent variables. Based on these results, quality was an important factor in the consumption of beer all on levels of consumption. This variable was nega tively related to beer consumption behavior. Therefore, as quality decreases, then beer cons umption increases and vice versa. Emotion was positively related with beer consumption behavior s. Therefore, the more passionate a respondent was about their favorite brand of beer, the more beer they were likely to consume more. Product category involvement also had a positive rela tionship meaning that as product category involvement increases, then consumption beha vior increases. Finally, health had a low, positively related impact on consumption behaviors. On Table 4-11, the classification matrix for beer consumption behavi or and demonstrates that there were 77.14% correctly classified cas es for light beer drin kers. There were 51.52% correctly classified cases for moderate beer drinkers and 57.1 4% correctly classified for heavy beer drinkers. The largest predic tion errors occurred in the mode rate beer drinker group. There

PAGE 82

82 were 16 cases incorrectly classified (48.48% e rrors). Therefore, 62.50% of the classification matrix was correctly classified. The value of the hit rate was 5.29 ( = .05) and was statistically significant. Further analysis was used in determining wh ich independent variable s have the greatest impact on the dependent variable, beer consump tion behavior. This analyze also followed up in order to clear up problems invol ving independence and noncollinear predictor variables used from a multiple discriminant analysis. Table 4-13 illustrates the results of the stepwise discriminant analysis containi ng unstandardized and st andardized coefficients, Wilks Lambda, F-ratio, eigenvalues, and p-value, for each consumption behavior, centroids for the function, and the Wilks Lambda and Chi-square for the canoni cal discriminant function. One variable out of the fifteen variables was selected when the Wilk s lambda (.80) was the lowest and the canonical correlation was .45. There was one function discriminated between all three beer drinking levels, light, moderate, and heavy. The function was found to be statistical ly significant (p .05). The canonical correlation (R) for the function is .45. This means that between the independent variables included in the analysis, there is a mode rate correlation between the discriminant function and the independent vari ables. The Wilks Lambda for the canonical discriminant function is .80. This means that 80% of the discriminant scores for variance are not explained by group differences. The Wilks Lambda (.80) and Chi-Square test (21.04) were found to be statistically significan t. The Chi-Square test has a pvalue of .00 with 2 degrees of freedom. This means that there can be inferences made from the data on to the general population based on these results. The average discriminant sc ores for the function (centroids) with respect to consumption levels were the following: light beer drinkers = -.64, moderate beer drinkers =

PAGE 83

83 .26 and heavy beer drinkers = .50. This means that the function does a good job of discriminating between all three different levels of consumption behaviors. The two most important variables in terms of individual variables for Wilks Lambda were emotion and product category involvement. These two variables had the smallest Wilks Lambda values of .80 for emotion and .84 for product category involvement. This means that 79.8% of emotion was not explained by group differences and that 83.6% of product category involvement was not explained by group differences In addition, these two variables have the highest F-ratios among all other variables and are statistically signifi cant (p-value <= .05). However, based on the stepwise discriminant analysis, all independent va riables except emotion were removed for analysis. In the discriminant function, emotion is the most important factor (standardized coefficient = 1.00) and was the only independent variab le analyzed. This relationship was positive as in the multiple discriminant analysis. Therefore, the more a college student feels emotionally toward a brand, then th e more likely they to exhibit increased beer consumption behavior. On Table 4-14, the classification matrix dem onstrates that there were 57.14% correctly classified cases for light drinkers. There were 52.94% correctly classified cases for moderate drinkers and 24.14% correctly classified for h eavy drinkers. The largest prediction errors occurred in the heavy drinker group. There were 53 cases incorrectly clas sified (54.1% errors). Therefore, 45.9% of the classifi cation matrix was correct ly classified. The value of the hit rate was 2.71 ( = .05). This hit rate was statistically significant.

PAGE 84

84Table 4-9 Multiple Discriminant Analysis: Beer Consumption Behavior Function 1 Function 2 Unstandardized Discriminant Coefficients Standardized Coefficients Unstandardized Discriminant Coefficients Standardized Coefficients Wilks Lambda F-ratio p-value* (Constant) -4.81 -5.85 Quality -.06 -.06 -.21 -.22 .92 4.20 .02 Price .10 .10 -.27 -.27 .96 2.15 .12 Social -.02 -.03 .07 .09 .98 1.12 .33 Emotion .55 .56 .18 .18 .80 11.80 .00 Environment -.36 -.31 -.01 -.01 .97 1.23 .30 Brand Benefits Health .15 .16 .04 .04 .94 3.04 .05 Risking Taking -.13 -.16 .35 .48 .97 1.57 .21 Curiosity Seeking .30 .31 .37 .47 .97 1.50 .23 Exploratory Shopping Behaviors Variety Seeking -.10 -.14 .33 .33 .97 1.64 .20 Interpersonal Influence .15 .15 .25 .25 .99 .56 .57 Product Category Involvement .32 .49 -.10 -.15 .84 9.15 .00 Gender -.90 -.39 -.88 -.38 .98 .99 .38 Employment .18 .09 .86 .43 .98 .91 .41 Greek Membership -.98 -.46 .18 .08 .99 .71 .49 Consumer Factors Age .12 .28 .10 .23 .97 1.68 .19 Function Eigenvalue % of Variance Cumulative % Canonical Correlation 1 .51 82.3 82.3 .58 2 .11 17.7 100.0 .31

PAGE 85

85Table 4-9 Continued After Function Wilks Lambda Chi-square df Significance* 1 through 2 .60 44.47 30 .04 2 .90 8.96 14 .83 Beer Consumption Behavior Function 1 Function 2 Light Beer Drinkers -.89 -.13 Moderate Beer Drinkers .26 .43 Heavy Beer Drinkers .80 -.35 F-ratio is bolded for significance (p-value<=.05) *Significance .05

PAGE 86

86 Table 4-10 Structure Matrix for Beer Consumption Behavior Function 1 2 Emotion .70* .18 Product Category Involvement .62* -.15 Quality .42* -.08 Health .36* .05 Price .30* -.08 Age .26* -.06 Gender -.20* .00 Greek -.17* -.05 Risk-Taking -.01 .56* Variety-Seeking -.12 .50* Curiosity-Seeking .14 .45* Employment -.04 .41* Social .14 .36* Environment .19 .28 Susceptibility to Interpersonal Influence .08 .28 Table 4-11 Group Classification: Beer Consumption Behavior Mult iple Discriminant Analysis Predicted Group Membership Consumption Behavior Light Beer Drinker Moderate Beer Drinker Heavy Beer Drinker Number of Cases Light Beer Drinker Moderate Beer Drinker Heavy Beer Drinker 27 (77.14%) 12 (36.36%) 3 (10.71%) 7 (20.00%) 17 (51.52%) 9 (32.14%) 1 (2.86%) 4 (12.12%) 16 (57.14%) 35 33 28 *Percentage of group cases classified: 62.5%

PAGE 87

87 Table 4-12 Mean and Standard Devi ation: Beer Consumption Behavior Light Beer Drinker Moderate Beer Drinker Heavy Beer Drinker Mean Standard Deviation Mean Standard Deviation Mean Standard Deviation Quality Index* 5.06 1.27 5.52 .94 5.81 .77 Price Index* 4.59 1.23 4.89 .80 5.12 .91 Social Index* 2.82 1.24 3.31 1.49 3.04 1.31 Emotion Index* 3.82 1.24 4.75 .96 5.00 .80 Environment Index* 3.77 .88 4.09 .80 3.99 .90 Brand Benefits Health Index* 1.97 1.03 2.44 1.18 2.61 1.00 Risking Taking Index* 3.93 1.32 4.32 1.32 3.76 1.16 Curiosity Seeking Index 4.64 1.07 5.06 .98 4.79 .95 Exploratory Shopping Behaviors Variety Seeking Index* 4.49 1.42 4.68 1.40 4.05 1.28 Interpersonal Influence Index* 3.40 .90 3.65 .96 3.48 1.08 Product Involvement Index* 2.26 1.35 3.21 1.69 3.89 1.54 Gender .83 .38 .73 .45 .68 .48 Employment .54 .51 .64 .49 .46 .51 Greek Membership .77 .43 .67 .48 .64 .49 Consumer Factors Age 20.14 1.42 20.76 2.45 21.18 2.83 *Measured on a 7-poi nt Likert scale Variables were coded as bivariates (0 and 1).

