<%BANNER%>

The Effect of Brand-Consumer Congruency on Brand Adoption Behavior in Social Networking Sites

Permanent Link: http://ufdc.ufl.edu/UFE0041833/00001

Material Information

Title: The Effect of Brand-Consumer Congruency on Brand Adoption Behavior in Social Networking Sites
Physical Description: 1 online resource (115 p.)
Language: english
Creator: Riediger, Stefanie
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: adoption, behavior, branding, congruity, facebook, gratifications, motivation, networking, sns
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: THE EFFECT OF BRAND-CONSUMER CONGRUENCY ON BRAND ADOPTION BEHAVIOR IN SOCIAL NETWORKING SITES This study investigates the self congruity theory and social networking site (SNS) usage motivations as contributing factors of branded posting behavior on SNSs through regression analysis. This research revealed that brand attitudes, self brand congruity, and motivational effects are unique per brand, making context and relevancy of brands important factors for marketing campaigns on SNSs. Compatibility was not significantly related to branded SNS posting behavior, but a significant positive relationship occurred between self brand congruity and adoptive posting behavior likelihood. All four motivations tested had significant relationships with adoption behavior likelihood individually, but none of the motivations interacted significantly with self brand congruity in predicting behavior. Thus, to increase SNS word-of-mouth likelihood, individuals must feel as though the typical user of a brand mirrors their own self image. Independent of self brand congruity, SNS usages for entertainment, information, social interaction, and convenience purposes can also indicate increased adoption behavior, uniquely per brand.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Stefanie Riediger.
Thesis: Thesis (M.Adv.)--University of Florida, 2010.
Local: Adviser: Sutherland, John C.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041833:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041833/00001

Material Information

Title: The Effect of Brand-Consumer Congruency on Brand Adoption Behavior in Social Networking Sites
Physical Description: 1 online resource (115 p.)
Language: english
Creator: Riediger, Stefanie
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: adoption, behavior, branding, congruity, facebook, gratifications, motivation, networking, sns
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: THE EFFECT OF BRAND-CONSUMER CONGRUENCY ON BRAND ADOPTION BEHAVIOR IN SOCIAL NETWORKING SITES This study investigates the self congruity theory and social networking site (SNS) usage motivations as contributing factors of branded posting behavior on SNSs through regression analysis. This research revealed that brand attitudes, self brand congruity, and motivational effects are unique per brand, making context and relevancy of brands important factors for marketing campaigns on SNSs. Compatibility was not significantly related to branded SNS posting behavior, but a significant positive relationship occurred between self brand congruity and adoptive posting behavior likelihood. All four motivations tested had significant relationships with adoption behavior likelihood individually, but none of the motivations interacted significantly with self brand congruity in predicting behavior. Thus, to increase SNS word-of-mouth likelihood, individuals must feel as though the typical user of a brand mirrors their own self image. Independent of self brand congruity, SNS usages for entertainment, information, social interaction, and convenience purposes can also indicate increased adoption behavior, uniquely per brand.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Stefanie Riediger.
Thesis: Thesis (M.Adv.)--University of Florida, 2010.
Local: Adviser: Sutherland, John C.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041833:00001


This item has the following downloads:


Full Text





THE EFFECT OF BRAND-CONSUMER CONGRUENCY
ON BRAND ADOPTION BEHAVIOR IN SOCIAL NETWORKING SITES




















By

STEFANIE RIEDIGER


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

2010

































2010 Stefanie Riediger









ACKNOWLEDGMENTS

I would like to express a heartfelt thank you and much gratitude to everyone who

has helped me in achieving my Master of Advertising degree. First and foremost, I am

greatly in debt to the extended invaluable and diligent hands-on assistance of my thesis

chair, Dr. John Sutherland. He agreed to take on advisor duties and pick up where

another advisor left off when it was not required of him. His unyielding patience and

encouragement was monumental. His guidance pushed me to produce quality work and

helped me believe in my abilities to do so. I would also like express my gratitude to Dr.

Jorge Villegas, Dr. Jon Morris and Dr. Robyn Goodman for their time and guidance with

regard to my study. Further thanks go to Jody Hedge for her administrative help.

I would also like to thank my loving parents who have been supportive of me every

step of the way, pushing me to do my best. Their continued love and support helped

remind me of what was important when the going got tough and discouragement set in.

I could always count on them to keep me focused on my goals and what needed to be

done to achieve them. I would also like to extend my thanks to my sister, Allison, my

brother-in-law, Kevin, and many wonderful friends who were sources of encouragement

in this process. They were always there for me.

Lastly, I would like to thank those from my past who helped me realize my

capabilities as a student, a professional, and most importantly as a person. They made

me see that I really can work through anything by taking things one step at a time.









TABLE OF CONTENTS

page

ACKNOW LEDGMENTS ........ ......... ......... .... ............... ............... 3

L IS T O F T A B L E S ........................................................................ ........... ........ ....... 6

LIS T O F F IG U R E S .................................................................. 7

A B S T R A C T ........................................................... .. ....................................... 8

CHAPTER

1 INTRO DUCTIO N .............. ............... .............. .......... .. ........... 9

2 LITERATURE REVIEW ............................. ..... ....................... 13

A Digital Revolution ........................................ 13
Participatory Cultures......................................... ............... 13
Uses and Gratifications.......................................... ............... 18
Online Profiling ................. ........ ....................... 20
The New W ord-of-Mouth......................... .......... ............... 23
Self-Congruity Theory ................................................................ 25
Hypotheses .................... ........ ............... 28

3 M ET H O D ............... ................................................................................... 34

R research D design ............... ......................................................................... 34
Participants ............................................................ 35
Data Collection and Procedure ....................................... ....... .......... .... 36
Measures and Instrument ...................................... ......... 40
Independent Variable ............... ........................... 42
Dependent Variable.............................. ............... 45
Motivation Interaction Variables ........ ........................ ........ ........... 47

4 R ESU LTS ......... ..... ......... ....................................................................... 48

D ata A na lysis ............................................................ .................................. 4 8
S a m p le P ro file ......... .............. ................................................ ............... 4 8
Facebook Tendencies/Activity.................... ............ ......... 49
Brand Attitudes, Congruity, and Compatibility ............ ....... ....... ......... 50
Independent, Dependent, and Interaction Variables ...................................... 51
Dependent Variable: Adoptive Behavior.............................. ............... 52
Hypothesis Testing .................................................................. ........ .................. 53
Interaction Effect of SNS Motivations ................ ........ ..................... 57
R research Q questions ...... .......... ......... .......... ....................... ............ 6 1
L im ita tio ns ......... ...... ........... ............................................ ... 6 6









5 DISCUSSION AND CONCLUSIONS................ ............................. 82

D is c u s s io n .............. ..... ............ ................. ............................................. 8 2
Im p licatio ns ........................... .............. ...... 8 3
Im plications Sum m ary ......... ........... .......................... ..... .......... 97
Future Research ............. ..... ............................. 99

APPENDIX

A IN V ITA T IO N ....................................................... 103

B ONLINE SURVEY QUESTIONNAIRE ................................ ......... ... ........... 104

LIST O F R EFER ENC ES ......... .......................................................... ............... 110

BIOGRAPHICAL SKETCH ............... ..... ........ ................. 115









LIST OF TABLES


Table page

4-1 Sample profile summary statistics ...... .................. ............... 68

4-2 Current Facebook profile behavior (direct expression)...................................... 70

4-3 Current Facebook profile behavior (indirect expression) ................................ 71

4-4 Brand attitudes, brand congruity, and compatibility ....................................... 72

4-5 Summary statistics for independent and motivation variables............................ 73

4-6 Dependent variable component correlations ................................................. 73

4-7 Dependent variable m easures.................................... ........................... ....... 74

4 -8 C o rre la tio n s ............................................................................. 7 5

4-9 Self-brand congruity correlations ............... .................................. ............... 75

4-10 Compatibility correlations ........... .. ................................... 76

4-11 Multiple regression for entertainment motivation variable (H3)....................... 77

4-12 Multiple regression for information motivation variable (H4)............................. 78

4-13 Multiple regression for social interaction motivation variable (RQ1)................... 79

4-14 Multiple regression for convenience motivation variable (RQ2)...................... 80

4-15 Summary of multiple regression significant relationships.............. ............... 81









LIST OF FIGURES

Figure page

2-1 O overview of the study .............................. ........................ .............. 33









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

THE EFFECT OF BRAND-CONSUMER CONGRUENCY
ON BRAND ADOPTION BEHAVIOR IN SOCIAL NETWORKING SITES

By

Stefanie Riediger

August 2010

Chair: John Sutherland
Major: Advertising

The emergence of social networking sites as a medium for communication has

provided a wave of word-of-mouth opportunities for advertisers. This study investigates

the self-congruity theory and social networking site (SNS) usage motivations as

contributing factors of branded posting behavior on SNSs through regression analysis.

This research revealed that brand attitudes, self-brand congruity, and motivational

effects are unique per brand, making context and relevancy of brands important factors

for marketing campaigns on SNSs. Compatibility was not significantly related to

branded SNS posting behavior, but a significant positive relationship occurred between

self-brand congruity and the likelihood of adoptive posting behavior. All four motivations

tested also had significant relationships with adoption behavior likelihood acting

independently, but none of the motivations interacted significantly with self-brand

congruity in predicting behavior. Thus, to increase likelihood of SNS word-of-mouth,

individuals must increasingly feel as though the typical user of a brand mirrors their own

self-image. Independent of self-brand congruity, SNS usages for entertainment,

information, social interaction, and convenience purposes can also indicate increased

adoption behavior, depending on the brand in question.









CHAPTER 1
INTRODUCTION

To further develop the most successful brand communication strategies, it is

helpful to study the connections made by a consumer that can ultimately lead to

branded behavior and purchase decisions. This study is designed to observe social

networking site (SNS) posting behavior and the intricacies of both consumer

perceptions and attitudes toward a brand to explain what factors contribute to the

likelihood of branded adoption behavior on SNS. This objective was achieved by

examining the effects of self-brand congruity and brand-SNS compatibility on branded

posting behavior within an SNS.

Previous studies have examined congruity in terms of a person's actual self-image

in comparison with his/her ideal image (Parker, 2005; Sirgy et al., 1997). Parker (2005)

also applied this comparison to brand personalities. The most relevant portion of this

research, however, involves the integration of these ideas by looking at the alignment

between self-image and brand-user image. Measures of congruity between one's self-

image and a brand-user image are determined as a result of this integration (Parker,

2005; Sirgy et al., 1997). The resulting self-brand congruity influences consumer

behavior as a function of the brand attitude elements of familiarity, attachment, and

loyalty (Parker, 2005). The self-congruity model within the realm of consumer behavior

will act as a pivotal variable, along with the perceived compatibility of a brand with a

particular SNS, as possible explanations for SNS posting behavior.

To facilitate the explanation of why consumers interact with brands within SNSs,

the uses and gratifications theory was adapted to the model. The assumption involved

in this theory is that media users are active users who take actions in search of a goal









(Blumler & Katz, 1974). Furthermore, Ko, Cho, and Roberts (2005) argued that these

people "are aware of their needs, and select the appropriate media to gratify their

needs" (p. 60). The following investigation of the translation of consumer identification

with a brand into adoption behavior of a brand relies heavily on formulated insight based

on this theory's foundational assumptions. The assumption to be explored is that the

higher a person's self-brand congruity (or the more a consumer identifies his/her actual

and ideal self with the typical user of a brand), the more likely that person is to adopt the

brand into his/her SNS posting behavior. In addition, the uses and gratifications theory

is applied with the prediction that a person's motivation for using an SNS could affect

how that person perceives brands within that context and ultimately how that person

posts.

Thus, the final element of this study, which contributes a new perspective to the

research on consumer attitudes and behaviors, is the medium within which consumers

are observed. The Internet has a growing and encouraging capacity for interaction that

has intensified the uses and gratifications theory (Ko, Cho, & Roberts, 2005). Not only

has the Internet been a natural incubator for many types of relationships and interaction,

but more specifically, SNSs have emerged as prominent stimulators of social activity

(Boyd, 2006). Until the development of online communities, connectivity and interaction

was limited to one-way or two-way communication (Shirky, 2009). However, with the

development of SNSs in recent years, this capacity has multiplied exponentially due to

the power of mass speech given to the individual. Webs of one-way and many-way

communication are formed simultaneously, creating an entire new world in which









marketers can immerse themselves and capitalize on an opportunity for premium

word-of-mouth communication value (Shirky, 2009).

One outcome of the social networking age is the practice of online profiling. The

analytical nature of the Internet makes a succinct, yet encompassing first impression a

commanding deliverable. The resulting trend of profiling oneself on Internet

communities continues to grow, as SNSs such as Facebook facilitate a diverse range of

communications (Willett, 2009). As these types of sites become increasingly popular,

new advantages of reaching consumers through this medium present themselves to

marketers. The self-branding that millions of SNS members engage in invites an

opportunity to research a brand's self-brand congruity role in the process. This study

aims to explain self-brand congruity's role by answering some unexamined fundamental

questions involving the consumer's core drive for branded posting behavior on SNSs.

Many studies have been executed regarding usage of the Internet, but a relatively

limited number of these have been specified to SNSs. Even fewer studies have related

the self-congruity theory to SNS behavior. The following research expands upon

knowledge regarding posting behavior on SNSs, particularly Facebook. These cognitive

connections are important for marketers because their past habits are becoming

obsolete as strategies transition from mostly push to mostly pull by the consumer. It is

also important to gain a better understanding of the thought process behind brand

adoptive posting behavior because these actions carry word-of-mouth trust that is

greatly valued in today's marketing practices.

The following research collected data from undergraduate students at the

University of Florida and registered users of the SNS, Facebook. The data were









collected by an online survey method using www.surveymonkey.com. The

questionnaire involved was designed to measure Facebook familiarity and attitudes,

brand familiarity and attitudes, self-brand congruity, brand-Facebook compatibility, and

motivations for using Facebook as possible influential factors of branded adoption

behavior likelihood. Facebook was chosen as the SNS of interest in the hope of

projecting key findings onto SNSs as a whole. The two brands observed in this study

were Starbucks and Dunkin' Donuts. The resulting data were analyzed with the help of

the computer software formerly known as Predictive Analytics Software (PASW

Statistics 18 or PASW), and now known as IBM SPSS Statistics (SPSS). Correlations

and relationships between variables were measured using regression analysis to test

hypotheses and explore research questions that emerged from the review of the

literature of consumer behavior, social psychology, brand management, and new media

communications.









CHAPTER 2
LITERATURE REVIEW

The introduction of social networking sites (SNSs) to the Internet has brought with

it unchartered territory. For those with mass communication interests, SNSs have

introduced questions and possibilities in areas such as consumer behavior, social

psychology, brand management, and new media communications. The present study

aims to provide answers to a few of these questions by looking principally at motivations

and self-congruity as possible explanations for SNS behaviors.