PAGE 88

88 Table 4-13 Stepwise Disc riminant Analysis: Beer Consumption Behavior Function 1 Unstandardized Discriminant Coefficients Standardized Coefficients Wilks Lambda F-ratio p-value* (Constant) -4.35 Quality .92 4.20 .02 Price .96 2.15 .12 Social .98 1.12 .33 Emotion .97 1.00 .80 11.80 .00 Environment .97 1.23 .30 Health .94 3.04 .05 Risking Taking .97 1.57 .21 Curiosity Seeking .97 1.50 .23 Variety Seeking .97 1.64 .20 Susceptibility to Interpersonal Influence .99 .56 .57 Product Category Involvement .84 9.15 .00 Gender .98 .99 .38 Employment .98 .91 .41 Greek Membership .99 .71 .49 Age .97 1.68 .19 Function Eigenvalue % of Variance Cumulative Canonical Correlation 1 ..25 100.0% 100.0% .45 After Function Wilks Lambda Chi-squared D.F. Significance 1 .80 21.04 2 .00 Beer Consumption Behaviors Function 1 Light Beer Drinkers -.64 Moderate Beer Drinkers .26 Heavy Beer Drinkers .50 F-ratio is bolded for significance (p-value<=.05); Test of Equality of Group Means: df for all variables are (2, 93), N = 96 These variables were not used in analysis fo r unstandardized and standardized coefficients. *Significance .05

PAGE 89

89 Table 4-14 Group Classification: Beer Consumption Behavior Step wise Discriminant Analysis Predicted Group Membership Beer Brand Choice Light Drinker Moderate Drinker Heavy Drinker Number of Cases Light Beer Drinker Moderate Beer Drinker Heavy Beer Drinker 20 (57.14%) 9 (26.47%) 6 (20.69%) 15 (42.86%) 18 (52.94%) 16 (55.17%) 0 (0%) 7 (20.59%) 7 (24.14%) 35 34 29 *Percentage of group cases classified: 45.9%

PAGE 90

90 CHATPER 5 DISCUSSION The three brands used in this study were excellent brands to analyze based on the differences in their positioning and marketing strategies. Bud Light is positioned as a convenient, light beer that focuses on fun and great taste. Bud Light is a brand extension from the popular brand, Budweiser. Bud Light ut ilizes massive marketing effort s (television, radio, internet, outdoor, sponsorships, print, etc.) by focusing humor and popular culture references used by their consumers. This brand has the most mark et share among the three brands, and based on pretest and main study results, Bud Light was the most popular brand among all other brands investigated. The positioning for Blue Moon is a premium, craft beer. The marketing for Blue Moon focuses on word of mouth and point of sale. Most consumers associate the brand with the slice of orange that is traditionally placed on the side of the glass when served. This is a key differentiation of the brand among other beers. C onsumers know that a pint of beer with an orange slide garnished on the side of the gla ss is Blue Moon. The brand is not highly marketed and does not have a presence in most marketing me dia. This strategy appeals to their consumers that love premium, craft beers. The positioning for Corona is escapism. Corona has traditionally been known as an import brand from Mexico, and the brand has moved into the idea of escaping to an exotic destination. The brand tries to completely differentiate itself from all other beers. The brand competes with premium beers based on its import status in the U.S. It has something similar with Blue Moon based on its garnish, a sliced lime pushed through its clear bottle. The marketing efforts focus on this idea of escaping to a relaxing place, whether it is a beach or just calm ocean waters. Corona

PAGE 91

91 creates a mood with their branding. They focus heavily on television and print advertising for their marketing efforts. All three brands are completely different in their marketing and positioning strategies. The prices of a six pack of twelve ounce bottles for all three brands from a local retailer were the following: Bud Light = $4.54, Blue Moon = $6.97, and Corona = $6.97. Therefore, based on using these three brands, there we re different segments of cons umers within the sample with different marketing strategies for each. Therefore, these three brands prov ided variance in their positioning and marketing strategy among the college market. Results from this study demonstrated variab les of significance for both beer brand choice and beer consumption behavior. These variable s were quality, price, emotion, environment, health, risk-taking behavior, a nd product category involvement. However, other variables were not found to be significant. Subjective norm variable s such as social and interpersonal influence may not have been deemed significant because this truly may not be a variable that influenced this market. It could also be po ssible that this market has difficu lty answering questions in this area. College students did not seem to be aide d in a beer consumption or beer brand choice decision based on social/interperson al factors. Previous research studies have found that there were social and interpersonal factors infl uence drinking alcohol (Wall et al. 1998, July; Trafimow, 1996; OCallaghan et al. 1997, September). However, this study concluded that they were not found to be influential in beer cons umption behavior and brand choice. In addition, retail marketing could potentially have very lit tle effect on this market based on these results. The exploratory shopping behavi ors of curiosity-seeking and variety-seeking did not have significant results as well. Variet y-seeking consumers looked for a lternatives that the consumer was familiar with. College students appeared to either be loyal to a brand or be a risk-taker. They

PAGE 92

92 did not look for alternatives to their favorit e brands. However, they may be looking for something new and different to become their ne w favorite. Curiosity-motivated behavior meant that the consumer sought out information thr ough shopping and interperso nal means. The fact that this is not found to be significant reitera tes the subjective norm re sults found from other variables. Study findings provided evidence that there was not a significant association between brand choice of beer among co llege students and situational variation. Although, some of the results had consistency with brand positioning (i.e. respondents drank Corona by the pool), there was not significant data found situation and bran d choice. College students appeared to either remain brand loyal or strayed to other brands based on other factors. Significant factors influencing beer brand choice were price, risk-taking behavior, and environment. The main factor that influenced brand choice was price. This supports previous research in terms of college st udents purchasing beer based on pr icing for that particular product category. As price goes up, then the brand choice for the brand declines. Inversely, as the price decreases, then brand choice for the brand incr eases. Risk-taking behaviors were found to be prevalent among college students with brand choi ce. These behaviors co incided with pricing because college students were open to choice w ith the product category of beer. Therefore, college students were willing to take risks when choosing a brand of beer, and pricing could be a major factor influencing this risk-taking beha vior. For example, a st udent could potentially retract from buying their favorite beer and partake in risk-taking behavior in order to save some money by purchasing a cheaper bra nd of beer that is not their favorite. Price and risk-taking behavior had an inverse relations hip among the two most important variables. Therefore, as the price of a beer brand decreases, then the likeli hood of a college student to exhibit risk-taking

PAGE 93

93 behavior increases. College student are very pri ce sensitive. Some loyalties and favorites exist. However, most college students cannot turn down a bargain. Therefore, in terms of impacting brand switching behaviors, the most influential f actor is pricing strate gies and promotions. It was unclear if marketing in a particular situation would be eff ective because there was not a relationship discovered that influences a beer brand choice with situation. However, the situation does influence the amount of beer consumed. However, marketers must be very cautious if they market to these heavy to modera te drinking behaviors and maintain a responsible outlook when attempting to reach these high beer consumption consumers. Beer consumption behaviors varied based on the situations that college students were involved. College students consumed lighter amounts of beer in situations such as special events, beach/pool areas, at a restaurant or relaxing at home, or in any situation that is not a party, bar, or club setting. Heavy and moderate beer consumption occurred mostly in bars clubs, or parties. College students tend to exhibit consumption beha vior based on the situation that they are in. Variables that were significant with beer cons umption behavior were quality, emotion, health, and product category involvement. The most in fluential factor was emotion. Based on beer consumption behaviors, a light beer drinker is more likely to not have a strong emotional attachment to brands. However, a heavy to mode rate drinker has more of a strong emotion for brands of beer. In addition, the product category involvement and quality also seems to coincide with these behaviors. If college students were highly involved with beer, then they consume more and had a greater interest in the overall quality of the beer. College students had mild concerns regarding health, and th is could deal with light beer consumption. The health concerns associated with drinking beer could potentially be curving the behavior with the consumption.