A Digital Revolution

Social Networking sites are creating a new revolution within the broader Internet

revolution that has taken place during the last 20 years (Shirky, 2009). The Internet has

conveniently provided the platform for social networks to develop. In the following

excerpt, Shirky (2009) discussed the social networking implications of the Internet from

a filmed presentation titled How Social Media Can Make History:

The Internet is the first medium in history that has native support for groups
and conversation at the same time. Whereas the phone gave us the one-to-
one pattern, and the television, radio, magazines, books gave us the one-
to-many pattern. The Internet gave us the many-to-many pattern.... As all
media gets digitized, the Internet also becomes the mode of carriage for all
other media. That means that every medium is right next door to every
other medium. Internet. .is increasingly more a site of coordination. (3:47)

Participatory Cultures

While increased coordination has certainly facilitated globalization and changed

cognitive connections, people are primarily concerned with the cultures in which they

participate (Boyd, 2006). Thus, the focus has evolved on participatory cultures. The

Convergence Culture Consortium (2006) defines participatory culture as follows:

Participatory culture describes the way consumers interact with media
content, media producers and each other as they explore the resources









available to them in the expanded media landscape. Consumers become
active participants in shaping the creation, circulation and interpretation of
media content. Such experiences deepen the consumer's emotional
investment in the media property, and expand their awareness of both
content and brand. (p. n.p.)

This increased awareness and active participation involves uploading content, adapting

existing content, working collaboratively with others, and forming online connections as

core activities (Willett, 2009). As participants in this respect, Internet users are

considered to be "active agents" who display competency, personality, and knowledge

(Beavis, 2009).

Social networking sites play an important role in this agency and participatory

development, taking the Internet's coordination function mentioned by Shirky (2009) a

step further. An SNS is a type of online community and participatory culture which

typically follows a basic format consisting of a personal profile, a blog and/or message

board, a private messaging utility, and a news feed that reports friends' recent activities

(Boyd & Ellison, 2007; Li, 2008). Boyd and Ellison (2007) identified three defining

features of SNSs. They allow individuals to 1) construct a profile that is at least semi-

public, 2) formulate a list of other users with which they share a connection, and 3)

navigate and interact with other users' profiles within that list (Boyd & Ellison, 2007).

Many different SNSs are available these days for a variety of purposes. The

nature and nomenclature of connections via SNSs may vary from site to site, but they

all generally serve the basic function of facilitating connections (Boyd & Ellison, 2007).

For example, Linkedln, founded in 2002 (Company History, 2008), is an SNS with the

specific purpose of making business and career connections. Some SNSs serve to

unite those with certain health conditions such as Juvination, which brings Type 1

diabetics together to share stories, tips, and information. Furthermore, some sites such









as Friendster, which originally aimed to inspire more romantic connections, were

transformed by entertainers as a platform for their promotions (Boyd & Ellison, 2007).

Friendster ultimately ran into its demise by not embracing this new direction, but

MySpace was able to successfully capitalize on the networking opportunity in its wake

(Boyd & Ellison, 2007). Moreover, Facebook was initially intended to serve a niche in

the Harvard University-only community before its popularity transitioned it into a global

phenomenon.

Since their launches, MySpace and Facebook have emerged as the leading

communities with the highest number of members (Lopez, 2008). As of 2006, MySpace

totaled 100 million users and has continued to linger around 100 million unique users in

the period since (Arrington, 2009). Likewise, MySpace's highest rival, Facebook,

grossed 18 million users in 2007 with "an estimated 90% of all undergraduates at

colleges and universities where Facebook is available are registered users of the site.

Of those users, 60% log on daily" (Lopez, 2008, p. 25). In June 2009, Facebook tallied

more than 200 million active users with more than 100 million of those logging on at

least once per day (Statistics, 2009). Most recently, less than one year later, Facebook

reported more than 350 million active users in December 2009 with the average user

spending 55 minutes of his/her day on the site (Statistics, 2009).

Facebook, initially launched in February 2004, is an SNS that aims to help people

connect with others and the world around them (About Facebook, 2004). As with many

other SNSs, Facebook allows users to share information and media such as photos,

videos, and links to other online media. By providing these functions, Facebook allows

users to develop and maintain relationships with friends and family on a consistent









basis (About Facebook, 2004). Typically, these personalized networks are mostly

comprised of connections previously made offline, meaning that SNSs, Facebook in

particular, are less frequently used to make new connections (Boyd & Ellison, 2007).

Facebook is unique from other SNSs, however, in that it originally began as an SNS

solely for Harvard University students before expanding to other universities. A person

was required to have a verifiable college email address in order to register with the site.

Facebook later expanded its registration requirement to all valid email addresses by

September 2006 (Company Timeline, 2009).

According to the third party audience analysis site, www.quantcast.com (2010),

facebook.com traffic is skewed slightly female (55%) and heavily Caucasian (75%), with

the highest trafficking age group being 18 to 34 year olds (42%). Furthermore, when

compared with similar SNSs, Facebook has the highest percentage of traffic to have a

college education, accounting for 53% of Facebook traffic (facebook.com Quantcast

Audience Profile, 2010). The initial exclusivity to college students, the foundation of

college-educated users, and its top-ranked popularity among SNSs are all reasons

Facebook was chosen for this study. For a sample of primarily undergraduate college

students and recent graduates, Facebook is most likely to present respondents already

familiar with the SNS.

With such social utilities as the ones provided by Facebook at hand, SNSs can be

used for a variety of purposes. According to Boyd (2008) on the subject of political

action and SNS behavior, our society is "status-obsessed and narcissistic" and "typical

social network site users are more invested in adding glitter to pages and SuperPoking

their 'friends' than engaging in any form of civically driven collective action" (p. 241).









Research has indicated that approximately 40% of SNS users choose to include a

self-description, invitation to contact, hobbies and interests, and references to friends

and relationships in their online profiles (Van Cleemput, 2009). Posting photographs is

another highly employed self-identifying behavior observed. Lange and Lampe (2008)

discovered that 88.9% of the participants they observed displayed an identifying

personal photograph. With so much personal information available in one location,

another study found, not surprisingly, that "participants use Facebook to learn more

about others, as the level of disclosure on Facebook may be higher and have more

detail than disclosure in real life" (Lange & Lampe, 2008, p. 15).

Interchangeability between online and offline communication and activity is also a

common theme in SNS posting behavior. For example, participants tend to use

Facebook to coordinate offline, real-life activities (Lange & Lampe, 2008). Transposable

offline worlds for online worlds are corroborated by the fact that most SNS users draw

on SNSs "to gather with friends when physical co-presences is impossible or

impractical .... For many of the most active participants on social network sites,

networked publics substitute for physical publics because physical publics are

inaccessible, untenable, heavily regulated, or downright oppressive" (Boyd, 2008,

p. 242). These are a few of the SNS posting trends observed thus far. However, many

questions have been left unanswered as new trends add and transform existing ones. A

basic understanding from pre-behavior explanations to post behavior observations

would help provide detailed answers to these questions. Thus, in looking at a

foundational measure, motivations and the uses and gratifications perspective can help

attach meaning to these purposes for SNS behavior.









Uses and Gratifications

Research has indicated that motivations can be of great consequence to SNS

behavior (Beavis, 2009). The uses and gratifications theory states that varying reasons

exist to use a mass medium (Blumler & Katz, 1974). This theory explores intrinsic needs

and motivations that lead to an individual's choices to perform specific behaviors

(Blumler & Katz, 1974). The behaviors chosen serve to satisfy those needs; needs

which can vary greatly between Internet usage and for other mediums (Ko, Cho, &

Roberts, 2005). This theory assumes that media users are active users who take

actions in search of a goal (Blumler & Katz, 1974). Furthermore, Ko, Cho, and Roberts

(2005) argued that these people "are aware of their needs, and select the appropriate

media to gratify their needs" (p. 60). For example, Wu (2009) stated that regarding

political behavior on Facebook, "endorsement of either an entertainment celebrity or a

political figure indeed is first and foremost built upon strong affective affiliation, hence

rendering a strong urge in the supporters to vocally cheer their heroes" (p. 17). This

urge to express support for their heroes represents the need in this case, and since the

features of Facebook provide extended opportunity to share political ideas and other

original political materials such as videos and transcripts of speeches (Wu, 2009),

Facebook became the appropriate medium used to gratify this need.

Thus, the Internet is a growing and encouraging capacity for interaction that, as a

result, has intensified the uses and gratifications theory (Ko, Cho, & Roberts, 2005).

Korgaonkar and Wolin (1999) looked at 41 items related to Internet usage and concerns

and found indications that consumers use the Internet for retrieving not only information,

but also entertainment and escape. Furthermore, a previous study looked at the

influence of Internet usage motivations on adoption behavior toward online services.









This study showed, depending on the type of website (information, infortainment,

shopping), that different motivators were key influencers on behavior (Lin, 1999).

Applying this same logic to SNSs would indicate that different users of SNSs could

view SNSs in different capacities depending on their motivation for utility. Just as with

various websites, SNSs can serve many functions within themselves. Therefore, it

should follow that the same concept of matching different motivational influencers to

varying levels or types of behavior may be applicable within the more specific context of

SNSs. As a result, some motivations may have a greater interaction effect on SNS

behavior than other motivations.

In concert with this, the impact of motivations is further amplified by the

interactivity of the Internet. The Internet allows users to participate at an elevated level

by giving them control, according to their preferences and needs, over the advertising

messages they receive, the amount of information they give and/or receive, and the

presentation method of such content (Hoffman & Novak, 1996).

According to Papacharissi and Rubin (2000), a key component to the uses and

gratifications theory is audience activity, which is heavily impacted by motivations. Ko,

Cho, and Roberts (2005) extended this idea to say that the four main motivations of

Internet usage are entertainment, information, social interaction, and convenience, and

that each of these motivations were strongly related with distinct types of interaction.

Several other motivations for using SNSs have been explored, including time-passing,

entertainment, relationship development, relationship maintenance, trend-following

(Hall, 2009), impression management, self-disclosure, perceived reciprocity (Park, Jin,

& Jin, 2009), communication maintenance between friends, and social compensation









(Barker, 2009). Upon review of these previously researched motives for using SNSs,

the majority categorize themselves nicely within the four main motivations to use the

Internet (entertainment, information, social interaction, and convenience) set forth by

Papacharissi and Rubin (2000) in a way that best suits the purpose of this study. With

SNSs mirroring many of the same functions as the Internet, and the vast majority of

researched motivations for using SNSs fitting into these four categories, it seems

appropriate to apply the motivation categories of entertainment, information, social

interaction, and convenience to SNS behavior for the purpose of this study.

Online Profiling

The manifestation of behaviors and motivations on SNSs begins with the heart

and soul of a social networking site: the construction of a profile page (Boyd & Ellison,

2007). This personal profile serves to represent the participant on a continual basis and

acts like a personal hub for each user's participation (Goodman et al., 2008). Through

this profile creation, users provide "understandings of themselves, about themselves

and for themselves in relation to others" (Dowdall, 2009, p. 78), including in relation to

brands. This personal profile becomes a continuation of their offline social existence,

which is extended into a crafted online identity. This morphing of offline and online life

even extends to daily conversations. Conversations begun in offline situations are

continued online and vice versa. The boundaries often become blurred (Davies, 2009;

Dowdall, 2009; Goodman et al., 2008).

The idea of identity creation is not limited to basic profile creation, but also extends

to other forms of posting behavior. Willett (2009) stated that by posting content,

particularly media-related content, today's youth can be seen as creating identities.

Although previous research has shown that identities implied by online profiles are









strongly related to their matching offline identities (Van Cleemput, 2009), this inclination

does not eliminate tendencies to play with it to some degree. "Social-networking sites

which combine blogs, profiles and photo and video-sharing can be viewed as cultural

resources which are used by young people as a way of performing and perhaps playing

with their identity" (Willett, 2009, pp. 55-56). In this way, users take resources available

to them and modify, reapply and recontextualize them into a bricolagee" (Levi-Strauss,

1974) that creates a new meaning and a unique identity (Willett, 2009).

Material objects, products, and brands are often used as markers of identity (Lury,

1996). This idea is part of a bigger fan culture. The Consortium (2006) labels this "Brand

Culture" and defines these as "communities of committed consumers of specific

products and services (such as BMW drivers, iPod users and Coke collectors)" (p. n.p.).

In conjunction with this, Quart (2003) proclaimed that consumer culture pegs teens as

target consumers, but that the reverse is also true in that teens are branded objects

themselves. As a result, a new landscape is being formed where products and services

are no longer simply developed and presented to the consumer in a unidirectional flow.

Instead, these fan bases are now providing feedback productions of media, text, and

interpretations (Jenkins, 2006). Products and the content that accompanies them are

now being modified uniquely by individuals who identify with them. These personal

connections and personal influencers have more impact than previously considered.

Beavis (2009) found in her research on convergence that "with respect to convergences

of 'fan, brand and style culture' [such as on social networking sites], brand and fan

culture had less of a role to play but 'style culture', and [students'] friendships and self-

concepts mattered considerably" (p. 32).









This trend is particularly apparent among younger users. In a PBS documentary

about youth today on SNSs, a high school boy claimed, "You need to have the Internet

on to talk to your friends [because] everybody uses it. It's like a currency. If you [do not]

use it you're going to be at the loss" (Goodman et al, 2008, 6:14). The online world has

become a familiar parallel universe to many people, and as such "social networking

sites are also increasingly the place where kids hash out their conflicts" (Goodman et

al., 2008, 14:46), once again blurring the lines between online and offline life. One

student interviewed about her Facebook profile replied, "Pretty much everyone has one.

It's like a section of the Internet that is your own. Like, you can make it your personality

exactly" (13:13). One expert in the document, Pascoe, explained:

in a way, the social networking sites are a digital representation of what we
think of as adolescence. So what teens are doing are going around and
trying on these different identities. "I'm a goth", or "I'm a surfer" or "I'm a
punk rocker", or "I'm a 'this' or 'that'", and the Internet's allowed them to
display that identity in a very dramatic and very distinct way." (18:26)

This kind of play is adopted by adults as well, particularly through connectivity.

"Connectivity is increasingly part of adults' working and social lives; like the young, they

use the Internet as a way of exploring new ways of connecting socially ." (Davies,

2009, p. 110).

Pascoe suggested that an opportunity for play with regard to identity is one

component of profiling (Goodman et al., 2008). Thus, profiling can be described as

more of a means-oriented practice as opposed to a goal-oriented practice that can be

applied by SNS users of all ages (Bauman, 2005). As a means-oriented practice, a user

is likely to utilize a very liquid structure, in which a profile can be modified as often and

as much as a user prefers. This means that a profile is continually evolving, just as the









actual user. In this case, content remains only so long as the user still feels it is

congruent with his/her own current identity to some degree (Dowdall, 2009).

The New Word-of-Mouth

Social networking sites have also impacted word-of-mouth practices. As a result of

the many-to-many pattern discussed previously (Shirky, 2009) and the practice of

identity creation using products and brands as markers of such identity, SNSs are a

prime medium for advanced word-of-mouth opportunities, which can be appealing to

marketers. Word-of-mouth is already a highly valued asset for marketers. In the SNS

environment, reach is significantly large, snowball effects thrive from multi-directional

conversations, and electronic recording of activity translates into word-of-mouth that is

more traceable and measureable than ever before (Trusov, Bucklin, & Pauwels, 2009).