PAGE 94

94 For example, if a college student is concerned with gaining weight due to carbohydrates, then the student would be more likely to consume less beer. College students are a vast market that c ontains different segments involving several factors that influence their brand choice. It was apparent that there are divisions among their beer consumption behaviors. Light beer drinkers app eared to be willing to try different brands and base purchasing on pricing. Heavy to moderate b eer consumption behaviors seemed to be more passionate emotionally about brands and were more involved with the product category of beer. This group could potentially exhibi t brand loyalties. This ideal is similar in marketing to the 80/20 rule, where 20% of the market for a brand re presents 80% of the sales for that particular brand. Among group differences in analysis in beer consumption behaviors, groups contrast because moderate to heavy beer drinkers tend to be less likely to take a risk in their shopping behaviors. The light beer drinkers were vast in numbers, but they are less likely to carry loyalties to brands of beer. Marketers should recognize these factors in the college student population. A very specific, targeted message involving emoti on for a beer brand could affect buying behavior for this heavy to moderate beer consumer. Thes e college students want to be recognized on a personal, emotional level and have a strong passion for beer. In ma rketing to the vast segment of college students influenced by pricing and risk -taking behaviors, the message should involve competitive pricing strategies and emphasize diffe rentiation. Brand extensions and new beer products would be of interest for this group. With the Theory of Reasoned Action (TRA) mode l in mind, it could be potentially feasible to present information from this study in order to seek certain behavior s from the college student market. Advertising to this market provides th e opportunity to influence brand attitudes,

PAGE 95

95 intention, and ultimately attempt to result in a purchasing behavior for a particular brand of beer. In addition, advertising may have some indir ect impact on subjective norms by reaching opinion leaders or individuals among a social group. The model also create s areas of investigation where further research could be utilized in order to determine influentia l factors that were not explored which have an impact beer cons umption and beer brand choice. This research study further validates the no tion of using brand be nefits instead of measuring product attributes for brand choice res earch. In addition, insights from the college market were discovered from this research study. This study has expa nded upon previous brand choice studies. In addition, this study adds to th e research studies involved with the topic of college students and consumption behavior. Ther e has been evidence provided that influential factors do influence beer brand choice and beer consumption behavior among college students. However, this study was only exploratory in natu re. This research creat es a starting point for further areas of investigation i nvolving situation variation, the college student marketing, beer brand choice, and beer consumption behavior.

PAGE 96

96 CHAPTER 6 LIMITATIONS The influential factors on beer brand choi ce with brand benefits among other variables have not been explored previously. More res earch is needed in order to have a better understanding of the college market. In this st udy, none of the demogra phic factors were found to be significant. This could have been attributed to the sample. It is evident that there were vast differences in the college market when it involv es making a beer brand choice. In addition, the population used in this study was not a true repr esentation of the college market. Sampling was limited to courses in the College of Journalism and Communications at the University of Florida. Therefore, there were differences among gender, Greek membership, employment, and age could have been affected by this. In addition, two measures used in the study scor ed low reliability scores and were reduced to two items. Health (reduced fr om four items to two) and curi osity-motivated behavior (reduced from three items to two) scored reliability scor es of .74 and .53 respectivel y. This reliability may have hindered the results for these variables in volved in this study. Health was found to be statistically significant for beer consumption behaviors and curios ity-motivated behavior was not found to be significant for neither brand choice or consumption behavior. These variables may need further investigation and research. However, these measures were found to have an acceptable reliability in prev ious studies (Orth, 2005). Based on the broken down sample of 222 to 98, the sample could not be further narrowed in order to achieve more reliable results based on multiple and stepwise discriminant analyses. This study does not use previously recognized al cohol consumption levels used in previous studies. Beer brand consumption be havior levels used in this st udy differed in contrast to other studies involving both the genera l population and other studies using college students.

PAGE 97

97 Situational variation and beer brand choice was not found to be significant. This could have resulted from many possible problems. The sample size could have been too small in order to create differentiated results, the constructi on of measuring variables may not have been the best way to indicate situ ational variation, or there could have been a problem with recalling the exact situation in which the respondent was last drinking their favorite beer. The fact that the beer brand choice situation question asked coll ege students where they last consumed there favorite beer also creates some li mitations as well. It is possible that a college student may drink their favorite beer more routinel y in other locations other than simply the last location they consumed it. Therefore, in future research st udies, the measure of this variable should be carefully analyzed in order to ga in significant and valid findings. Multiple discriminant analyses were used in the main analysis for the study. However, a stepwise discriminant analysis was used in orde r to limit multicollinearity. Stepwise discriminant analysis has been criticized in ot her research studies. The issues that have been associated with this analysis are incorrect degrees of freedom based on computer packages to analyze data, sampling error capitalization, and failure to select the best subset of variables in a given size. Problems with this research coul d be addressed and improved with more research on the topic, recalculations, improved sampling, further analyses on data, and better reliability for construct scales would aid in improving the current research (Thompson, 1995). The Theory of Reasoned Action (TRA) model wa s used as a theoretic al foundation for this research study. However, the model does involve some limitations. One limitation of the TRA model is based on ignoring internal or exte rnal conditions influencing or hindering the individuals behavior. Extrane ous variables such as alcohol tolerance, opportunity, and money could influence behavior (Netemeyer et al. 1991) Also, the model does not recognize more than

PAGE 98

98 one behavior at a time. It is pos sible for an individual to have si multaneous intentions leading to multiple behaviors. The model uses attitudes and subjective norms to predict behavior. However, based on this model, there is limited amount of learning that can be obtained (Ajzen and Fishbein, 1980; Schlegel et al. 1977; Hays, 1985).

PAGE 99

99 APPENDIX A PRETEST

PAGE 100

100Hello, My name is Dave Ritter, and I am a graduate student in advertis ing at the University of Florida. This is a research study ques tionnaire for my thesis project. My topic involves the study of attitude s, perceptions, and decisions involving brand choice for the pro duct category of beer. Please take a few moment s to read over the questions carefully. There are no right or wrong answers, so ple ase answers the questions honestly. Your results will remain confidential, and you will be anonymous upon the completion of this s urvey. I would just like to thank you very much for your participation in this study. Returning the survey indicates your cons ent for use of the answers you supply. If you have any questions, you may contact Dave Ritter at 309-531-3509. You may also reach me at dritter@ufl.edu Please fill o ut the follow information in order to receive credit for participating in this st udy. This section will be separa ted from the survey upon completion to protect your anonymity. Name: ________________________________ UF E-Mail: ________________________________ UF ID: ________________________________

PAGE 101

101Construct: Brand Choice 1. Think about the brands of beer you have consumed in the past month What brands were they? Plea se check as many or as few as apply. _____Amber Bock _____Amstel Light _____Becks _____Blue Moon _____Bud Extra _____Bud Light _____Bud Select _____Budweiser _____Busch _____Coors Light _____Corona _____Dos Equis _____Fosters _____Guinness _____Heineken _____Keystone Light _____Killians _____Michelob _____Michelob Ultra _____Miller High Life _____Miller Lite _____Natural Light _____Old Milwaukees Best _____Pabst Blue Ribbon _____Red Stripe _____Rolling Rock _____Sam Adams _____Schlitz _____Steel Reserve _____Stella _____Yuengling _____Other, Please Specify ________ _____None of the Above Construct: Consumption Behavior 2. On average how many days a week do you drink b eer (place a circle around th e appropriate number)? 0 1 2 3 4 5 6 7

PAGE 102

102Construct: Situational Factors 3. Please indicate by circling how likely you are to drink beer in the following situations (place a circle around a number for each situation). Not Very Likely Very Likely In a bar or club with a date 1 2 3 4 5 6 7 In a bar or club with friends 1 2 3 4 5 6 7 At a sporting event 1 2 3 4 5 6 7 Family events 1 2 3 4 5 6 7 At a restaurant with a date 1 2 3 4 5 6 7 At a restaurant with friends 1 2 3 4 5 6 7 Party with friends at home 1 2 3 4 5 6 7 Alone at home 1 2 3 4 5 6 7 Construct: Situational Factors 4. In addition to these situations, what are other occasions or situations that you drink beer? Please indicate your response in the blanks provided below. ____________________________________________________________________________________________________________ ____________________________________________________________________________________________________________ ____________________________________________________________________________________________________________