Attaining this word-of-mouth advantage presents a powerful opportunity. Due to the

augmented coordination capacity of the Internet and online identity creation trends, as

mentioned earlier, users can now act as producers and consumers (Shirky, 2009). In

this way, traditional advertising efforts are being restructured to a pull strategy, which

means that companies are increasingly placing consumers in control.

Lester Wunderman, considered by many to be the creator of today's direct

marketing, thought direct mail to be similar to a game of shooting a target in the sense

that the mail is the bullet that hits the target (selected consumers) every time

(Steinbock, 2000). The Internet presents itself to this utility as uniquely opportunistic due

to its adoption rate, ubiquity, and constant access to and ability to distribute virtually

limitless content (Arndt, 2001). The realm of consumer relationships has consequently

been altered by the moderation of the Internet. Regarding the birth of the Internet as an

advertising medium, Wunderman progressively stated, "I wanted something different. I









wanted a medium where the advertiser would become the target and the consumer

would become the shooter. That's what's happening. It is a profoundly different

marketplace" (Steinbock, 2000, p. 127).

This vision has matured with the emergence of major SNSs such as MySpace and

Facebook in the period since Wunderman's statement in 2000. The interaction changes

accelerated by SNSs have redefined how people engage with brands by allowing the

relationships to be more personal and individualized as if the company is simply another

person to befriend (Boyd, 2006).

Wunderman stated that "with the rise of interactivity, more companies would have

to become relationship marketers and win new clients one at a time. In the process, ...

computers would substitute human interactivity" (Steinbock, 2000, p. 129). Social

networking sites facilitate this individualized relationship marketing by attaining clients

and consumers one "friend" or "fan" at a time, creating more customized and focused

relationships with consumers who freely elect to do so. The creator and CEO of

Facebook, Mark Zuckerberg, observed this trend by saying,

[In] the last hundred years ... the way to advertise was to get into the mass
media and push your content. .. In the next hundred years information
[will not] just be pushed out to people, it will be shared among the millions
of connections people have. Advertising will change. You will need to get
into these connections. (Holzner, 2009, p. 1)

Discovering the factors behind SNS branded posting behavior will bring marketers one

step closer to being in the heart of these connections.

As previously stated, the consumer is therefore increasingly in control, but the

many-to-many communication pattern is also allowing brands to mold their products and

marketing efforts to better fit their consumers according to this consumer feedback

(Shirky, 2009). Previously marketers simply strove to influence and promote









word-of-mouth about their brands. Consumers are showing-through participation-that

they are willing to talk. They are willing to adopt brands, ideas, and concepts as their

own, as defining characteristics of themselves (Boyd & Ellison, 2007). With the

development of SNSs, marketers can still promote branded word-of-mouth, but now

have the power of the many-to-many discussions that characterize SNSs behind them

and an increased ability to compactly observe and quantify results (Trusov, Bucklin, &

Pauwels, 2009).

Boyd (2008) argued on behalf of social activists that "given the typical friend

overlap in most networks, many within those networks hear the same thing over and

over until they believe it to be true [which gives] the impression that activists have

spread a message further than they have" (p. 243). This not only applies to social

activists, but also branded word-of-mouth. In addition to this increased effectiveness,

consumers voluntarily open doors on SNSs for brands to share valuable content and

promotions by electing to be "friends" or by becoming a "fan" of the brand. While

marketers' goals are still to get people to talk about their brand, the difference is that on

an SNS those people have a place highly conducive to discussions about that brand,

and brands can even join the conversation. "Just as different national identities have

been mixed in the hybrid [of online and offline worlds], so too the realms of business

and culture are converging in novel ways" (Consalvo, 2006, p. 120). These transitions

make understanding consumer connections, motivations, and behavior even more

relevant.

Self-Congruity Theory

Two theories, the uses and gratifications theory and the self-congruity theory, help

explain what factors contribute to an individual's branded activity on SNSs. As









previously discussed, the uses and gratifications theory provides a focus on how

motivating factors are likely to influence the usage of a medium to gratify specific needs

(Blumler & Katz, 1974). Thus, in alignment with this theory, it should follow that a

person's motivation for using an SNS is likely to impact his usage behavior.

To supplement this approach in a socially influenced setting, image concepts and

perceptions can be very influential on public behavior decisions (Shaw & Costanzo,

1970). Thus, a self-congruity theory perspective could contribute insight into posting

behavior. The self-congruity theory applies self-congruity effects on self-expression to

the attitude theory by suggesting favorable predispositions toward brands that enhance

or confirm perceptions of their own self-image (Sirgy, 1986). Essentially, it is a social

cognition theory due to the fact that an awareness of "self" serves a fundamental role

and application in life experience (Rychlak, 1981). The attitude theory explains

consumers' favorability (likes and dislikes) in connection with consumer needs, which

ultimately drives consumption behavior (Sirgy, 1986). Subsequently, self-congruity is

the match or mismatch perception associated with a consumer's comparison between

self-image and other stimuli such as brand image, product image, or company image

(Sirgy, 1986). Moreover, Parker (2005) explained the implications of this in terms of

brand attitudes:

Self-brand congruity, the comparison of self to brand, affects brand
attitudes particularly when the social signaling value of a brand is high (i.e.,
used in a public situation) and when symbolic, self-expressive motivations
are involved. Positive brand attitudes should result as the similarity
between brand image and the consumer's self-image increases (p. 4)

Self-brand congruity as used by Parker (2005) and Sirgy (1986), is the congruity match

between self-image and brand-image. Sirgy et al. (1997) used the phrasing "typical user

of a brand" to measure brand-image instead of simple and direct references to one's









image of a brand itself because "personal images of a product reflect the stereotype of

the generalized users of that product and are determined by a host of factors such as

advertising, price, and other marketing and psychological associations" (p. 229).

In his research Sirgy (1986) examined self-image and its comparison to the actual

perceptions of characteristics of the "I" or "me," as Rogers (1959) defined "self," as well

as the perception of the ideal self, which refers to how a person would like to view

himself. Self-image is a subset of "self" based on self-awareness that can change in

different social roles (Sirgy, 1986). This construct is a key indicator for understanding

consumer behavior (Parker, 2005). Rogers (1959) stated that individuals are motivated

by a basic and fundamental actualizingg tendency" that one hopes will, at the very least,

maintain, if not enhance a person's self-image. Furthermore, Franken (1994) indicates

the structural basis on how the concept of self becomes actualized:

There is great deal of research which shows that the self-concept is,
perhaps, the basis for all motivated behavior. It is the self-concept that
gives rise to possible selves, and it is possible selves that create the
motivation for behavior. (p. 443)

Thus, both the uses and gratifications theory and self-congruity theory provide

rounded reasoning for a resulting conclusion that the more a person thinks that a brand

is congruent with his own self-images, the more he will be motivated to behave in ways

that reinforce those self-images on SNSs for various gratification purposes. If a person's

natural tendency is to self-actualize according to his motivations as Rogers (1959)

claimed, it seems probable that such a person would display these behaviors publicly

via his own profiles and through social interaction on SNSs.









Hypotheses

This study expanded upon previous theoretical models involving antecedents of

media usage, particularly SNSs, and congruity between an individual and a brand within

such a medium. The constructs here are dependent upon familiarity and loyalty with the

brand and the medium. Familiarity and loyalty are then used to shed insight on a

person's motivation and likelihood to integrate a brand on his SNS profile as a

personally identifying characteristic. The operational definitions in this case are

important to clarify due to the colloquially interchangeable use of terms such as "online

communities," "social networking utilities," "online social networks," and "social

networking sites." For the purposes of this study "social networking sites" (SNSs) will be

the term used. Accordingly, the concepts to be measured in this theoretical

development include self-brand congruity, brand-SNS compatibility, brand attitude, SNS

usage motivations, and finally the resulting brand adoption behavior to one's SNS

activity.

A brand, according to the American Marketing Association is a "name, term, sign,

symbol, or design, or a combination of them, intended to identify the goods and services

of one seller or group of sellers and to differentiate them from those of competition"

(Keller, 2008, p. 2). However, Keller (2008) widened this definition to a more universal

industry concept by stating that a brand is also "something that has actually created a

certain amount of awareness, reputation, prominence, and so on in the marketplace"

(Keller, 2008, p. 2). Though he limited this brand development to the marketplace,

people in general are increasingly branding themselves through the phenomenon of

online profiling; creating the awareness, reputation, and prominence that Keller (2008)

referred to. It is this consumption need of self-expression that often drives consumers to









"purchase brands (e.g., Gucci and Rolex) that communicate a particular image or social

role" (Parker, 2005, p. 10). Thus, they are creating a brand for themselves often through

a selected compilation of other brands in the marketplace. Some insight into this self-

expression practice can be provided by the self-brand congruity concept.

Self-brand congruity is based upon attitudes regarding oneself and a brand, or

rather, the typical user of a brand. Attitudes are typically researched in marketing and

advertising because they have shown to perceptively predict consumer behavior

(Mitchell & Olson, 1981). The core influences of behavior are the cognitive building

blocks, or beliefs, which are considered to be fundamental in developing attitudes

(Fishbein & Ajzen, 1975).

Thus, an individual's beliefs are likely to be highly translated into his introspective

attitudes and their attitudes toward brands. Since attitudes have been shown to mold

behavior in past consumer behavior research, it is not unreasonable to presume that

these beliefs and attitudes can be translated into measureable, personally identifying

actions (Parker, 2005). Self-brand congruity is defined by this perceived alignment of

attitudes toward a brand and the beliefs that make up their identity. For the purposes of

this study, the resulting behavior likelihood is defined as the likely or unlikely adoption of

a brand to be presented in any fashion on an individual's Facebook activity or profile

page, referred to as "brand adoption behavior likelihood" in this study. This behavior is

proposed to be indicative of a person's actual and ideal self-brand congruity with the

particular brand.

S Hypothesis 1: Self-brand congruity (both actual and ideal) will be positively related
to the likelihood of SNS brand adoption behavior.









It is possible, however, that self-brand congruity may be present, yet not be

exhibited on an SNS because the brand or product category may not be considered

suitable for a public persona so openly broadcasted. For example, a product may not be

deemed as relevant, applicable, appropriate, or natural for Facebook topics. Thus, the

individual may not consider a brand or product category compatible with Facebook.

This possible lack of compatibility must be addressed in an effort to make this research

more reliable. Thus, the more an individual considers a brand as being relatable and

suitable for Facebook, the more likely that person will be to display such accordance.

S Hypothesis 2: Brand-SNS compatibility will be positively related to the likelihood
of SNS brand adoption behavior.

As previously stated, in order for a brand to be selected by the individual for such

self-expressive profiling, it is important that he have a favorable consumer attitude

toward it. The same could also be true for the reverse situation. If a person has a strong

unfavorable attitude toward a brand, he may consider that anti-congruity to be an

integral part of his personal characteristic set as well. With this assumption in mind, the

current study will assumed this association to be true and executed evaluations solely in

positive terms.

Parker (2005) declared, "In general, consumers tend to favor brands and products

that satisfy their needs and wants better than competitive choices" (p. 10). This

statement implies a working application of the uses and gratifications theory. The four

primary motivations for Internet usage laid out by Kaye and Johnson (2001) and

Papacharissi and Rubin (2000) were information, convenience, entertainment, and

social interaction. As previously discussed, these four fundamental Internet usage









motivations can be applied more specifically to the use of SNSs to derive insight on

their effect on behavior in this new context.

Within these four motivations, previous studies have indicated that media with

higher interactive value or usage value (such as entertainment or information), tend to

induce higher motivation in consumers to use them (Ko, Cho, & Roberts, 2005).

Concurrent with this finding, the entertainment and information motivations should

inspire a positive influence on brand adoption behavior likelihood.

* Hypothesis 3: An entertainment SNS motivation will positively interact with both
actual and ideal self-brand congruity in predicting SNS adoption behavior
likelihood.

* Hypothesis 4: An information SNS motivation will positively interact with both
actual and ideal self-brand congruity in predicting SNS adoption behavior
likelihood.

Furthermore, social interaction is highly impacted by perceived social

consequences. Associations are therefore carefully expressed. One study found that

when users express their political affiliations, they are typically prompted to justify their

choices (Wu, 2009). Such accountability and social judgment would likely inflict a sense

of reserve and prudence in freely admitting to associations unless the individual feels

very strongly about them. Moreover, if a person is using an SNS out of convenience

motivations, it seems unlikely that the person would exhibit the effort of associating

himself with a brand in extended behavior or exhibiting critical evaluation involved in

justifying such associations (Wu, 2009). Therefore, both of these motivations can be

expected to stimulate a negative influence on the addition of a brand to a person's SNS

posting behavior. Thus the following research questions arise:

* Research Question 1: Do social interaction SNS motivations negatively interact
with actual and ideal self-brand congruity in predicting SNS brand adoption
behavior likelihood?









* Research Question 2: Do convenience SNS motivations negatively interact with
actual and ideal self-brand congruity in predicting SNS brand adoption behavior
likelihood?









































Figure 2-1 Overview of the study.


Entertainment
(H3)

Information
(H4)

Social Interaction
(RQ1)

Convenience
(RQ2)


Interaction Effect
(H3, H4, RQ1, RQ2)









CHAPTER 3
METHOD

The purpose of this study was to explain branded posting behavior on SNSs. Such

answers would ideally provide brand management answers for how to excel using

social media as a pivotal resource. The goal was to gain insight from active SNS

participants and consumers that could unlock new approaches to brand management.

This study executed this goal primarily by looking at self-brand congruity measures, but

also by observing participants' motivation(s) for posting on an SNS as an interacting

effect. The data for this research were collected using an online survey with self-

administered questionnaires. The variables measured in reference to the leading SNS,

Facebook, included 1) familiarity with Facebook and the brand, 2) attitudes toward

Facebook and the brand, 3) current Facebook behavior, 4) image perceptions regarding

self-brand congruity as well as brand-SNS compatibility, and 5) their resulting adoption

behavior likelihood.

Research Design

This study employed an online survey to fulfill its research purpose. The

questionnaires for this survey were available for completion on www.surveymonkey.com

from May 15, 2009 through May 26, 2009 (Appendix B). Sample participants received

an invitation to complete a questionnaire, which directed them to the appropriate link

(Appendix A).

In completing the questionnaire, each participant was asked to evaluate himself,

an individual brand, and Facebook on qualities such as familiarity, image perception,

self-congruity, and compatibility between the brands and Facebook. These constructs

were assessed using Likert scales with five levels. Variables were typically coded with









the most positive response equaling one, and more negative responses were assigned

incrementally increasing values. The data for each questionnaire were coded and

recorded into a SPSS (formerly PASW) database. The data were then analyzed for

correlations with actions taken in reference to the brand on Facebook through

regression analysis. Descriptive and frequency analyses were also executed to

supplement these observations.

Participants

The sampling method employed in this case was a non-probability purposive

convenience sample. Purposive sampling depends on the researcher to select

participants-within their convenience and judgment-who most appropriately represent

the targeted sample (Babbie, 2007). To obtain a valid sample of SNS users within the

scope and constraints of this study, purposive sampling was combined with the

snowball sampling method, where "a researcher collects data on the few members of

the target population he or she can locate, then asks those individuals to provide the

information needed to locate other members of that population whom they happen to

know" (Babbie, 2007, p. 185). This combination was especially appropriate for this study

because it asked the population, Facebook users, to participate in a way that mirrored

the kind of SNS behavior in which the study was interested. Thus, to fulfill the purpose

set out for this research, a total of 151 respondents participated in the study. Of these

151 participants, 74 responses qualified for inclusion in analysis of this survey. These

participants were registered users of Facebook and coffee drinkers who were at least

18 years of age.