PAGE 103

103Construct: Consumption Behavior 5. How much beer on average do you drink (one drink = 12 oz bottle and/or can) in one occasion? (Please do not include mixed drinks or other drinks that are not c onsidered beer for your answer) _____________________________________________________________________________ The following questions ask about your perceptions and attitudes regarding Budweiser Please answer them to the best of our knowledge. Construct: Quality/Performance Benefits 6. Please indicate the extend of your agreement with the following statements about the quality of Budweiser beer (place only one circle for each statement). Strongly Disagree Strongly Agree Budweiser has consistent quality. 1 2 3 4 5 6 7 Budweiser is well made. 1 2 3 4 5 6 7 Budweiser has an acceptable standard of quality. 1 2 3 4 5 6 7 Budweiser has poor craftsmanship. 1 2 3 4 5 6 7 Budweiser is a brand that woul d last a long time among other br ands of beer. 1 2 3 4 5 6 7 Budweiser would perform consistently. 1 2 3 4 5 6 7 Construct: Health Benefits 7. Please indicate in the bla nks below all the health benefits from Budweiser that you can think of. ____________________________________________________________________________________________________________ ____________________________________________________________________________________________________________

PAGE 104

104Construct: Price/Value for Money Benefits 8. Please indicate the extend of your agreement with the following statements about the pricing of Budweiser beer (place only one circle for each statement). Strongly Disagree Strongly Agree Budweiser is reasonably priced. 1 2 3 4 5 6 7 Budweiser offers value for money. 1 2 3 4 5 6 7 Budweiser is a good product for the price. 1 2 3 4 5 6 7 Budweiser is economical. 1 2 3 4 5 6 7 Construct: Normative/Personal (Social) Benefits 9. Please indicate the extend of your agreement with the following statements about the social impact of Budweiser beer (place only one circle for each statement). Strongly Disagree Strongly Agree Budweiser helps me feel acceptable. 1 2 3 4 5 6 7 Budweiser improves the way I am pe rceived by others. 1 2 3 4 5 6 7 Budweiser makes a good impression on other people. 1 2 3 4 5 6 7 Budweiser gives its owner social approval. 1 2 3 4 5 6 7 Construct: Emotion Benefits 10. Please indicate the exte nd of your agreement with the following statements about your attitudes and enjoyment of Budweiser beer (place only one circle for each statement). Strongly Disagree Strongly Agree Budweiser is a product that I would enjoy. 1 2 3 4 5 6 7 Things about Budweiser make me want to use it. 1 2 3 4 5 6 7 Budweiser makes me feel relaxed. 1 2 3 4 5 6 7

PAGE 105

105Construct: Emotion Benefits Budweiser makes me feel good. 1 2 3 4 5 6 7 Budweiser would give me pleasure. 1 2 3 4 5 6 7 Budweiser evokes thoughts of happiness. 1 2 3 4 5 6 7 Budweiser soothes me. 1 2 3 4 5 6 7 Budweiser eliminates all fear. 1 2 3 4 5 6 7 Budweiser eliminates all anger. 1 2 3 4 5 6 7 Budweiser makes me anxious. 1 2 3 4 5 6 7 Budweiser makes me want to use it. 1 2 3 4 5 6 7 Construct: Environment (Stewardship) Benefits 11. Please indicate the ex tend of your agreement with the following statements about the environmental impact of Budweiser beer (place only one circle for each statement). Strongly Disagree Strongly Agree Budweiser is produced in an environmentally friendly way. 1 2 3 4 5 6 7 Budweiser is made without polluting the environment. 1 2 3 4 5 6 7 Budweiser is made by people who care for the environment. 1 2 3 4 5 6 7 Construct: Health Benefits 12. Please indicate the ex tend of your agreement with the following statements about the health implications of Budweiser beer (place only one circle for each statement). Strongly Disagree Strongly Agree Budweiser comes with lots of health benefits. 1 2 3 4 5 6 7 Budweiser promotes ones health when enjoyed in moderation. 1 2 3 4 5 6 7

PAGE 106

106Construct: Brand Choice 13. Think about the brands of beer that you plan to consume in the next month Please rank order the following brands of beer by placing a next to the brand of beer that you will most likely consume in the next month, a next to the brand you feel t he next likely to be consumed, through 32 (including Other) which would represent the brand that you are least likely to consume in t he next month. If you do not plan to consume any of the following brands, then please chec k None of the Above. _____Amber Bock _____Amstel Light _____Becks _____Blue Moon _____Bud Extra _____Bud Light _____Bud Select _____Budweiser _____Busch _____Coors Light _____Corona _____Dos Equis _____Fosters _____Guinness _____Heineken _____Keystone Light _____Killians _____Michelob _____Michelob Ultra _____Miller High Life _____Miller Lite _____Natural Light _____Old Milwaukees Best _____Pabst Blue Ribbon _____Red Stripe _____Rolling Rock _____Sam Adams _____Schlitz _____Steel Reserve _____Stella _____Yuengling _____ Other Specify _____________ _____ None of the Above

PAGE 107

107Please take a few moments to fill to me a little about yourself by circling the appropriate responses. Please keep in mind tha t your information will remain complete ly confidential and anonymous. 14. What is your gender? 1) Male 2) Female 15. What is your current age (write in number in years)? _________ 16. What is your current year as a student? 1) Freshman 2) Sophomore 3) Junior 4) Senior 5) Masters Graduate Student 6) PHD Student 7) Other Please Specify _________________________ 17. What is your curren t employment status? 1) Full Time 2) Part Time 3) Unemployed

PAGE 108

10818. What is your current major? 1) Public Relations 2) Advertising 3) Telecommunications 4) Journalism 5) Other Please Specify _________________________ 19. Please circle your curr ent living situation? 1) Live with Parents 2) Live alone (Apartment or House) on campus 3) Live alone (Apartment or House) off campus 4) Live with roommate(s) (Apa rtment or House) on campus 5) Live with roommate(s) (Apart ment or House) off campus 6) Fraternity or Sorority House 7) Live in a Dorm Alone 8) Live in a Dorm with roommate(s) 9) Other Please Specify _________________________ 20. Are you a member of a frat ernity or sorority? 1) Yes 2) No

PAGE 109

109 APPENDIX B MAIN STUDY

PAGE 110

110Hello, My name is Dave Ritter, and I am a graduate student in advertis ing at the University of Florida. This is a research study ques tionnaire for my thesis project. My topic involves the study of attitude s, perceptions, and decisions involving brand choice for the pro duct category of beer. Please take a few moment s to read over the questions carefully. There are no right or wrong answers, so ple ase answers the questions honestly. Your results will remain confidential, and you will be anonymous upon the completion of this s urvey. I would just like to thank you very much for your participation in this study. Returning the survey indicates your cons ent for use of the answers you supply. If you have any questions, you may contact Dave Ritter at 309-531-3509. You may also reach me at dritter@ufl.edu Please fill o ut the follow information in order to receive credit for participating in this st udy. This section will be separa ted from the survey upon completion to protect your anonymity. Name: ________________________________ E-Mail: ________________________________ UF ID: ________________________________

PAGE 111

111Please take a moment to think about the last time you consumed your favorite beer Construct: Brand Choice 1. What brand of beer is it (Please write only one beer brand in the blank provided below)? _____________________________________________________ Construct: Situational Factors 2. For the brand you mentioned in question 1 which of the following situations did you last consume this beer (Please circle only one choice for your answer)? 1) In a bar or club with friends 2) In a bar or club with a date 3) In a bar or club alone 4) In a restaurant with friends 5) In a restaurant with a date 6) In a restaurant alone 7) At home with friends (not a party) 8) At home with a date (not a party) 9) At home alone (not a party) 10) Sporting event/Concert with friends 11) Sporting event/Concert with date 12) Sporting event/Concert alone 13) Beach/Pool with friends (not a party) 14) Beach/Pool with date (not a party) 15) Beach/Pool alone (not a party) 16) Party (not at home) with friends 17) Party (not at home) with date 18) Party (not at home) alone 19) Party at your home 20) Other; Please Specify _________________________