Data Collection and Procedure

Surveys are used for studies with descriptive, explanatory, and exploratory

purposes where the units of study are typically individuals as opposed to groups or

interactions (Babbie, 2007). Babbie (2007) also stated that "survey research is probably

the best method available to the social researcher who is interested in collecting original

data for describing a population too large to observe directly," and it can be an

"excellent vehicle for measuring attitudes and orientations in a large population"

(p. 244). Privacy limitations of SNSs and the vast population of individual SNS users

eliminate the possibility of observing the entire population. As a result, survey research

that relies on a representative sample to characterize the population was the most

effective method to provide the attitudinal and behavioral insight desired in this study.

This method was utilized to gain exploratory insight on participants due to the

succinctness of both the method and the medium. Focus groups are typically most

appropriate for exploratory inquiries where limited research on the specific subject in

question has been done previously and inductive reasoning is required. However, they

carry a risk of the chameleon affect occurring. The chameleon affect is the

nonconsciouss mimicry of the postures, mannerisms, facial expressions, and other

behaviors of one's interaction partners, such that one's behavior passively and

unintentionally changes to match that of others in one's current social environment"

(Chartrand & Bargh, 1999, p. 893).

Instead, an online survey allowed for much of the same questions to be asked as

in a focus group or personal interview, but with distinct and definitive answers that

mirrored the SNS medium in question. Thus, minimal, if any, valuable insight was lost.

The objective of this exploratory study was to obtain measurable and meaningful









information from participants by identifying motivational, attitudinal, and behavioral

patterns.

To provide sound explanation for the sampling method chosen, it is important to

clarify why Facebook was the social networking site chosen for this research. As

explained earlier, SNSs often target specific groups of users. Therefore, in selecting a

manageable representation of SNSs, the scope of this research was narrowed down to

Facebook for a variety of reasons. First, Facebook is an SNS with one of the highest

number of members and the most registered activity by facilitating interaction and

sharing of news and media between individuals and groups (Lopez, 2008). Beyond this

top-ranked popularity, Facebook's users are fairly diverse demographically, but are

highly saturated among college students due to its initial exclusivity as a utility solely for

college students (About Facebook, 2004). Facebook's qualities position it as most

aligned with the available convenience sample population, but also most likely to

provide respondents who are familiar with the specific social utility. The sample chosen

mirrors the users of Facebook.

The first execution of this sampling was done via an invitation requesting

Facebook users' participation in the study which was posted directly on Facebook.

Facebook viewers of the invitation were encouraged to repost it and pass it along to

others within their network. This invitation explained what the study regarded, as well as

what participation meant for them in terms of time, effort, and confidentiality.

A second purposive convenience sample was used in conjunction with the first

sample. The sample population for this group was college undergraduate students

between the ages of 18 and 25, enrolled in Journalism and Communication classes at









the University of Florida for the 2008-2009 academic year. The same invitation to

participate was distributed to students in three classes within the University of Florida's

Mass Communication department. Students participated voluntarily and, at the

discretion of each class's professor, the classes were given the opportunity to earn

course credit for the completion of the questionnaire. To keep the questionnaire within a

reasonable length, one questionnaire was dedicated to each of the two brands used.

Participants were directed, according to last name, to one of the two questionnaires to

complete.

Similar to the carefully chosen SNS for this research, the brands observed in the

questionnaires were also well considered. Without establishing a level of familiarity with

a brand, answers to subsequent questions regarding that brand would carry little

meaning, if answers were possible at all. Thus, familiarity was an important founding

construct of this study, and finding brands with which participants were highly familiar

improved the research validity. As a result, two brands were chosen: Starbucks, a

well-known brand internationally, and Dunkin' Donuts, a well-known coffee provider in

the United States, but with a slightly different consumer base. The product category of

coffee distributors was chosen for its prominence among college students, as it remains

a staple in many college students' lives. It was also ubiquitous enough that regardless

of coffee consumption levels, the sample recruited would likely have attitudes and

beliefs regarding the brands within the category. In fact, in a study about the differences

between in-home coffee drinking and in-store consumers, 46% of respondents ordered

non-coffee products, which implies additional reasons for developing attitudes toward









coffee shops than the mere liking of coffee (Henson, 2007). Furthermore, according to

Parker's (2005) protests, Starbucks was among the top 25 publicly consumed brands.

A pre-test was executed to determine the most appropriate brands for the study.

Brands tested in the pre-test included two national companies, Starbucks and Dunkin'

Donuts, and three local companies in Gainesville, Florida, Barnie's Coffee & Tea Co.,

Maude's Classic Cafe, and Lollicup Coffee & Tea. The pre-test was very brief and

included questions evaluating familiarity, attitudes, self-brand congruity, and brand

compatibility with Facebook. Starbucks and Dunkin' Donuts proved to be the most

appropriate brands for this study. They both had similar high levels of familiarity, yet

attitude, congruity, and compatibility answers showed that they did not have the same

group of consumers.

Further analysis for the actual study included a review of each brand's set of

responses and entering the data into PASW. Frequency analysis and simple summary

statistical tests were applied to each brand group individually, as well as the combined

sample data, to provide a descriptive sample profile and a summary of existing

Facebook activity. Subsequently, several comparisons of means analyses were

executed to establish significant differences between the two brands regarding

demographics, behaviors, and psychographics. Among the variables assessed in these

comparisons were independent variables of brand attitude, actual and ideal self-brand

congruity, and brand-Facebook compatibility. Next, bivariate correlation and multiple

regression tests were performed to determine whether to accept or reject each

hypothesis, as well as to explore possible answers to the research questions. From this

synthesized information, the statistical tests were evaluated on whether the data









provided sufficient answers regarding the correlation between each independent or

interaction variable (explained in more detail in the following sections) and the

dependent variable-adoption behavior likelihood on Facebook. The results of this

research are detailed in Chapter 4.

Measures and Instrument

The materials required for this study included a questionnaire (see Appendix B), a

computer with Internet access, and an invitation to complete the questionnaire (see

Appendix A). The invitation was brief, describing the general context of the study and

participation, as well as providing the link where the students could find and complete

the questionnaire.

As previously stated, this study included two sets of questionnaires: one set of

questions regarding Starbucks and the same set of questions regarding Dunkin' Donuts.

This split was done to shorten the length of the questionnaire with the intention of

increasing completion rates. Furthermore, within the questionnaires, it was important to

set background context with a list of questions regarding the participants' behavior

tendencies in their current Facebook usage. The questionnaire also targeted topics

about attitudes, congruity, compatibility, and adoption behavior. Some questions asked

participants to role play in a given situation; other questions simply asked participants to

what degree they agree or disagree with an item.

Each questionnaire consisted of eight sections. These sections aimed to establish

1) participant qualification; 2) motivations for using Facebook; 3) familiarity with, usage

level of, and attitudes toward Facebook; 4) existing user habits on Facebook; 5)

familiarity and attitudes toward the brand; 6) self-brand congruity and brand-Facebook

compatibility; 7) adoption behavior likelihood of the brand; and 8) a demographic profile.









More specifically, the survey questionnaire began with a series of qualification and

background questions regarding Facebook. To qualify to complete the entire

questionnaire, participants were required to be coffee drinkers, and also have a

registered Facebook account. Question 3 asked for the date of initial membership with

Facebook to discover the possible level of expertise and give more meaning to activity

on these sites on the basis of membership length.

Questions 17 through 28 focused on the participants' level of familiarity with,

usage of, and attitudes toward Facebook. This section consisted of 5-point semantic

differential questions based on questionnaires by Douglas (1999) and Kang (1999). Of

these differential questions, Questions 19 through 22 ascertained how active

participants are on the site. Question 21 indicated participants' approximate total

activity time on Facebook per week. Although this question was separated into two

variables for hours and minutes in data collection, the two were converted into a single

hourly measurement and reported in hours. Questions 29 through 38 established the

user's current habits on Facebook and operationalize the current Facebook behavior

variable. In doing so, the results of the 10 questions for each participant were summed

to create a current Facebook behavior index where a maximum score of "44" indicates

the least behavior and familiarity and a minimum score of "10" indicates the most

intense behavior practices. This measure helped to give proper significance to results of

the research. For example, if a participant starts out at a high level of activity, it shows

that this individual is willing to display information about himself. The same idea applies

to the other extreme.









Independent Variable

Two official independent variables were established in this study. These were 1)

self-brand congruity (Hypothesis 1) where both actual and ideal self-brand congruity

were observed, and 2)brand-Facebook compatibility (Hypothesis 2). However, an added

interest was observed in brand attitude as a corroborating independent variable. This

observation was implemented to confirm the theoretical progression from brand attitude

to self-brand congruity and finally, adoption behavior as laid out by previous research.

Brand attitude and differences between brand groups were explored for a

preliminary assessment of the appropriateness of measures with alignment to previous

research and an assessment of different brands' influences on self-brand congruity and

brand-Facebook compatibility. Confirming a lack of significant differences between

brand groups lends more justification to examining the variables of self-brand congruity

and brand-Facebook compatibility outlined for Hypothesis 1 and Hypothesis 2. On the

other hand, confirmed significant differences between brand groups would indicate a

need to keep brand influence in mind when considering the results of these tests.

Furthermore, to rule out significant differences between brands would mean that both

brand samples could be evaluated together as one combined sample. This would add to

the overall sample size and reliability of the research.

To give further depth to the evaluation of the independent brand groups, familiarity

with the brand was assessed briefly, and brand attitude was considered to a further

extent. To establish familiarity with the brand and measure brand attitude, the chosen

brands, Starbucks and Dunkin' Donuts, were introduced in each parallel questionnaire

by asking semantic differential questions (Questions 39 through 43). Familiarity was

addressed with a single semantic differential on a 5-point scale from "1" ("very familiar")









to "5" ("very unfamiliar"). Furthermore, attitude toward the brand was also measured via

four semantic differential questions (40 through 43) rating the following four qualities for

the brand: 1) good/bad, 2) pleasant/unpleasant, 3) favorable/unfavorable, and 4)

likable/unlikable. These questions were also adopted from questionnaires by

Papacharissi and Rubin (2000), Douglas (1999), Kang (1999), and Parker (2005). This

method of evaluation is commonly used to measure brand attitude (MacKenzie, Lutz, &

Park, 1989). The results of the four questions were averaged to create an index for

operationalization of the brand attitude variable with a reliable Cronbach's alpha

calculated in previous research as .83.

Next, the independent variable of self-brand congruity was addressed in the latter

half of the questionnaire. This section linked the individual, the environment, and the

brands by asking questions regarding the participants' image perceptions and level of

congruity with the brands within the context of Facebook. The initial questions for this

section (questions 44 through 48) were adopted from the global self-congruity scales

method proposed by Sirgy (1997) and further executed in Parker's self-congruity

research (2005). This method first determined brand user imagery descriptors and

subsequently measured self-brand congruity based on the descriptors cataloged. The

respondent was first asked to provide two adjectives to describe the typical user of the

brand in question. Next, the respondent was asked questions to establish the level of

congruity he felt between his own actual and ideal self-images and the descriptors he

just used to describe the typical user of the brand as follows:

Take a moment to think about [Brand x]. Think about the kind of person
who typically uses the brand. Imagine this person in your mind and then
describe this person using one or two personal adjectives to describe the
typical user of the brand. Now indicate your agreement or disagreement to









the following statements: The typical user of [Brand x] is consistent with
how I see myself (actual self congruity); The typical user of [Brand x] is
consistent with how I like to see myself (ideal self congruity); The [Brand x]
brand is compatible with Facebook (brand-Facebook compatibility).

As labeled above, the first of these questions is intended to measure actual

self-congruity, while the second question is intended to measure ideal self-congruity.

The congruity indicator scores collected from both of these questions were averaged to

create a solitary self-brand congruity index to simplify analysis and discussion, provided

no significant difference occurred between the actual and ideal congruity responses. If

significant differences occurred, however, actual and ideal congruity were treated

separately as individual indicators of self-brand congruity. This method measures

self-brand congruity directly, as opposed to performing separate evaluations of self-

image and brand (or product-user) image (Sirgy et al., 1997). The method proposed by

Sirgy et al. (1997) also took a holistic and global perspective on the measurement by

allowing participants to conjure up their own perception descriptors for typical brand

users instead of presenting them with predetermined image perceptions to evaluate

congruity (Sirgy et al., 1997). High or low self-brand congruity was determined by the

level to which the participant indicated he perceived a match (or mismatch) between his

own self-image (actual and ideal) and the perceived image of the typical user of the

brand. In this study, high self-brand congruity was indicated with scores closer to 1,

while scores closer to 5 indicated low self-brand congruity.

Brand-Facebook compatibility, another independent variable, was subsequently

measured as part of this set by asking the participant to indicate his level of agreement

or disagreement that the brand is compatible with Facebook. This measurement used a

solitary 5-point Likert-scale ranging from "1" ("strongly agree") to "5" ("strongly









disagree"). The identical questions were asked for both Starbucks and Dunkin' Donuts

on their respective questionnaires for each of these independent variables.

Dependent Variable

The dependent variable in this study was brand adoption behavior likelihood on a

Facebook user's posting activity. This variable was examined in the final portion of the

questionnaire with two questions. Questions 49 and 50 asked the respondent to role

play in various Facebook situations. First, in Question 49, the respondent was simply

asked whether he would add the brand to his profile to which he must answer his level

of agreement on a 5-point scale which was coded from "1 ("no, definitely not") to "5"

("yes, definitely"). Thus, a higher number indicated a higher likelihood of adoption

behavior.

Since research is limited on specific posting behavior intention, this first dependent

variable question was based on measures of intended voting behavior. For example,

the following question was utilized to determine candidate preference in a study that

examined the impact of negative political television commercials, "If you were voting in

this election, after seeing this commercial, would you vote for the sponsoring

candidate?" Respondents were then asked to indicate their selection on a scale ranging

from "no, definitely not" (-3) to "yes, definitely" (+3) (Tinkham & Weaver-Lariscy, 1993).

As reinforcement to this measurement, the next question asked about more

detailed posting behavior decisions. A list of 18 possible brand adoption behaviors, with

positive, neutral, and negative brand attitudes, were presented to the respondent, along

with an "Other" option. Ten of these items were denoted as positive actions regarding

the brand, five were considered neutral, and three were considered to be negative

actions. Neutral and negative actions were included to provide a more well-rounded









pool of possible actions reflective of typical Facebook behavior. This list of 18 behaviors

was chosen because these behaviors provided a possible range typical of Facebook

users, which gave the opportunity for a user to publicly express or display a brand in his

Facebook activity. The participant was asked to check all the actions he was likely to

take. Since this study aimed to identify positive effects on behavior, the 10 positive

action items were the focus for data analysis. To operationalize this dependent variable,

a value of "1" was assigned if the participant indicated he would likely take that action,

and a value of "0" was assigned if the action was not selected. Responses to the 10

positive items were tallied to create an aggregate positive action score. This aggregate

score from "0" to "10" was added to the score from the previous question to create a

total adoption behavior likelihood index. This index had possible scores from "1" to "15"

as a result. In this case, the higher the adoption behavior likelihood index score, the

more likely a person was to take action regarding the brand in his Facebook activity.