PAGE 112

1123. For the brand you mentioned in question 1 when did you most recently consume this beer (Please circle only one choice for your answer)? 1) A week ago (within the last 7 days) 4) More than one month but less than 3 months (30-90 days) 2) More than one week but less than two weeks (8-14 days) 5) More than 3 months (90+ days) 3) More than two weeks but less than a month (15-29 days) 6) Does not apply 4. For the brand mentioned in question 1 if you purchased this beer, then what was the cost per beer (Please estimate cost for one beer, where one beer = 12 ounces)? If you did not purchase, then please indicate by writing N/A in the blank provided. ______________________________________________________________ Construct: Quality/Performance Benefits 5. Please indicate the extent of your agreem ent with the following statements about your favorite beer ( as mentioned in question one ). Please place only one circle for each statement. Strongly Disagree Strongly Agree It has consistent quality. 1 2 3 4 5 6 7 It is well made. 1 2 3 4 5 6 7 It has an acceptable standard of quality. 1 2 3 4 5 6 7 It has poor craftsmanship. 1 2 3 4 5 6 7 It would last a long time among othe r brands of beer. 1 2 3 4 5 6 7 It would perform consistently. 1 2 3 4 5 6 7

PAGE 113

113Construct: Price/Value for Money Benefits 6. Please indicate the extent of your agreem ent with the following statements about your favorite beer (as mentioned in question one). Please place only one circle for each statement. Strongly Disagree Strongly Agree It is reasonably priced. 1 2 3 4 5 6 7 It offers value for money. 1 2 3 4 5 6 7 It is a good product for the price. 1 2 3 4 5 6 7 It is economical. 1 2 3 4 5 6 7 Construct: Normative/Personal (Social) Benefits 7. Please indicate the extent of your agreem ent with the following statements about your favorite beer (as mentioned in question one). Please place only one circle for each statement. Strongly Disagree Strongly Agree It helps you feel acceptable. 1 2 3 4 5 6 7 It improves the way you are perceived by others. 1 2 3 4 5 6 7 It makes a good impression on other people. 1 2 3 4 5 6 7 It gives you social approval. 1 2 3 4 5 6 7

PAGE 114

114Construct: Emotion Benefits 8. Please indicate the extent of your agreem ent with the following statements about your favorite beer (as mentioned in question one). Please place only one circle for each statement. Strongly Disagree Strongly Agree It is a product that you enjoy. 1 2 3 4 5 6 7 Things about it make you want to use it. 1 2 3 4 5 6 7 It makes you feel relaxed. 1 2 3 4 5 6 7 It makes you feel good. 1 2 3 4 5 6 7 It gives you pleasure. 1 2 3 4 5 6 7 It evokes thoughts of happiness. 1 2 3 4 5 6 7 It soothes you. 1 2 3 4 5 6 7 It eliminates all fear. 1 2 3 4 5 6 7 It eliminates all anger. 1 2 3 4 5 6 7 It makes you anxious. 1 2 3 4 5 6 7 It makes you want to use it. 1 2 3 4 5 6 7 Construct: Environment (Stewardship) Benefits 9. Please indicate the extent of your agreem ent with the following statements about your favorite beer (as mentioned in question one). Please place only one circle for each statement. Strongly Disagree Strongly Agree It is produced in an environmentally friendly way. 1 2 3 4 5 6 7 It is made without polluting the environment. 1 2 3 4 5 6 7 It is made by people who care for the environment. 1 2 3 4 5 6 7

PAGE 115

115Construct: Health Benefits 10. Please indicate the extent of your agreement with the following statements about your favorite beer (as mentioned in question one). Please place only one circle for each statement. Strongly Disagree Strongly Agree It comes with lots of health benefits. 1 2 3 4 5 6 7 It promotes ones health when enjoyed in moderation. 1 2 3 4 5 6 7 It is a good way to relieve your stress. 1 2 3 4 5 6 7 It glorifies unhealthy drinking behaviors. 1 2 3 4 5 6 7 Construct: Exploratory Shopping Behaviors 11. Please indicate the extent of your agreem ent with the following statements (pl ace only one circle for each statement). Strongly Disagree Strongly Agree I have little interest in fads and fashions. 1 2 3 4 5 6 7 I like to shop around and look at displays. 1 2 3 4 5 6 7 I shop around a lot for my clothes just to find out more about the latest styles. 1 2 3 4 5 6 7 I hate window shopping. 1 2 3 4 5 6 7 12. Please indicate the extent of your agreem ent with the following statements (pl ace only one circle for each statement). When I eat out, I like to try the most unusual items the rest aurant serves, even if I am not sure I would like them. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree When I go to a restaurant, I feel it is sa fer to order dishes I am familiar with. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree

PAGE 116

116 I am very cautious in tryi ng new/different products. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree I would rather wait for others to try a new store or restaurant than try it myself. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree I never buy something I dont know about at the risk of making a mistake. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree 13. Please indicate the extent of your agreem ent with the following statements (pl ace only one circle for each statement). A new store or restaurant is not somethi ng I would be eager to find out about. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree Investigating new brands of gr ocery and other similar products is generally a waste of time. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree When I hear about a new store or restaurant, I take advant age of the first opportunity to find out more about it. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree Construct: Susceptibility to Interpersonal Influence 14. Please indicate the extent of your agreem ent with the following statements (pl ace only one circle for each statement). I often consult other people to help choose the be st alternative available from a product class. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree If I want to be like someone, I often tr y to buy the same brands that they buy. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree It is important that others lik e the products and brands I buy.

PAGE 117

117 Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree To make sure I buy the right product or brand, I often observe what others are buying and using. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree I rarely purchase the latest fashion styles until I am sure my friends approve of them. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree I often identify with other pe ople by purchasing the same products and brands they purchase. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree If I have little experience with a produc t, I often ask my friends about the product. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree When buying products, I generally purchase those brands that I think others will approve of. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree I like to know what brands and prod ucts make good impressions on others. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree I frequently gather info rmation from friends or family about a product before I buy. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree If other people can see me using a product, I often purchase the brand they expect me to buy. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree I achieve a sense of belonging by purchasing the same products and brands that others purchase. Strongly Disagree 1 2 3 4 5 6 7 Strongly Agree

PAGE 118

118Construct: Consumption Behavior 15. On average how many days a week do you drink beer (place a ci rcle around the appr opriate number)? 0 1 2 3 4 5 6 7 16. How much beer on average do you drink (one drink = 12 oz. bottle and/or can) in a one occasion ? (Please do not include mixed drinks or other drinks that are not c onsidered beer for your answer) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17+ Construct: Product Category Involvement 17. Please indicate the extent of your agreement with the following statements (place only one circle for each statement). Strongly Disagree Strongly Agree Generally, I am someone who finds it important what beer he or she buys. 1 2 3 4 5 6 7 Generally, I am someone who is interested in th e kind of beer he or she buys. 1 2 3 4 5 6 7 Generally, I am someone whom it means a lot wh at beer he or she buys. 1 2 3 4 5 6 7 Please take a few moments to tell us about yourself by circling the appr opriate responses. Please keep in mind that your information will remain completely c onfidential and anonymous. 18. What is your gender? 1) Male 2) Female 19. What is your current age (write in number in years)? _________

PAGE 119

11920. What is your current year as a student? 1) Freshman 2) Sophomore 3) Junior 4) Senior 5) Masters Graduate Student 6) PHD Student 7) Other Please Specify _________________________________________ 21. What is your curren t employment status? 1) Full Time 2) Part Time 3) Unemployed 22. What is your current major? 1) Public Relations 2) Advertising 3) Telecommunications 4) Journalism 5) Other Please Specify _________________________ 23. Please circle your curr ent living situation? 1) Live with Parents 2) Live alone (Apt/Home) on campus 3) Live alone (Apt/Home) off campus 4) Live with roommate(s) (Apt/Home) on campus 5) Live with roommate(s) (Apt/Home) off campus 6) Fraternity or Sorority House 7) Live in a Dorm Alone 8) Live in a Dorm with roommate(s) 9) Other Please Specify ___________________________ 24. Are you a member of a frat ernity or sorority? 1) Yes 2) No Thank you for your participation in this study and remem ber if you do drink, then please drink responsibly!!!

PAGE 120

120 REFERENCES Aaker, D. A., Batra, R. and Myers, J. G. (1992), Advertising Management 4th ed, Prentice Hall, Englewood Cliffs, NJ. Ajzen, I. and Fishbein, M. (1980), Understanding Attitudes and Predicting Social Behavior Englewood Cliffs, NJ: Prentice-Hall, Inc. Alcohol makers tread tricky path in marketing to college students. (2005) USA Today Money, pp. 01b. Alvarez, B. A. and Casielles, R. V. (2005), Consum er evaluations of sales promotion: the effect on brand choice, European Journal of Marketing Vol. 39, No. pp. 54-70. Andorfer, J. B. (2005), Got Beer? A-B plots health pitch, Advertising Age Vol. 76, Issue 42. Auger, P., Burke, P., Devinney, T. M., and Louvier e, J. J. (2003), What will consumers pay for social product features?, Journal of Business Ethics Vol. 42, No. 3, pp. 281-304. Babor, T. F., and Mendelson, J. H., Greenberg, I., a nd Kuehnle, J. (1978), Experimental analysis of the happy hour: effects of purch ase price on alcohol consumption, Psychopharmocology, Vol. 58, pp. 35-41. Bagozzi, R. P. (1986), Principles of Marketing Management Science Research Associates, Chicago, IL. Bagozzi, R. P., Baumgartner, H., and Yi, Y. (1992), State versus action orientation and the theory of reasoned action: application to coupon usage, Journal of Consumer Research Vol. 18, March, pp. 505-518. Ballantyne, R., Warren, A., and Nobbs, K. (2006), The evolution of brand choice, Brand Management Vol. 13, No. 4/5, pp. 339-352. Bandura, A. (1977). Social Learning Theory New York: Prentice Hall. Basler, R. P. (1953), Collected works of Abraham Lincoln, Vol. 1, New Brunswick, New Jersey: Rutgers University Press. Batra, R., Homer, M.P., and Kahle, L. R. (2001) Values, susceptibility to normative influence, and attribute importance Weight s: A Nomological Analysis, Journal of Consumer Psychology, Vol. 11, No. 2, pp. 115-128. Baumgartner, B. (2003), Measuring changes in brand choice behavior, Schmalenbach Business Review Vol. 55, pp. 242-256.

PAGE 121

121 Bearden, W. O. and Etzel, M. J. (1982), Reference group influence on product and brand purchase decisions, Journal of Consumer Research Vol. 9, No. 2, pp. 183-194. Bearden, W., Netemeyer, R. G., and Tell, J. E. (1989), Measurement of consumer susceptibility to interpersonal influence, Journal of Consumer Research Vol. 15, No. 4, pp. 473-481. Beer has an image crisis; wi ne and spirits gain. (2005) USA Today 1AP. Beers Bad Rap for Carbs Unjus tified. (2006), USA Today Magazine Vol. 134, Issue 2731. Belch, G. E. and Belch, M. A. (1995), Introduction to Advertising and Pr omotion: An Integrated Marketing Communications Perspective 3rd ed., Boston, MA: Irvin. Belk, R. (1974), An exploratory assessment of situational effects in buyer behavior, Journal of Marketing Research Vol. 11, pp. 156-163. Bentz, Y. and Merunka, D. (2000), Neural networ ks and the multinomial logit for brand choice modelling: a hybrid approach, Journal of Forecasting Vol. 19, pp. 177-200. Berden, T. P. J., Brombacher, A. C., and Sander, P. C. (2000), The building bricks of product quality: An overview of some basic concepts and principles, International Journal of Production Economics Vol. 67, pp. 3-15. Bern, C., Contias, M., Elorz, M., and Mgi ca, J. M. (2004), The use of retail store database for brand choice analysis, Int. Rev. of Retail, Distribution and Consumer Research Vol. 14, No. 1, pp. 19-29. Boulard, G. (2005), Drinking too much, too: trying to find an answer to the persistent habit or binge drinking among young peopl e vexes the nations policymakers. State Legislatures pp. 12-15. Bowlby, J. (1979), The Making and Breaking of Affectional Brands London: Tavistock. Bozinoff, L. and Victor, R. J. (1984), Recall and recognition memo ry for product attributes and benefits, Advances in Consumer Research Vol. 11, No. 1, pp. 348-352. Braglia, M. and Petroni, A. ( 2000), Stakeholders influence and internal championing of product stewardship in the Italia n food packaging industry, Journal of Industrial Ecology Vol. 4, No. 1, pp. 75-92. Bruen, J. (2002), Product liability: The role of the product steward, Risk Management Vol. 49, No. 2, pp. 34-35. Calantone, R. and Knight, G. (2000), The critical role of product quality in the international performance of industrial firms, Industrial Marketing Management Vol. 29, pp. 493-506.

PAGE 122

122 Charlton, P. and Ehrenberg, A. S. C. (1973) McConnells experimental brand choice data, Journal of Marketing Research Vol. X, pp. 302-307. Chib, S., Seetharaman, P. B., and Strijnev, A. (2004), Model of brand choice with a no purchase option calibrated to scanner-panel data, Journal of Marketing Research Vol. XLI, pp. 184-196. Choi, C. S. and Coughlan, A. T. (2006), Private label positioning: Quality versus feature differentiation from the national brand, Journal of Retailing Vol. 82, No, 2, pp. 79-93. Cioletti, J. (2006), Spread the brews, Beverage World Vol. 125, No. 1. Christie, J., Fisher, D., Kozup, J. C., Smith, S., Burton, S., and Creyer, E. H. (2001), The effects of bar-sponsored alcohol beverage promo tions across binge and nonbinge drinkers, Journal of Public Policy and Marketing Vol. 20, No. 2, pp. 240-253. Coate, D. and Grossman, M. (1988), Effects of alcoholic beverage prices and legal drinking ages on youth alcohol use, Journal of Law and Econometric Evidence pp. 145171. Collins, R. L., Schell, T., Ellickson, P. L., and McCaffrey, D. (2003), Predictors of beer advertising awareness among eighth graders, Addiction, Vol. 98, pp. 1297-1306. Coxe, D. (2004), Sex is out, carbs are in, Macleans Vol. 117, No. 5. Day, J. C. and Jamieson A. (2005) School enrollment: 2000, U.S. Census Bureau. Washington, DC: U.S. Government Printing Offi ce. Retrieved April 12, 2006 from http://www.census.gov/prod/2003pubs/c2kbr-26.pdf Dom estic ad spending by category (2006), Advertising Age, from TNS Media Intelligence. Retrieved September 27, 2006, from www.adage.com Eagly, A. H. and Shelly, C. (1993), The Psychology of Attitudes Orlando, FL: Harcourt Brace Jovanovich, Inc. Ennew, C. T., Reed, G. V., and Binks, M. R. (1993), Im portance-performance analysis and the measurement of service quality, European Journal of Marketing Vol. 27, No. 2, pp. 5970. Erdem, T. and Swait, J. (2004), Brand cr edibility, brand consideration, and choice, Journal of Consumer Research Vol. 31, pp. 191-198. Faden, Vivian B. (2006), Trends in initiation of alcohol use in the United States 1975 to 2003, Alcoholism: Clinical and Experimental Research Vol. 30, No. 6, pp. 1011-1022, June, from NHSDA 1998.