Question 50 was originally intended to modify the model developed by Curry

(2004) which measured decision-making bases in which each answer was assigned a

positive, negative, or zero value. These values were aggregated to create four

composite scores for each of the items measured in Curry's research (Curry, 2004).

This aggregated scoring was adapted to the current study with a few changes. The

main modification made to this model was that essentially one item was examined in

this section of the questionnaire and multiple response options were not mutually

exclusive. Thus, answer options were increased and one composite score was created

as a result of this modification. This model was chosen because users could perform

more than one of the proposed behaviors simultaneously, any of which could potentially









express varying attitudes toward the indicated brand on a positive-negative continuum.

As with Curry's model, positive and negative aggregate scoring could be applied in

future research to determine a correlation between negative brand attitude, self-brand

congruity, and behaviors with negative attitudes, or to observe inaction tendencies as

opposed to posting with positive or negative attitudes toward a brand.

Motivation Interaction Variables

The interaction variables in this research explored the interacting effects of various

motivational reasons for using Facebook. A set of 13 questions addressed the four main

motivations set forth by Papacharissi and Rubin (2000). These motivations were

information (Questions 4 through 6), convenience (Questions 7 through 9),

entertainment (Questions 10 through 13), and social interaction (Questions 14 through

16). Each of the four motivations had three questions dedicated to its measurement

except for entertainment which included four questions in the set. The participant was

asked to indicate his level of agreement with these 13 statements on a 5-point scale

from "1" ("strongly agree") to "5" ("strongly disagree"). Results for each question within a

motivation set were averaged to create an index score for that motivation. As a result,

lower motivation indexes indicated that a participant was more motivated by that usage

purpose. This model indicated sufficient reliability with a collective Cronbach's alpha

score of .78 (Ko, Cho, & Roberts, 2005). Motivational indices were then multiplied by

the self-brand congruity indices to create an interaction variable for each motivation.

Self-brand congruity indices, individual motivation indices, and interaction variables

were all entered into multiple regression analyses for each of the four motivations

examined in Hypothesis 3, Hypothesis 4, Research Question 1, and Research

Question 2.









CHAPTER 4
RESULTS

Data Analysis

Sample Profile

This study recruited 151 participants for the online survey. However, due to the

nature of the study and the qualifying measures required to produce valid data, 77

respondents were eliminated from the final analysis of results. Respondents were

removed for two reasons: 1) respondents' voluntary incompletion of the survey and 2)

respondents' not drinking coffee and/or not having a registered Facebook account.

Therefore, many respondents were removed because they did not fit one or both of

these qualifications, and thus were unable to answer the questions relevant to the core

purpose of this study. Taking these eliminations into account, the final valid sample

includes 74 respondents. Of these 74 participants, 41 took the questionnaire with

questions regarding Starbucks, and 33 took a questionnaire with questions regarding

Dunkin' Donuts.

Table 4-1 displays the summary description of the Starbucks, Dunkin' Donuts, and

total sample. The Starbucks and Dunkin' Donuts groups were not significantly different

(Table 4-1). Mostly female, participants ranged in age from 18 to 41 with the majority 18

to 24. These statistics are in alignment with Facebook's original target market age range

of 18 to 24. Just over 80% of the participants were in the process of earning an

undergraduate degree or had already completed a degree. More than half of the

respondents were Caucasian.









Facebook Tendencies/Activity

Means for familiarity, acceptability and likelihood to visit Facebook were high for

both groups (Table 4-1), with no significant difference between groups. Likewise, with

no difference between groups, 85.2% of the respondents use Facebook at least on a

daily basis (once per day plus multiple times per day). Participants' activity time on

Facebook averaged 6.09 hours per week.

Participants' level of public expression on Facebook was measured for direct

(Table 4-2) and indirect expression of information (Table 4-3). The direct expression

informational categories typically completed on Facebook included Basic Information,

Personal Information, Contact Information, and Educational and Work Information.

Among these direct expression behaviors, the most completed and detailed section was

the Basic Information section with 25% of the respondents stating it to be fully

completed, whereas respondents typically indicated they provide very limited

information in the Personal, Contact, and Work and Education Sections. This would

imply a general sense of privacy and inaction with respect to directly providing

descriptive details about themselves. This also supports the tendency for members to

primarily use SNSs to maintain current relationships (with those who would likely

already have such information such as contact, work, education and personal

information) rather than to search out new relationships (Boyd & Ellison, 2007).

The indirect expression of information was represented by a display of

associations with other items that they either searched for or were prompted to add by

other Facebook users. These items included Groups, Fan items, Bumper Stickers, Gifts,

and Other items. Groups came out as the most common behavior by far for each of the

three samples with 91.9% of the combined sample noting that they had elected to









become a member of at least a few groups, if not many groups. Gifts sent from other

Facebook users, Other Items, and Fan Items also presented a majority of users who

reported they showed a few of these items on their profile. Bumper stickers found by

the participants themselves exhibited the least common behavior with a majority

choosing to display none in that category.

Overall Facebook behavior was calculated by summing each respondent's current

profile behavior with a maximum possible score of "44" (indicative of lesser activity) and

a low possible score of "10" (indicative of greater activity) (Table 4-3). This

measurement resulted in total current activity scores between 23.0 and 24.0 for both

Starbucks and Dunkin' Donuts individually, as well as for the combined sample. On the

scale in this study from most active ("10") to least active ("44") these scores fell on the

more active side of the median, demonstrating moderate to high behavioral tendencies

for each group. No significant differences occurred between groups for total current

Facebook behavior or any of the individual behaviors.

Brand Attitudes, Congruity, and Compatibility

Consistent with previous use of the brand attitude measures and suggested

acceptable levels, between .7 and 1.0 (Davis, 1997), the items used for the brand

attitude index produced a Cronbach's alpha of .95 overall, .92 for the Starbucks group,

and .97 for the Dunkin' Donuts group (Table 4-4). While the data showed no significant

difference of familiarity between the brands, the attitude toward Starbucks was

significantly more positive than the attitude toward Dunkin' Donuts. Self-brand congruity

components and brand-Facebook compatibility also returned significant differences

between Starbucks and Dunkin' Donuts (Table 4-4).









Although no significant differences occurred between actual and ideal self-brand

congruity ratings for Dunkin' Donuts, Starbucks's actual self-brand congruity ratings

were significantly greater than Starbucks's ideal self-brand congruity ratings. Since at

least one of the two observed brands revealed significant differences between these

self-brand congruity components, both actual and ideal self-brand congruity measures

will be used as separate variables in each sample group for the remainder of this

study's statistical analysis.

In addition to significant differences between actual and ideal self-brand congruity,

significant differences were also present between groups. Starbucks's respondents

rated Starbucks as significantly more congruent with their actual and ideal selves than

Dunkin' Donuts' respondents rated their actual and ideal selves. Lastly, Starbucks was

perceived to be significantly more compatible with Facebook than Dunkin' Donuts.

Since participants also had significantly more positive attitudes toward Starbucks than

Dunkin' Donuts, this could be indicative that if a person feels negative associations

toward a brand, he feels more strongly that the brand is not appropriate for SNS

discussion or favorable in making positive social impressions and vice versa.

Independent, Dependent, and Interaction Variables

Table 4-5 presents a summary of this study's key variables, means, and reliability

measures. For multiple item indicators, Cronbach alphas were reported, and the

Pearson correlation coefficient was reported for the dual components (actual and ideal)

of self-brand congruity. Compatibility was a single item measure. While most of the

alphas met the acceptable minimum (.70) (Davis, 1997), several did not (Convenience

and Social Interaction). However, given the sample sizes' link to unstable correlations, it

was decided that it was better to include the scales rather than delete them-keeping









the low alphas in mind when the data were analyzed. Significant differences between

brands were observed for each of the independent variables and for one motivational

variable (Information). The other three motivations showed no significant difference

between brands.

Dependent Variable: Adoptive Behavior

The dependent variable, adoption behavior likelihood, consisted of two measures.

The first was a single item scale asking the respondent to indicate if he would either

visually or verbally add the brand to his Facebook ("1" = "no, definitely not" to "5" = "yes,

definitely") (see "Likely of Adding" in Table 4-7). The second measure, which was added

to the first measure to create the calculated dependent variable, consisted of the sum of

all the positive actions a participant indicated he was likely to take regarding the brand

on Facebook. These two behavioral measures were significantly and moderately-

strongly correlated for the combined sample and for each brand individually (Table 4-6),

indicating these were appropriate measures.

Table 4-7 shows descriptive statistics and frequencies for each of the two

measures used to calculate the dependent variable, adoption behavior likelihood. Total

positive actions were significantly higher for Starbucks than for Dunkin' Donuts. In a

more detailed look at each of the positive actions, two action items displayed

significantly higher results for Starbucks than for Dunkin' Donuts. These action items

included sending a brand logo gift to someone and adding a bumper sticker of the brand

logo that was sent by another user.

The three actions most frequently indicated for the combined group were to

"accept and display a brand logo gift that someone sent you," "write/post/display

something positive about a brand in your profile or status," and "become a fan of the









brand on Facebook." These items also appear within the most frequent responses for

both the Starbucks and Dunkin' Donuts sample groups. The action least frequently

indicated was to actively search out and join a group with a positive attitude toward the

brand. This item was followed closely by the act of searching for the brand's Facebook

profile and more information regarding the brand. The results in Table 4-7 show that

participants were least likely to fulfill action items requiring a participant to actively go

out of his way to seek out the brand or media feature.

Hypothesis Testing

This study explored possible relationships among a number of variables within the

concepts of consumer behavior, social psychology, brand management, and new media

communications. The constructs at the heart of this study, which have been briefly

explored thus far, included self-brand congruity, brand-SNS compatibility, and the four

SNS motivation variables of entertainment, information, social interaction, and

convenience. For the scope of this study, the best way to uncover answers to the

hypotheses between these variables was by executing regression analyses to

determine relationships.

* Hypothesis 1: Self-brand congruity (both actual and ideal) will be positively related
to the likelihood of SNS brand adoptive behavior.

First, a bivariate correlation test was performed examining indexes for actual and

ideal self-brand congruity, current Facebook behavior, brand attitude, and adoption

behavior likelihood (Table 4-8). Adoption Behavior Likelihood is inversely coded from

the remaining variables. Negative correlation values are therefore indicative of positive

relationships.









This bivariate correlation examination highlighted a few key observations. First,

current Facebook behavior showed no significant correlation, and thus no relationship

with either self-brand congruity or brand attitude. Furthermore, though a significant

correlation was observed between current behavior and adoption behavior likelihood for

the combined sample, evidence showed only a weak positive relationship (R = -.27).

These apparent discrepancies suggest that positive attitudes and highly perceived

self-brand congruity may be highly correlated with likely adoption behaviors indicated by

participants, but this, in turn, may not necessarily translate strongly into fulfilled

behavior.

Second, brand attitude displayed a significant moderate relationship with actual

self-brand congruity, ideal self-brand congruity, and adoption behavior likelihood for all

three sample groups. This result further confirms previous research and exemplifies

these variables as good measures. Third, since brand attitude and adoption behavior

likelihood had a significantly moderate relationship, it reasonably follows that actual and

ideal self-brand congruity would also illustrate a significant moderate relationship with

the dependent variable. This was, in fact, the case for the combined sample and

Starbucks. Dunkin' Donuts, however, showed only ideal self-brand congruity to have a

significant moderate relationship with adoption behavior likelihood. The important

overall take-away from these observations was that as brand attitude and both actual

and ideal self-brand congruity increased positively, adoption behavior likelihood also

increased, revealing a significant and positive moderate relationship.

With these confirmations in mind, Hypothesis 1 further examined the components

of self-brand congruity using bivariate correlations (Table 4-9). For the purpose of this









test, actual and ideal self-brand congruity measures were recorded such that "1" =

"strongly disagree" and "5" = "strongly agree", in order to have positive correlations

depict a positive relationship and vice versa. Also, since preliminary statistical tests

showed significant differences between groups for both actual and ideal self-brand

congruity, bivariate correlation tests were run for both the Starbucks and Dunkin' Donuts

samples in addition to the combined sample.

Three separate correlations were examined with regard to both actual and ideal

self-brand congruity. The first set looked at participants' intent to add the brand to their

Facebook profile either verbally or visually. This measurement showed significant

moderately positive correlations with ideal self-brand congruity for Starbucks, Dunkin'

Donuts, and the combined sample. But for actual self-brand congruity this was only the

case for the Starbucks and combined sample.

The second set observed the correlation with the sum of positive actions which

participants indicated they would likely take. Both actual and ideal self-brand congruity

revealed significant, slightly weaker moderate correlations than in the previous

observation set for both Starbucks and the combined sample. However, Dunkin' Donuts

was not significant at the p < .05 level for either actual or ideal self-brand congruity this

time.

The third set calculated dependent variable for this study-adoption behavior

likelihood-was compared with the self-brand congruity components. As previously

stated, adoption behavior likelihood was calculated by adding the score of the user's

intent to add the brand to his Facebook activity (min = 1, max = 5) (Question 49 of the

questionnaire) and the sum of positive actions (min = 0, max = 10) which the participant









indicated he would take regarding the brand on Facebook (Question 50 of the

questionnaire). The bivariate correlation results were similar to the first set regarding

intent to add the brand to Facebook activity. Both actual and ideal self-brand congruity

correlations, with adoption behavior likelihood were significantly positive with moderate

strength for both Starbucks and the combined sample. Dunkin' Donuts however,

returned ideal self-brand congruity as significantly and moderately correlated with ideal

self-brand congruity, but not with actual self-brand congruity. These moderately positive

correlations indicate that as a participant perceives increased self-brand congruity (both

actual and ideal), his adoption behavior likelihood also increases. Starbucks's

correlation strengths were consistently greater for actual self-brand congruity, whereas

Dunkin' Donuts showed the opposite to be true, though only two of the Dunkin' Donuts

correlations were significant. Overall, this test showed that both actual and ideal

self-brand congruity were positively related to adoption behavior likelihood. Hypothesis

1 was accepted as a result.

S Hypothesis 2: Brand-SNS compatibility will be positively related to the likelihood
of SNS brand adoption behavior.

Just as in the analysis for self-brand congruity in Hypothesis 1, brand-Facebook

compatibility for Hypothesis 2 was also recorded such that increased compatibility was

associated with increasing numerical values ("1" = "strongly disagree" and

"5" = "strongly agree") to keep positive correlations depicting a positive relationship.

Hypothesis 2 was explored with the same process as the latter part of Hypothesis 1 with

bivariate correlations. As with Hypothesis 1, comparisons with brand-Facebook

compatibility were observed with both intent to add the brand to one's Facebook profile

and positive action items indicated by the participant that they would likely take, as well









as the calculated adoption behavior likelihood (Table 4-10). Since t-tests showed

significant differences between groups for brand-Facebook compatibility, each brand

was observed separately in addition to the combined sample.

Observations of the intent to add the brand to their Facebook activity or profile

showed a lack of significance for weakly correlated Starbucks, Dunkin' Donuts, and

combined samples. These same results were similarly repeated showing no

significance for the sum of positive adoptive actions and also for adoption behavior

likelihood. Thus, compatibility did not prove to be correlated, positively or negatively,

with brand adoption behavior likelihood, which denoted that a relationship between

these two variables was not present. Unlike self-brand congruity results in Hypothesis 1,

brand-Facebook compatibility did not demonstrate a significant positive relationship with

adoption behavior likelihood, resulting in the rejection of Hypothesis 2.

Interaction Effect of SNS Motivations

The secondary aspect explored in this study was the interaction effect of four

usage motivations for SNSs, including entertainment, information, social interaction, and

convenience. The dependent variable for the following two hypotheses and two

research questions was the calculated overall adoption behavior likelihood addressed

previously in Hypotheses 1 and 2. Since the independent and interaction variables were

inversely coded from the dependent variable, a negative coefficient indicated a positive

relationship, and vice versa, for each of the four motivational effects explored in

Hypothesis 3, Hypothesis 4, Research Question 1, and Research Question 2.

Differences between Starbucks and Dunkin' Donuts's samples for the individual

entertainment, social interaction, and convenience motivations were not significant. As a

result, a greater focus was placed on the combined sample for these examinations than









for the information motivation. The differences between groups for the information

motivation were significant, meaning that a greater weight needed to be assigned to

each brand individually, especially for the interaction with self-brand congruity

measures. Since significant differences occurred between groups for both actual and

ideal self-brand congruity, brand groups were still given individual attention for not only

the information motivation, but also the entertainment, social interaction, and

convenience motivations.

S Hypothesis 3: An entertainment SNS motivation will positively interact with both
actual and ideal self-brand congruity in predicting SNS adoption behavior
likelihood.

Hypothesis 3 addressed interaction effects between an entertainment motivation

and self-brand congruity measures with a multiple regression model (Table 4-11). The

independent variables for this analysis were actual and ideal self-brand congruity,

entertainment motivation, and the interaction effect between the entertainment

motivation and both actual and ideal self-brand congruity.

The combined sample equation was statistically significant [F(5,68) = 4.06*,

*p < .01]. This model produced an R value of .48, indicating a moderate correlation

between actual and ideal self-brand congruity, entertainment motivation, both their

interaction effects, and the dependent variable, adoption behavior likelihood. R-squared

was .23, meaning that 23% of variance in these five variables is explained by adoption

behavior likelihood. Likewise, the Starbucks equation was also statistically significant

[F(5,35) = 5.12*, *p < .01] with a moderate correlation of .65 and 42% of variance

explained by the dependent variable. Since both of these equations were significant, a

linear relationship occurred between the variables for each sample group and these two

equations can be projected onto the population. In contrast to the Starbucks and









combined samples, the Dunkin' Donuts equation was not statistically significant, and

thus, no linear relationship was present for this sample.

Although both the Starbucks and combined sample equations proved significant,

independently, the only significant relationship with adoption behavior likelihood was for

entertainment motivation for the Starbucks sample (t = -2.96, p < .05). However, this

means that for each sample group, both the individual self-brand congruity variables

and the interaction effects between entertainment motivation and actual and ideal

self-brand congruity were not significant. Thus, the two interactions between the

entertainment motivation and both actual and ideal congruity have no unique

contribution to predicting a Y-value for the regression equation. As a result, Hypothesis

3 is rejected.

S Hypothesis 4: An information SNS Motivation will positively interact with both
actual and ideal self-brand congruity in predicting SNS adoption behavior
likelihood.

The same multiple regression tests from Hypothesis 3 were applied in the

exploration of Hypothesis 4 with entertainment motivation replaced with information

motivation. The independent variables for this analysis were actual and ideal self-brand

congruity, information motivation, and the two interaction effects between information

motivation and the two self-brand congruity components (Table 4-12). Significant

differences between brands for both the individual information motivation variable and

the self-brand congruity measures provided reason to carefully consider each brand on

an individual basis.

For this hypothesis, Dunkin' Donuts returned a regression equation that was not

significant. In addition, this examination produced no significant individual relationships

with adoption behavior likelihood for the Dunkin' Donuts sample. However, as with the









previous motivation, both Starbucks [F(5,35) = 7.66*, *p <.01] and the combined sample

[F(5,68) = 6.13*, *p <.01] produced significant linear regression equations. Thus, linear

relationships are present for these sample groups and these regression equations can

be projected onto the population. Starbucks showed a strong correlation (R = .72)

between the inputted self-brand congruity, motivation, and interaction variables, and the

dependent variable, with 52% of variance explained by adoption behavior likelihood.

Likewise, the combined sample revealed a moderate correlation between these

variables (R = .56), explaining 31% of the variance present.

The individual information motivation variable was significantly related to adoption

behavior likelihood for both the combined sample (t = -3.12, p < .01) and the Starbucks

sample (t = -3.04, p < .01). In fact, information motivation was the only variable with a

significant relationship with the dependent variable for the combined sample. The

negative coefficients for the information variable for these sample groups indicate a

positive effect on adoption behavior likelihood.

Furthermore, while Dunkin' Donuts produced no significant individual relationships,

Starbucks returned actual self-brand congruity (t = -2.87, p < .05) and the interaction

variable between actual congruity and information motivation (t = 2.18, p < .05) to also

have significant relationships with adoption behavior likelihood. Although not significant

for the combined sample, these variables were the most important variables for the

model with Beta values of -1.43 and 1.35, respectively. Partially consistent with

predictions for these variables, the negative coefficient for actual congruity implied that

actual self-brand congruity is positively related to adoption behavior likelihood.









However, the positive coefficient for the interaction between actual congruity and

information motivation indicated a negative effect.

The other half of the self-brand congruity measures (ideal congruity) showed no

significant correlations with adoption behavior likelihood individually nor by interacting

with information motivation. Thus, these variables had no unique contribution to

predicting a Y-value for the model's regression equation. As a result, a positive

information motivation interaction with both actual and ideal self-brand congruity cannot

be statistically confirmed and Hypothesis 4 is rejected. Supplementary to this, the

differences between Starbucks and Dunkin' Donuts' results signify that the effect of

information motivation and the interaction effect between information motivation and

self-brand congruity vary per brand, so each brand needs to be considered on an

individual basis in future studies.

Research Questions

The research questions that naturally follow these hypotheses are simply a

continuation of the interaction effects of the final two SNS usage motivations, social

interaction, and convenience. Since these variables were expected to have negative

impacts on behavior toward a brand, they were posed as research questions to be

explored, not to provide definitive statistical answers. However, multiple regression

analyses were performed for both the social interaction motivation and convenience

motivation in the same manner as in Hypothesis 3 and Hypothesis 4, but with expected

negative influences rather than positive influences.

S Research Question 1: Do social interaction SNS motivations negatively interact
with actual and ideal self-brand congruity in predicting SNS brand adoption
behavior likelihood?









Research Question 1 entered actual and ideal self-brand congruity, social

interaction motivation, and the two interactions between each self-brand congruity

component and social interaction motivation to have potential significant relationships

with adoption behavior likelihood (Table 4-13). Social interaction motivations were

expected to negatively impact the effect of both actual and ideal self-brand congruity on

adoption behavior likelihood. Significant differences between brands for the self-brand

congruity measures provided reason to carefully consider each brand on an individual

basis.

As with the combined samples of Hypothesis 3 and Hypothesis 4, the combined

sample returned a significant linear regression equation [F(5,68) = 5.28*, *p <.01], but

no individual significant individual relationships, meaning that they had no unique

contribution to predicting a Y-value for the regression equation. However, Starbucks

and Dunkin' Donuts showed varying results.

Starbucks returned both a significant regression equation [F(5,35) = 9.29*,

*p < .01], as well as individual significant correlations between independent variables

and adoption behavior likelihood. The regression equation indicated a strong correlation

between the independent and dependent variables (R = .76), accounting for 57% of the

variance present. The significant relationships with adoption behavior likelihood for this

sample were with actual self-brand congruity (t = -3.39, p < .01), the independent social

interaction motivation (t = -3.66, p < .01), and the interaction between social interaction

motivation and actual congruity (t = 2.82, p < .01). Consistent with expectations, the

positive coefficient for the interaction between ideal self-brand congruity and social

interaction motivation indicates a negative influence on adoption behavior likelihood.









Dunkin' Donuts also returned significant individual variables, including both actual

congruity (t = 2.11, p < .05), ideal congruity (t = -2.05, p < .05), and the interaction effect

between social interaction motivation and actual congruity (t = -2.01, p < .05). The

negative coefficient for the interaction between actual self-brand congruity and social

interaction motivation reveals a positive influence on adoption behavior likelihood, which

contradicts the predicted negative impact, and contrasts the Starbucks result. However,

the overall regression equation for the Dunkin' Donuts sample was not significant. Thus

no linear relationship existed between the inputted variables and adoption behavior

likelihood, and the equation cannot be projected onto the population.

While the Starbucks and combined sample models were statistically significant,

individual variables complicated this analysis. For both brands, social interaction did

significantly interacted with actual self-brand congruity. However, besides the fact that

the directional influences of these two results contradicted each other, the second

self-brand congruity component, ideal congruity, revealed that it did not have a

significant interaction effect with social interaction motivation for either brand.

Consequently, the data do not confirm that social interaction motivation negatively

interacts with both self-brand congruity components in predicting adoption behavior

likelihood. However, the significant interaction effects between actual self-brand

congruity and social interaction motivation give reason to explore this avenue further.

Actual self-brand congruity may be a significant factor in this model, whether it is

independently or interacting with social interaction motivation.

* Research Question 2: Do convenience SNS motivations negatively interact with
actual and ideal self-brand congruity in predicting SNS brand adoption behavior
likelihood?









The final motivation observed was convenience motivation and its interaction

effect with actual and ideal self-brand congruity on adoption behavior likeliness (Table

4-14). Variable effects were observed individually per brand due to significant

differences observed between groups for self-brand congruity measures.

As with the previous motivations, the equations for both Starbucks

[F(5,35) = 7.35*, *p < .01] and the combined sample [F(5,68) = 5.45*, *p < .01] were

statistically significant allowing the linear relationship to be projected onto the

population. The regression equation for Dunkin' Donuts was not significant and

therefore cannot be projected onto the population. Starbucks showed a strong

correlation (R = .72) accounting for 51% of variance. The combined sample followed

with a moderate correlation (R = .54) with 29% of variance accounted for by adoption

behavior likelihood.

Although Dunkin' Donuts did not return any of the independent or interaction

variables to be significant, Starbucks and the combined sample showed discrepancies

on this matter. Similar to the information motivation results, both Starbucks (t = -3.97,

p < .01) and the combined sample (t = -2.81, p < .05) returned the independent

convenience motivation as significantly correlated with adoption behavior likelihood.

This was the only significant relationship for the combined sample, but the data for the

Starbucks sample illustrated that in addition to the independent convenience motivation,

its interaction effect with ideal self-brand congruity also had a significant impact on

adoption behavior likelihood (t = 2.04, p < .05).

Negative coefficients for both the combined sample and Starbucks independent

convenience motivation variable implied a positive effect on adoption behavior









likelihood, contrary to expectations. On the other hand, the interaction effect between

convenience motivation and ideal congruity for the Starbucks sample displayed a

positive coefficient, meaning that convenience motivation negatively interacted with

ideal self-brand congruity in predicting adoption behavior likelihood.

Despite the fact that this component of the interaction variable set was consistent

with predictions for this research question, it was the only interaction effect for any of

the sample groups that was significant. The lack of significance for this motivation's

interactions with actual congruity for Starbucks, and with both actual and ideal congruity

for Dunkin' Donuts and the combined sample, indicate that they have no unique

contribution to predicting a Y-value in the regression model equation. Hence, this

multiple regression analysis revealed that while the convenience motivation may

individually have a significant relationship with adoption behavior likelihood, no

significant negative interaction effect between convenience motivation and self-brand

congruity components on adoption behavior likelihood was confirmed. This was true of

all four motivational interaction variables examined in this study. Also, similar to the

previous three motivations examined, the differences between Starbucks and Dunkin'

Donuts's results indicated that the effect of convenience motivation and its interaction

effect with self-brand congruity measures varied according to brand.

A summary of the significant relationships with adoption behavior likelihood for

each motivation's multiple regression analysis made in Hypothesis 3, Hypothesis 4,

Research Question 1, and Research Question 2 for the combined sample (labeled

as"'C.S."), Starbucks (labeled as "S"), and Dunkin' Donuts (labeled as "D.D.") indicates

that social interaction had the highest number of significant relationships, followed by









information, convenience, and lastly, entertainment (Table 4-15). A look at each sample

indicates that Starbucks returned the most variables significantly related to adoption

behavior likelihood, including all four independent motivation variables. Actual

self-brand congruity proved to have a significant relationship with adoption behavior

likelihood more often than ideal self-brand congruity. Interaction effects showed this

same trend.

Limitations

This study is limited by several factors. First, relatively low variances accounted for

by the models act as a noteworthy hindrance in this study, as do the low reliabilities of

the individual motivation and compatibility variables. Second, the sample includes

students from the College of Journalism and Communication at the University of Florida.

These students, being already clearly interested in communication, may be more

interested and involved with SNSs and higher posting behavior than students in other

disciplines. Third, the issue of online privacy could have been an influential factor, as

indicated by many responses about current profile habits, but was too broad to explore

in the scope of this study with due diligence.

Fourth, due to the nature of the study where qualifying components had to be met

to fulfill the questionnaire to completion, the resulting sample size was lower than

preferred which lead to questionable reliability measures. Fifth, for the portion of the

sample that was extended on Facebook, a more reliable sample would have been

beneficial from a more national and global audience. If this were the case, brands

chosen for observation would need to be represented in those selected geographical

areas. Sixth, due to limiting the length of the questionnaire, brands were chosen to be

restricted to one product category and participants were presented with only one of the









two brands. A broader brand collection and a direct comparison between brands

presented at once would have been helpful in achieving a broader perspective of

branded behavior.

The questionnaire itself included several limitations. In addition to the items

mentioned previously, if it had been possible to return to the questionnaire and modify it,

changes would have included discarding the coffee drinker qualification to increase

sample size, since many non-coffee drinkers also have developed attitudes toward

well-known brands and their typical users. A modified questionnaire also would have

included more questions to measure compatibility to obtain a reliability score for the

variable. Finally, the question regarding posting behavior was limited by an unequal

number of items for positive, negative, and neutral responses. As a result, only positive

actions were sufficiently observed. Future research observations regarding simple

action versus inaction and positive action versus negative action could provide added

valuable insight on this subject. Such research modifications would likely introduce new

variables in the effect of brand on self-brand congruity measures and adoptive SNS

posting behavior that were beyond the scope and capabilities of this study.