PAGE 123

123 Fall 2005 Product Beverages. (2006), MRI+ Mediamark Internet Reporter database 2006, accessed April 12, 2006 Farquar, P. H. (1989), Managing brand equity, Marketing Research Vol. 1, pp. 2433. Feltham, T. S. (1998), Leaving home: Br and purchase influences on young adults, Journal of Consumer Marketing, Vol. 15, No. 4, pp. 372-385. Fornell, C., Johnson, M. D., Anderson, E. W., Cha, J. and Bryant, B. E. (1996), The American customer satisfaction in dex: Nature, purpose, and findings, Journal of Marketing Vol. 60, pp. 7-18. Fry, J. N. (1971), Personality vari ables and cigarette brand choice, Journal of Marketing Research Vol. III, pp. 298-304. Green, C. A., Perrin, N. A., and Polen, M. R. ( 2004), Gender differences in the relationships between multiple measures of alcohol consumption and physical and mental health, Alcohol Clin Exp Res Vol. 28, pp. 754-764. Grossman, M., Coate, D., and Arluck, G. M. (1987) Price sensitivity of al coholic beverages in the United States, Control Issues in Alcohol Abuse Prevention: Strategies for States and Communities Harold H. Holder ed., JAI Press, Inc., Greenwich, Conn., pp. 169-198. Gruenewald, P., Madden, P., and Janes, K. (1992), Alcohol availability and formal power and resources of state alcohol beverage control agencies, Alcohol Clin Exp Res Vol. 16, pp. 591-597. Gruenewald, P., Miller, A., and Treno, A. (1993) Alcohol availability and the ecology of drinking behavior, Alcohol Health Res World Vol. 17, pp. 39-45. Haley, R. I. (1968), Benefit segmentation: A decision-oriented research tool, Journal of Marketing Vol. 32, pp. 30-35. Hartford, T. C., Weschsler, H., and Rohman, M. (1983), The structural context of college drinking, Journal of Studies on Alcohol Vol. 44, pp. 722-732. Havan, C. and Shaver, P. R. (1994), Attachment as an organizational framework for research on close relationships, Psychological Inquiry Vol. 5, pp. 1-22. Hays, R. (1985), An integrated value-expectan cy theory of alcohol and other drug use, British Journal of Addiction Vol. 80, pp. 379-384. Helliker, K. and Ellison, S. (2005), Anheuser wants world to know beer is healthy, Wall Street Journal (Eastern edition) New York, N.Y., pp. B1.

PAGE 124

124 Hoyer, W. D. and MacInnis, D. J. (2004), Consumer Behavior 3rd Edition, Boston: Houghton Mifflin Company. Hickle, G. T. and Stitzhal, D. (2003), Apportioning the responsibilities fo r product stewardship: A case for a new federal role, Environmental Quality Management Vol. 12, No. 3, pp.116. Hruschka, H. (2002), Market share analysis using semi-parametric attraction models, European Journal of Operational Research Vol. 138, pp. 112-225. Snider, M. (2006), IPods knock over beer mugs, USA Today ; Section: Life, pp. 09d. taken from Spring 2006 Lifestyle an d Media Study, Student Monitor. Is the glass half-empty or half-full? (2004), Tufts University Health and Nutrition Letter, Vol. 22, No. 9. Jessor, R. (1981), The perceived treatment environm ent in psychological theo ry and Research. In D. Magnusson (ed.), Toward a Psychology of Situations: An Interactionist Perspective Hillsdale, NJ: Lawrence Erlbaum Associates. Johnston, L. D., OMalley, P.M., Bachman, J. G., and Schulenberg, J. E. (2004), Monitoring the Future National Survey Results on Drug Use 1975-2003. Vol. II : College Students and Adults Ages 19-45 Bethesda, MD: National Institute on Drug Abuse, NIH Pubilication No. 04-5508. Jones-Webb, R., Toomey, T., Short, B., Murra y, D., Wagenaar, A., and Wolfson, M. (1997), Relationship among alcohol availability, drinking location, alcohol consumption, and drinking problem in adolescents, Subst Use Misuse Vol. 32, pp. 219-228. Kandel, D. B. (1980), Drug and drinking behavior am ong youth, In A. Inkeles, N.J. Smelser and R. Turner (eds.), Annual Review of Sociology Vol. 6, pp. 235-285, Palo Alto, CA: Annual Reviews, Inc. Keller, K. L. (1993), Conceptualizing, measuring, and managing customer-based brand equity, Journal of Marketing Vol. 57, No. 1, pp. 1-22. Klasky, A. L. (2006), Drink to your health?, Scientific American Special Edition Vol. 16, No. 4. Knibbe, R. A., Oostveen, T., and Van de Goor, I. (1991), Young peoples alcohol consumption in public places: Reasoned behavior or related to the situation? British Journal of Addiction Vol. 86, pp. 1425-1433. Kondo, K. (2004), Beer and health: Preventative eff ects of beer components on lifestyle related diseases, BioFactors Vol. 22, pp. 303-310.

PAGE 125

125 Kuo, M., Wechsler, H., Greenberg, P., and Lee, H. (2003), The marketing of alcohol to college students: The role of low pr ices and special promotions, American Journal of Preventive Medicine, Vol. 25, No. 3, pp. 204-211. Lai, A. (1991), Consumption situation and produc t knowledge in the adoption of a new product, European Journal of Marketing Vol. 25, No. 10, pp. 55-67. Lambert, Z. V. (1972), Price and choice behavior, Journal of Marketing Research Vol. IX, pp. 35-40. Lancaster, K. J. (1971), Variety, Equity and Efficiency New York: Columbia Press. Laurent, G. and Kapferer, J. (1985), M easuring consumer involvement profiles, Journal of Marketing Research Vol. 22, No. 1, pp. 41-53. Lger, J. and Scholz, D. (2004) The fickle beer consumer, Marketing Magazine, Vol. 109, No. 17. Levy, S. and Sheflin, N. (1983), New evidence on controlling alcohol use through price, Journal of Studies on Alcohol, Vol. 44, pp. 929-937. Levy, S. and Sheflin, N. (1985), The demand for alc oholic beverages: An aggregate time series analysis, Journal of Public Policy and Marketing Vol. 4, No. 1, pp. 47-54. Loftus, P. (2005), Study brews on health benefits of hops, Wall Street Journal (Eastern edition) pp. B4C. Mayhew, G. E. and Winer, R. S. (1992), An empirical analysis of internal and external reference prices using scanner data, Journal of Consumer Research Vol. 19, June, pp. 62-70. McGuire, W. J. (1968), Personality and susceptibility to social influence. In E. F. Borgetta and W. W. Lambert (eds), Handbook of Personality Theory and Research Chicago: Rand McNally, pp. 1130-1187. Miller, K. E. and Ginter, J. L. (1979), An investigation of situational variation in brand choice behavior and attitude, Journal of Marketing Research Vol. XVI, pp. 111-123. Netemeyer, R. G., Burton, S., and Johnston, M. (1991), Volitional and goal-directed behaviors: A confirmato ry analysis approach, Social Psychology Quarterly Vol. 54, No. 2, pp. 87-100. OCallaghan, F. V., Chant, D. C., Callan, V. J., and Baglioni, A. (1997), Models of alcohol use by young adults: An examination of various attitude -behavior theories, Journal of Studies on Alcohol Vol. 58, No. 5, pp. 502-507.

PAGE 126

126 Orth, U. R. (2005), Consumer personality and othe r factors in situational brand choice variation, Brand Management Vol. 13, No. 2, pp. 115-133. Orth, U. R., McDaniel, M. R., Shellhammer, T., and Lopetcharat, K. (2004), Promoting brand benefits: The role of consumer psychographics and lifestyle, Journal of Consumer Marketing Vol. 21, No. 4, pp. 31-47. Orth, U. R., McGarry Wolf, M., and Dodds, T. (2005), Dimensions of wine region equity and their impact on consumers preferences, Journal of Product and Brand Management Vol. 14, No. 4, pp. 477-489. sterberg, E. (1995), Do alcohol prices affect consumption and related problems? In: H. D. Holder and G. Edwards (eds.), Alcohol and Public Policy: Evidence and Issues New York: Oxford University Press, pp. 145-163. Papatla, P. and Krishnamurthi, L. (1996) Measuring the dynamic effects of promotions on brand choice, Journal of Marketing Research Vol. XXXIII, pp. 20-35. Park, C. W., Jaworski, B. J., and MacInnis, D. J. (1986), Strategic brand concept-image management, Journal of Marketing Vol. 50, pp. 621-635. Park, C. W. and Lessig, V. P. (1977), Students and housewives: Differences in susceptibility to reference group influence, Journal of Consumer Research Vol. 4, No. 2, pp. 102-110. Park, C. W. and Mittal, B. (1985) A theory of involvement in consumer behavior: Problems and issues, Research in Consumer Behavior Vol. 1, pp. 201-232. Puth, G., Mostert, P., and Ewing, M. (1999), Cons umer perceptions of mentioned product and brand attributes in magazine advertising, Journal of Product and Brand Management Vol. 8, No. 1, pp. 38-49. Quester, P. G. and Smart, J. (1998), The infl uence of consumption si tuation and product involvement over consumers use of product attributes, Journal of Consumer Marketing, Vol. 15, No. 3, pp. 220-238. Raju, P. S. (1980), Optimum stimul ation level: Its relationship to personality, demographics, and exploratory behavior, Journal of Consumer Research Vol. 7, No. 3, pp. 272-282. Ram, S. and Jung, H. (1989), The link between involvement, use innovativeness and product usage, Advances in Consumer Research Vol. 16, pp. 160-166. Redman, S., Sanson-Fisher, R. W., Wilkinson, C. Fahey, P., and Gibberd, R. W. (1987), Agreement between two measures of alcohol consumption, Journal of Studies on Alcohol Vol. 48, pp. 104-108.