Table 4-1. Sample profile summary statistics
Starbucks Dunkin' Donuts Total sample Sig.
(n=41) (n=33) (n=74)
Mean S. D. Mean S. D. Mean S.D.
Age 22.15 3.69 23.12 4.21 22.58 3.93 NS*
Age group # % # % # % NS**
18-24 35 85.4 25 75.8 60 81.1
25-34 5 12.2 7 21.2 12 16.2
35+ 1 2.4 1 3.0 2 2.7
Gender # % # % # % NS**
Male 15 36.6 11 33.3 26 35.1
Female 26 63.4 22 66.7 48 64.9
Education # % # % # % NS**
High School 10 24.4 2 6.1 12 16.2
AA in progress 1 2.4 -- -- 1 1.4
AA -- -- 3 9.1 3 4.1
Bachelor in progress 22 53.7 17 51.5 39 52.7
Bachelor 5 12.2 5 15.2 10 13.5
Graduate 2 4.9 6 18.2 8 10.8
PhD in progress 1 2.4 -- -- 1 1.4
Ethnicity # % # % # % NS**
Asian 1 2.4 2 6.1 3 4.1
Black/African American 4 9.8 4 12.1 8 10.8
Hispanic/Latino 9 22.0 3 9.1 12 16.2
White/Caucasian 26 63.4 24 72.7 50 67.6
Other 1 2.4 -- -- 1 1.4









Table 4-1. Continued


Starbucks Dunkin' Donuts Total sample Sig.
(n=41) (n=33) (n=74)
Mean S.D. Mean S. D. Mean S.D.
Facebook familiarity 1.51 .60 1.48 .83 1.50 .70 NS*
Facebook acceptability 1.76 .70 194 .79 1.84 .74 NS*
Facebook likely to visit 1.61 .89 1.30 .47 1.47 .74 NS*
Facebook frequency # % # % # % NS**
Multiple times/day 29 70.7 25 75.8 54 73.0
Once per day 4 9.8 5 15.2 9 12.2
Multiple times/week 4 9.8 3 9.1 7 9.5
Once per week 2 4.9 0 0.0 2 2.7
One per month 1 2.4 0 0.0 1 1.4
LT once per month. 1 2.4 0 0.0 1 1.4
Facebook current behavior # % # % # % NS**
Browsing 15 36.6 10 30.3 25 33.8
Interacting 22 53.7 17 51.5 39 52.7
Adding profile info 4 9.8 6 18.2 10 13.5
Mean S. D. Mean S. D. Mean S.D.
Facebook total activity time 6.69 10.58 5.35 4.43 6.09 8.39 NS*


*One-way ANOVA
**Chi-square test
aFacebook familiarity. Facebook acceptability and Facebook likely to visit were
measured on scale of 1 to 5 where "1" = very familiar, very acceptable, and definitely
will visit
bFacebook total activity time is reported in hours, e.g, 6.69 = 6 hours and .69 of an hour
(41.4 minutes)









Table 4-2. Current Facebook profile behavior (direct expression)
Starbucks Dunkin' Donuts Total sample Sig.
Facebook profile- basic # % # % # % NS
Full 8 19.5 11 33.3 19 25.7
Almost full 15 36.6 12 36.4 27 36.5
Very limited 16 39.0 10 30.3 26 35.1
None 2 4.9 -- -- 2 2.7
Facebook profile- personal # % # % # % NS
Full 2 4.9 4 12.1 6 8.1
Almost full 12 29.3 7 21.2 19 25.7
Very limited 23 56.1 21 63.6 44 59.5
None 4 9.8 1 3.0 5 6.8
Facebook profile- contact info # % # % # % NS
Full -- -- 1 3.0 1 1.4
Almost full 8 19.5 7 21.2 15 20.3
Very limited 27 65.9 21 63.6 48 64.9
None 6 14.6 4 12.1 10 13.5
Facebook profile- education/work # % # % # % NS
Full 8 19.5 8 24.2 16 21.6
Almost full 13 31.7 11 33.3 24 32.4
Very limited 18 43.9 12 36.4 30 40.5
None 2 4.9 2 6.1 4 5.4









Table 4-3. Current Facebook profile behavior (indirect expression)
Starbucks Dunkin' Donuts Total sample Sig.
Facebook profile- groups # % # % # % NS
Many 15 36.6 10 30.3 25 33.8
A few 23 56.1 20 60.6 43 58.1
None 3 7.3 3 9.1 6 8.1
Facebook profile- fan items # % # % # % NS
Many 4 9.8 5 15.2 9 12.2
A few 18 43.9 17 51.5 35 47.3
None 18 43.9 11 33.3 29 39.2
Not familiar 1 2.4 -- -- 1 1.4
Facebook profile- bumpers from others # % # % # % NS
Many 7 17.1 7 21.2 14 18.9
A few 18 43.9 8 24.2 26 35.1
None 15 36.6 18 54.5 33 44.6
Not familiar 1 2.4 -- -- 1 1.4
Facebook profile- bumpers by self # % # % # % NS
Many 2 4.9 3 9.1 5 6.8
Afew 10 24.4 3 9.1 13 17.6
None 28 68.3 27 81.8 55 74.3
Not familiar 1 2.4 -- --1 1.4
Facebook profile- gifts from others # % # % # % NS
Many 1 2.4 6 18.2 7 9.5
A few 29 70.7 9 27.3 38 51.4
None 11 26.8 18 54.5 29 39.2
Facebook profile- other # % # % # % NS
Many 2 4.9 3 9.1 5 6.8
A few 26 63.4 18 54.5 44 59.5
None 13 31.7 12 36.4 25 33.8
Mean S. D. Mean S. D. Mean S.D. Sig.
Total current Facebook activity 23.83 4.33 23.30 3.95 23.60 4.14 NS
aThe sum of all Facebook actions from Tables 4-2 and 4-3 where max = 44 and min = 10









Table 4-4. Brand attitudes, brand congruity, and compatibility


Brand familiarity
Brand attitude index
Good-bad
Pleasant-unpleasant
Favorable-unfavorable
Likeable-unlikeable
Alpha
Actual self-brand congruityb
Ideal self-brand congruityc
Pearson correlation
Compatibilityd


Starbucks
(n=41)
Mean
1.76
1.98
2.00
1.95
2.20
2.00
.92
2.56
2.76
.83
2.46


Dunkin' Donuts Total sample
(n=33) (n=74)


S.


1
1


D. Mean
.97 2.09
.59 2.35
.71 2.33
.74 2.39
.78 2.52
.71 2.42
.97
.05 3.18
.11 3.30
.81
.67 2.82


S


1
1


. D. Mean
.91 1.91
.80 2.15
.96 2.15
.70 2.15
.94 2.34
.97 2.19
.95
.16 2.84
.21 3.00
.83
.95 2.62


S.D. Sig.
.95 NS
.71 p<.05
.84 --
.75 --
.86 --
.85 --

1.14 p<.05*
1.18 p<.05**

.82 p<.05


aAttitude toward Starbucks was significantly more positive than attitude toward Dunkin'
Donuts (t = -2.29; df = 72; p < .05)
bStarbucks actual self rating was significantly greater than Starbucks ideal self rating (t = -
1.95, df = 40; p < .05) (paired t-test)
cDunkin' Donuts actual self rating was not significantly different from its ideal self rating
(paired t-test)
*Actual self-congruity was significantly greater for Starbucks than Dunkin' Donuts
(t = 2.39, df = 72; p < .05).
**Ideal self-congruity was significantly greater for Starbucks than Dunkin Donuts (t = 2.00,
df = 72; p < .05).
dCompatibility between Starbucks and Facebook was significantly higher than
compatibility between Dunkin' Donuts and Facebook (t = -1.88; df = 72; p < .05)









Table 4-5. Summary statistics for independent and motivation variables
Starbucks Dunkin' Donuts Total sample
Independent variables Mean Reliability Mean Reliability Mean Reliability Sig.
Brand attitude 1.98 a = .92 2.35 a = .97 2.15 a = .95 p < .05a
Actual congruity 2.56 3.18 2.84 p < .05b
Ideal congruity 2.76 3.30 3.00 p < .050
Compatibility 2.46 NA1 2.82 NA1 2.62 NA1 p < .05d
Motivation variables Mean Reliability Mean Reliability Mean Reliability Sig.
Entertainment 1.90 a =.72 1.89 a = .56 1.90 a = .65 NS
Convenience 2.17 a =.64 2.20 a =.68 2.19 a =.65 NS
Social interaction 2.59 a =.53 2.46 a =.48 2.54 a =.50 NS
Information 3.63 a =.86 3.27 a = .70 3.47 a =.82 p < .05e
aAttitude toward Starbucks was significantly more positive than attitude toward Dunkin'
Donuts (t = -2.29; df = 72; p < .05)
b Actual self-congruity was significantly greater for Starbucks than Dunkin' Donuts
(t = 2.39, df = 72; p < .05).
c Ideal self-congruity was significantly greater for Starbucks than Dunkin Donuts (t = 2.00,
df = 72; p < .05).
dCompatibility was measured by a single 1-5 item where 1 = Very Compatible.
Compatibility between Starbucks and Facebook was significantly higher than compatibility
between Dunkin' Donuts and Facebook (t = -1.88; df = 72; p < .05)
elnformation motivation was significantly higher for Dunkin' Donuts participants than for
Starbucks participants (t = 1.74; df = 72; p < .05)

Table 4-6. Dependent variable component correlations
Starbucks Dunkin' Donuts Total sample
Dependent variable
R Sig. R Sig. R Sig.
Intent to add brand to Facebook with .71 p < .05 .72 p < .05 .72 p < .05
adoptive behavior
Brand attitude with combined .71 p < .05 .38 p < .05 .56 p < .05
adoption behavior likelihood









Table 4-7. Dependent variable measures
Starbucks Dunkin' Donuts Total sample
Mean Alpha Mean Alpha Mean Alpha Sig.
Likely of adding (single semantic scale: 5
2.20 NA 1.82 NA 2.03 NA NS
= "Definitely")
Positive adoptive behavior (n=10) # % # % # % Sig.
Accept and display a brand logo gift that
16 39.0 8 24.2 24 32.4 NS
someone sent you
Write/post/display something positive
13 31.7 7 21.2 20 27.0 NS
about brand in profile or status
Become a fan of the brand on Facebook 9 22.0 9 22.0 18 24.3 NS
Send a brand logo gift to someone 10 24.4 3 9.1 13 17.6 p<.05
Add/display Facebook bumper sticker of
10 24.4 3 9.1 13 17.6 p<.05
brand logo someone sent you
Accept an invitation to join a group with
8 19.5 4 12.1 12 16.2 NS
positive attitude toward brand
Send someone else a bumper sticker of
7 17.1 2 6.1 9 12.2 NS
the brand logo
Add/display Facebook bumper sticker of
5 12.2 1 3.0 6 8.1 NS
brand logo you found
Search out brand Facebook profile and
4 9.8 1 3.0 5 6.8 NS
more info
Search out and join group with positive
2 4.9 1 3.0 3 4.1 NS
attitude toward the brand
Mean S.D. Mean S.D. Mean S.D. Sig.
Total positive behaviors 2.05 2.29 1.18 1.69 1.66 2.08 p<.05*
*Total positive behaviors regarding Starbucks were significantly higher than total positive
behaviors regarding Dunkin' Donuts (t = 1.87; df = 72; p < .05)










Table 4-8. Correlations
Self-brand Adoption
congruity Current Brand behavior
Actual Ideal behavior attitude likelihood1
Combined Actual congruity 1.00 .83** .07 .58** -.43**
sample Ideal congruity .83** 1.00 .12 .57** -.39**
Current behavior .07 .12 1.00 .13 -.27*
Brand attitude .58** .57** .13 1.00 -.56**
Adoption behavior likelihood -.43** -.39** -.27* -.56** 1.00
Starbucks Actual congruity 1.00 .83** -.03 .60** -.47**
Ideal congruity .83** 1.00 .06 .48** -.36*
Current behavior -.03 .06 1.00 .17 -.28
Brand attitude .60** .48** .17 1.00 -.71**
Adoption behavior likelihood -.47** -.36* -.28 -.71** 1.00
Dunkin' Actual congruity 1.00 .81** .25 .50** -.29
Donuts Ideal congruity .81** 1.00 .24 .60** -.37*
Current behavior .25 .24 1.00 .14 -.32
Brand attitude .50** .60** .14 1.00 -.38*
Adoption behavior likelihood -.29 -.37* -.32 -.38* 1.00
1Negative correlation values indicate positive relationships due to inverse coding
**. Correlation is significant at the 0.01 level (2-tailed)
*. Correlation is significant at the 0.05 level (2-tailed)

Table 4-9. Self-brand congruity correlations
Hypothesis 1 Starbucks Dunkin' Donuts Total sample
Actual self-brand congruity and intent to 49** 33 44**
add brand to Facebook
Ideal self-brand congruity and intent to add 39* 41*
brand to Facebook
Actual self-brand congruity and positive .42** .24 .38**
action items
Ideal self-brand congruity and positive .31* .33 35**
action items
Actual self-brand congruity and adoption 47** 29 43**
behavior likelihood
Ideal self-brand congruity and adoption .36* 37* 39**
behavior likelihood
**. Correlation is significant at the 0.01 level (2-tailed)
*. Correlation is significant at the 0.05 level (2-tailed)









Table 4-10. Compatibility correlations

Hypothesis 2
Compatibility and intent to add brand
to Facebook
Compatibility and positive action
items
Combined intention and adoption
behavior likelihood


Starbucks
R Sig.

.26 .10

.21 .19

.24 .13


Dunkin' Donuts
R Sig.

.20 .27

.28 .12

.26 .14


Total sample
R Sig.