PAGE 127

127 Reeves, C. A. and Bednar, D. A. (1994), Defi ning quality: Alternatives and implications, Academy of Management Review Vol. 18, No. 3, pp. 419-445. Rhodes, P. (2005), 3 cheers for beer, Natural Health Vol. 35, Vol. 9. Richins, M. L. and Bloch, P. H. (1986), After the new wears off: The temporal context of product involvement, Journal of Consumer Research Vol. 13, No. 2, pp. 280-285. Romaniuk, J. (2003), Brand attributes distribution outlets in the mind, Journal of Marketing Communications, Vol. 9, pp. 73-92. Romaniuk, J. and Sharp, B. (2003) Measuring brand perceptions: Testing quantity and quality, Journal of Targeting, Measurem ent and Analysis for Marketing Vol. 11, No. 3, pp. 218229. Schlegel, R. P., Crawford, C. A., and Sanburn, W. D. (1977), Correspondence and mediational properties of the Fishbein model: An a pplication to adolescent alcohol use, Journal of Experimental Social Psychology Vol. 13, pp. 421-430. Schwartz, R. H., Farrow, J. A., Banks, B., and Gi esel, A. E. (1998), Use of false ID cards and other deceptive methods to purchase al coholic beverages during high school, Journal of Addictive Diseases Vol. 17, No. 3, pp. 25-33. Simonson, I., Carmon, Z., and OCurr y, S. (1994), Experimental evid ence on the negative effect of product features and sales promotions on brand choice, Marketing Science, Vol. 13, No. 1, Winter, pp. 23-40. Singh, V. P., Hansen, K. T., and Sachin, G. (2005), Modeling prefer ences for common attributes in multicategory brand choice, Journal of Marketing Research Vol. XLII, pp. 195-209. Smith, W. (1956), Product differentiation and mark et segmentation as alternative marketing strategies, Journal of Marketing Vol. 21, pp. 3-8. Stafford, J. E. and Cocanougher, B. A. (1977), Reference group theory. Selected Aspects of Consumer Behavior, Washington, DC: Superintendent of Documents, U.S. Government Printing Office, pp. 361-380. Sweeney, J. C. and Soutar, G. N. (2001), Consum er perceived value: The development of a multiple item scale. Journal of retailing Vol. 77, No. 2, pp. 203-220. Thompson, B. (1995), Stepwise regression and stepwi se discriminant analysis need not apply here: A guidelines editioral, Educational and Psychological Measurement Vol. 55, No. 4, pp. 525-535.

PAGE 128

128 Thomson, M., MacInnis, D. J., and Park, C. W. (2005), The ties that bind: Measuring the strength of consumers emotional attachments to brands, Journal of Consumer Psychology, Vol. 15, No. 1, pp. 77-91. Trafimow, D. (1996), The importance of attitudes in the prediction of college students intentions to drink, Journal of Applied Social Psychology Vol. 26, No. 24, pp. 2167-2188. Trafimow, D. and Fishbein, M. (2001), The modera ting effect of behavior type on the subjective norm-behavior relationship, The Journal of Social Psychology Vol. 134, No. 6, pp. 755763. Uncles, M. D. and Ehrenberg, A. S. ( 1990), Brand choice among older consumers, Journal of Advertising Research August/September, Vol. 30, No. 4, pp. 19-22. U.S. market share leaders. (2007), Advertising Age from TNS Media Intelligence. Retrieved August 11, 2007, from www.adage.com Van Trijp, H. M. C. (1994), Product-related determ in ants of variety-seeking behavior for foods. Appetite, Vol. 22, pp. 1-10. Van Trijp, H. C. M., Hoyer, W. D., and Inman, J. (1996), Why switch? Product category level explanations for true va riety-seeking behavior, Journal of Marketing Research Vol. 33, No. 3, pp. 281-292. Vazquez, R., Belen del Rio, R., and Iglesias, V. (2002), Consumer-based brand equity: Development and validation of a measurement instrument, Journal of Marketing Management Vol. 18, No. pp. 27-48. Viral cant cure what ails beast. (2006), Advertising Age, Vol. 77, No. 42. Wagner, U. and Taudes, A. (1986), A multivariate polya model of brand choice and purchase incidence, Marketing Science, Vol. 5, No. 3, pp. 219-244. Wahlers, R. G., Dunn, M. G., and Etzel, M. J. (1986), The congruence of alternative OSL measures with consumer expl oratory behavior tendencies, Advances in Consumer Research Vol. 13, No. 1, pp. 398-402. Walker, R. (2004), The idea of a beer with a h ealth benefit is debatable, but thats clearly whats resonating, New York Times Magazine Vol. 153, No. 52746, pp. 18. Wall, A., Hinson, R. E., and McKee, S. A. ( 1998), Alcohol outcome expectancies, attitudes toward drinking and the theory of planned behavior, Journal of Studies on Alcohol Vol. 59, No. 4, pp. 409-419. Walsh, G. and Mitchell, V. (2005), Demographi c characteristics to consumers who find it difficult to decide, Marketing Intelligence and Planning Vol. 23, No. 2/3, pp. 281-295.

PAGE 129

129 White, T. and Pomponi, R. (2003), Gain a competitive edge by preventing recalls, Quality Progress Vol. 36, No. 8, pp. 41-49. Wilkie, W. L. (1986), Consumer Behavior, New York: John Wiley and Sons. Wilks, J. and Callan, V. J. (1990), A diary appr oach to analyzing young adults drinking events and motivations, Australian Drug and Alcohol Review Vol. 5, pp. 3-7. Witt, R. E. (1969), Informal social group influence on consumer brand choice, Journal of Marketing Research Vol. VI, pp. 473-476. Wood, L. M. (2004), Dimensions of brand purchas ing behavior: Consumers in the 18-24 age group, Journal of Consumer Behavior Vol. 4, No. 1, pp. 9-24. Woods, P. (2005), Good news about brews, Health Vol. 19, No. 8. Woodside, A. G. and Fleck Jr., R. A. (1979), Th e case approach to understanding brand choice, Journal of Advertising Research, Vol. 19, No. 2, pp. 23-30. Xiaohua, L. and Germain, R. (2003), Product quality orientation and its performance implications in Chinese state-owned enterprises, Journal of International Marketing Vol. 11, No. 2. Yang, S., Allenby, M. G., and Fennell, G (2002), M odeling variation in brand preference: The roles of objective environmen t and motivating conditions, Marketing Science, Vol. 21, No. 1, pp. 14-21. Zaichkowsky, J. L. (1985), Familiarity: Product use, involvement or expertise? In E. C. Hirschman and M. B. Holbrook (eds), Advances in Consumer Research Vol. 12, Provo, UT: Association for Consumer Research.

PAGE 130

130 BIOGRAPHICAL SKETCH Dave Ritter ( david_ritter_99@hotmail.com ) is a graduate student m ajoring in advertising at the College of Journalism a nd Communications at the Universi ty of Florida. His research focuses on brand choice, marketing, and consumer behavior. Dave completed his Associate of Science in business administration from Lakela nd College. In addition, Dave completed his Bachelor of Science in marketing fr om the University of Illinois.