.07 .57

.04 .76

.05 .68









regression for entertainment motivation variable (H3)


Combined sample (Constant)
Actual self-brand congruity
Ideal self-brand congruity
Entertainment
Actual congruity X
entertainment
Ideal congruity X entertain nment
Starbucksb (Constant)
Actual self-brand congruity
Ideal self-brand congruity
Entertainment
Actual congruity X
entertainment
Ideal congruity X entertain nment
Dunkin' Donuts. (Constant)
Actual self-brand congruity
Ideal self-brand congruity
Entertainment
Actual congruity X
entertainment
Ideal congruity X entertain n ment
aDependent variable = Adoption behavior likelihood
N = 74, R = .48, R2 = .23, F(5,68) =4.06*, *p <.01


B
11.77
-2.11
-.17
-2.75
-.63

.03
21.19
-3.40
-2.07
-7.76
.76


1.53
.44
-2.11
2.57
2.90
1.26


B t
3.73
-.82 -1.04
-.07 -.09
-.56 -1.65
.68 .60

.03 .03
4.36
-1.13 -1.23
-.73 -.75
-1.55 -2.96
.72 .56


1.51 1.09
.10
-.98 -.78
1.25 1.04
.64 1.22
1.69 .87


-1.94 -2.70 -1.43 .16


Y = 11.77 2.11 (Actual self-brand congruity) .17(Ideal self-brand congruity) -
2.75(Entertainment) -.63 (Actual self-brand congruity)(Entertainment) + .03(Ideal
self-brand congruity)(Entertainment)
bDependent variable = Adoption behavior likelihood
N = 41, R = .65, R2 = .42, F(5,35) = 5.12*, *p <01
Y = 21.19 3.40(Actual self-brand congruity) 2.07(Ideal self-brand congruity) -
7.76(Entertainment) + .76 (Actual self-brand congrui[y)(Entertainment) + 1.53(Ideal
self-brand cong ruity) (Entertai nment)
CDependent variable = Adoption behavior likelihood
N = 33, R = .48, R2 = .23, F(5,27) = 1.61, p >.05
Y = .44 -2.11(Actual self-brand congruity) +2.57(Ideal self-brand congruity) +
2.90(Entertainment) +1.26(Actual self-brand congruity)(Entertainment) 1.94(Ideal
self-brand congruity)(Entertainment)


Sig.
.00*
.30
.93
.10
.55

.98
.00*
.23
.46
.01"
.58


.28
.92
.44
.31
.23
.39


Table 4-11. Multiple









Table 4-12. Multiple regression for information motivation variable (H4)
B p t Sig.
Combined sample (Constant) 17.79 4.92 .00*
Actual self-brand congruity -3.69 -1.43 -1.52 .13
Ideal self-brand congruity -.44 -.18 -.19 85
Information -2.92 -.88 -3.12 .00"
Actual congruity X information .77 1.35 1.12 .27
Ideal congruity X information .01 .02 .02 .99
Starbucksb (Constant) 21.11 4.55 .00"
Actual self-brand congruity -7.96 -2.64 -2.87 .01"
Ideal self-brand congruity 3.36 1.18 1.26 .22
Information -3.55 -1.09 -3.04 .00"
Actual congruity X information 1.69 2.59 2.18 .04
Ideal congruity X information -.86 -1.41 -1.14 .26
Dunkin' Donutsc (Constant) 10.91 1.90 .07
Actual self-brand congruity .95 .44 .20 .84
Ideal self-brand congruity -3.23 -1.57 -.71 .48
Information -1.49 -.43 -.95 .35
Actual congruity X information -.30 -.65 -.21 .83
Ideal congruity X information .72 1.51 .52 .61
aDependent variable = Adoption behavior likelihood
N = 74, R = .56, R2 = .31, F(5,68) =6.13*, *p <.01
Y = 1 7.79 3.69(Actual self-brand congruity) .44(Ideal self-brand congruity) -
2.92(Information) + .77(Actual self-brand congruity)(Informati on) + .01 (Ideal self-brand
congruity)(Information)
bDependent variable = Adoption behavior likelihood
N = 41, R = .72, R2 = .52, F(5,35) =7.66*, *p <.01
Y = 21.11 7.96(Actual self-brand congruity) + 3.36(Ideal self-brand congruity) -
3.55(Informati on) + 1.69(Actual sel f-brand congruity)(Information) .86(Ideal sel f-brand
congruity)(Informati on)
cDependent variable = Adoption behavior likelihood
N = 33, R = .41, R2 = .17, F(5,27)= 1.12, p >.05
Y = 10.91 + .95(Actual self-brand congruity) -3.23(Ideal self-brand congruity) -
1 .49(I nformatio n) -.30(Actual self-brand congruity)(Information) + 72(Ideal self-brand
congruity)(In formation)









regression for social interaction motivation variable (RQ1)


Combined sample








StarbuLcksb








Dunkin' Donutsc


(Constant)
Actual self-brand congruity
Ideal self-brand congruity
Social interaction
Actual congruity X social interaction
Ideal congruity X social interaction
(Constant)
Actual self-brand congruity
Ideal self-brand congruity
Social interaction
Actual congruity X social interaction
Ideal congruity X social interaction
(Constant)
Actual self-brand congruity
Ideal self-brand congruity


12


-2
-2



22
-9
4
-6
3
-1
4
6
-5


Social interaction .53 .15 .35 .73
Actual congruity X social interaction -2.57 -4.19 -2.01 .05"
Ideal congruityX social interaction 1.87 2.87 1.67 .11
aDependent variable = Adoption behavior likelihood
N = 74, R = .53, R2 = .28, F(5,68) =5.28*, *p <.01
Y = 12.17 + .22(Actual self-brand congruity) 2.12(Ideal self-brand congruity) -
2.28(Social interaction) .34(Actual self-brand congruity)(Social interaction) + .72(Ideal
self-brand congruity)(Social interaction)
bDependent variable = Adoption behavior likelihood
N = 41, R = .76, R2 = .57, F(5,35) = 9.29", *p <.01
Y = 22.96 9.21 (Actual self-brand congruity) +4.17 (Ideal se lf-brand congruity) -
6.55(Social interaction) + 3.00(Actual self-brand congruity)(Social interaction) -
1.38(Ideal self-brand congruity)(Social interaction)
CDependent variable = Adoption behavior likelihood
N = 33, R = .53, R2 = .28, F(5,27) =2.09, p >.05
Y = 4.23 + 6.84(Actual self-brand congruity) 5.67(Ideal self-brand congruity) +
.53(Social interaction) 2.57(Actual self-brand congruity)(Social interaction) + 1.87(Ideal
self-brand congruity) (Social interaction)


B B t
.17 3.76
.22 .09 .10
.12 -.86 -1.11
.28 -.56 -1.76
.34 -.46 -.40
.72 .95 .92
.96 5.24
.21 -3.05 -3.39
1.7 1.47 1.57
.55 -1.56 -3.66
.00 3.59 2.82
.38 -1.71 -1.30
.23 1.10
.84 3.19 2.11
.67 -2.76 -2.05


Siq.
.00*
-92
.27
.08
.69
.36
.00*
.00*
.13
.00*
.01"
.20
.28
.04*
.05*


Table 4-13. Multiple









Multiple regression for convenience motivation variable (RQ2)


B B t Sig.
Combined (Constant) 13.34 5.36 .00"
sample Actual self-brand congruity -.4A -.19 -.23 .82
Ideal self-brand congruity -1.94 -.78 -.96 .34
Convenience -3.26 -.77 -2.81 .0i
Actual congruity X convenience -.14 -.19 -.15 .88
Ideal congruity X convenience .82 1.10 .89 .38
Starbucksb (Constant) 19.21 5.86 .00"
Actual self-brand congruity .59 .19 .21 .83
Ideal self-brand congruity -4.74 -1-67 -1.78 .08
Convenience -5.96 -1.28 -3.97 .00"
Actual congruity X conve nience -1.03 -1.17 -.85 .40
Ideal congruity X convenience 2.49 2.88 2.04 .05*
Dunkin' Donuts (Constant) 3.12 .75 .46
Actual self-brand congruity -1.32 -.61 -.47 .64
Ideal self-brand congruity 1.16 .56 .42 .68
Convenience 1.15 .33 .61 .55
Actual congruity X convenience .64 1.07 .51 .62
Ideal congruity X convenience -.93 -1.56 -.73 .47
aDependent variable = Adoption behavior likelihood
N = 74, R = .54, R2 = .29, F(5,68) =5.45*, *p <.01
Y = 13.34 48(Actual self-brand congruity) 1-94(Ideal self-brand congruity) -
3.26(Convenience) .14(Actual self-brand congruity)(Convenience) + .82(Ideal
self-brand congruity)(Convere nce)
bDependent variable = Adoption behavior likelihood
N = 41, R = .72, R2 = .51, F(5,35) =7.35*, *p <.01
Y = 19.21 + .59(Actual self-brand congruity) 4-74(Ideal self-brand congruity) -
5.96(Conve nience) 1.03(Actual self-brand congruity)(Convenience) + 2.49(Ideal
self-brand congruity)(Convene nce)
Dependent variable = Adoption behavior likelihood
N = 33, R = .41, R2 = .16, F(5,27) = 1.06, p >.05
Y = 3.12 1.32(Actual self-brand congruity) + 1 16(Ideal self-brand congruity) +
1.15(Convenience) + .64(Actual self-brand Congruity)(Convenience) .93(Ideal
self-brand congruity)(Convenience)


Table 4-14.









Table 4-15. Summary of multiple regression significant relationships
Self-brand congruity Individual Interaction effect
Actual Ideal motivation Actual Ideal
Motivations C.S. S D.D. C.S. S D.D. C.S. S D.D. C.S. S D.D. C.S. S D.D.
Entertainment X
Social interaction X X X X X X
Information X X X X
Convenience XX X









CHAPTER 5
DISCUSSION AND CONCLUSIONS

Discussion

The main purpose of this research was to determine if an opportunity occurs for

greater customer relationship management and to clarify a few of the fundamental

principles that can lead to successful relationships with consumers via SNSs. To do so,

this study examined the relationship of actual and ideal self-brand congruity, as well as

brand-SNS compatibility with consumer likelihood of publicly associating themselves

with a brand on Facebook. In conjunction with these variables, motivational interactions

with each self-brand congruity measure were examined to attempt to further explain

what impacts SNS posting behavior. The two brands utilized to exemplify these

potential relationships were Starbucks and Dunkin' Donuts within the SNS context of

Facebook.

Two theories were applied in shaping this research. In previous research,

self-congruity theory has been used to explore brand attitude. Since brand attitude is

often used in predicting branded behavior, the linkages between these three constructs

led to an investigation in the present study of self-brand congruity's relationships with

branded behavior within the SNS environment. This investigation was fulfilled by

observing both actual and ideal self-brand congruity perceived by participants and then

determining the presence or absence of relationships with brand adoption behavior

likelihood on Facebook. In the process, brand attitude was also measured as a source

of extra security. This examination presented results in alignment with previous

research findings, demonstrating that with an increase in brand attitude, both actual and

ideal self-brand congruity also increase.









Second, the uses and gratifications theory inspired an inspection of the interaction

effects of SNS usage motivations on posting behavior. Interaction effects were studied

for four motivation categories: entertainment, information, social interaction, and

convenience. These components were employed to observe each motivation's

interaction with each self-brand congruity element in predicting adoption behavior

likelihood.

These theoretical foundations provided structure for advancing previously

explained relationships in a complex and relatively young social media environment.

The following theoretical implications detail the conclusions derived from the statistical

tests completed in Chapter 4. Collectively, they help clarify which factors contributed to

brand adoption behavior likelihood, as well as give reasons for elimination of those that

do not contribute to this behavior, as originally expected. A discussion of the

implications of these findings for marketers and advertisers in the future follows to

provide guidance for the next steps to be taken regarding this topic.

Implications

This research was fueled by an inquiry into which factors contributed to SNS

posting behavior, which indicates personal adoption of a brand. The primary constructs

observed in the present research were self-brand congruity, brand-Facebook

compatibility, and motivations for using SNS. Both self-brand congruity (actual and

ideal) and brand-Facebook compatibility were examined in Hypothesis 1 and

Hypothesis 2, respectively, using bivariate correlation analyses with regard to adoption

behavior likelihood. The four motivational interaction effects were each assigned their

own examination. Hypotheses 3 and 4 targeted the entertainment and information

motivations, respectively. Research Questions 1 and 2 addressed the predicted









negative interaction effects of the social interaction and convenience motivations,

respectively. To achieve the purpose of this research, a combination of frequency

analyses, descriptive summaries, t-tests, bivariate correlations, and multiple regression

analyses were implemented. Statistical compilations of these tests illustrate existent and

non-existent linear relationships between the variables.

A precautionary assessment of each sample group was performed to make sure

the brand was not overlooked as an extraneous or spurious variable. Variables were

evaluated for significant differences between the Starbucks sample and the Dunkin'

Donuts sample. These tests showed no significant differences for any of the

demographic variables or variables addressing Facebook familiarity, attitudes, and

activity. In addition, familiarity with Starbucks was not significantly different from

familiarity with Dunkin' Donuts, which both averaged between "familiar" and "very

familiar" in responses. Thus, both Starbucks and Dunkin' Donuts participants mirrored

each other in this sample demographically, but also in that both groups can be assumed

to be equally familiar with both Facebook and their respective brand. This established

familiarity improved the reliability of the subsequent data. Moreover, the lack of

significant differences for these variables indicated that Facebook activity and attitudes

were not contingent on varying brands. Current Facebook behavior indicated that

participants demonstrated moderate to high levels of expression behavior overall.

However, attitudes toward Starbucks were significantly more positive than those

toward Dunkin' Donuts. Previous research establishing self-brand congruity as a

predictor of brand attitudes was further validated by this study's high correlations

between these constructs. Both actual and ideal self-brand congruity were significantly









greater for Starbucks than for Dunkin' Donuts. In this case, the results illustrate that

Starbucks participants perceived more congruity between both their actual and ideal

selves and typical Starbucks users than Dunkin' Donuts participants perceived between

both their actual and ideal selves and the typical Dunkin' Donuts user. Within this

Starbucks significantly greater sample, participants perceived significantly higher

congruency between their actual selves and the typical user of Starbucks than between

the typical user of Starbucks and how they would like to see themselves (ideal

congruity). However, no similar significant differences occurred between actual and

ideal self-brand congruity for the Dunkin' Donuts sample.

Significant differences between groups signified, as with attitudes toward different

brands, that individual brands have unique levels of self-brand congruity, which applies

to both actual and ideal congruity. This finding suggests that self-brand congruity

measures are more reliant on brand attitudes, and perhaps brand personality, than

originally considered. In turn, brands that manage to achieve a high level of actual and

ideal self-brand congruity with their current or target consumers, and also achieve highly

positive brand attitudes among these consumers, could observe increased adoptive

behavior and SNS word-of-mouth benefits over brands that fail to achieve these goals.

This implication is supported further by the results of Hypothesis 1 in the following

section. The implication that follows is a need to investigate self-brand congruity

components on a per brand basis in future research. Also, based on the findings of this

research, marketers should attempt to relate their typical user image with the actual and

ideal self-images of their current and target consumers for increased word-of-mouth

benefits from SNS activity.









Similar to the outcomes of self-brand congruity, compatibility also varies between

brands depicted by significantly higher perceived Starbucks compatibility with Facebook

than perceived Dunkin' Donuts compatibility with Facebook. This difference could

ultimately mean that some brands are considered more appropriate or acceptable to

talk about and display openly on an SNS than others, and that compatibility could be a

significant factor in the execution of branded Facebook behaviors.

Motivational interactions were also checked for significant differences between

brands. Among the four motivation variables, entertainment, social interaction, and

conveniences yielded no significant differences between brands. Thus, participants

were assumed to be equally likely or unlikely to identify these SNS usage motivations

for themselves regardless of the brands presented to them. Information was the only

motivation to exhibit a significantly higher information motivation for Dunkin' Donuts than

for Starbucks. Thus, some brands elicit greater information motivation for their SNS

usage than other brands. This could mean for future research purposes that those

pursuing informational purposes are more likely to actively search for brands and carry

out associated branded behavior.

Lastly, the dependent variable also showed that adoptive behavior of Starbucks

was significantly higher than adoption of Dunkin' Donuts into SNS behavior. Correlation

measurements showed significant strong correlations between the components used to

compute the variable. Significant moderate and strong correlations were also revealed

between the computed adoption behavior likelihood and brand attitudes. These

correlation tests substantiated the dependent variable components as good

measurements, although the relationship between brand attitude and adoption behavior









likelihood was significantly stronger for Starbucks than Dunkin' Donuts. A key

observation from frequency statistics of each individual proposed action was that

participants were least likely to fulfill actions which required them to exert increased or

initiated effort in completing the behavior.

One inconsistency to note, however, was the correlations between current

Facebook behavior reported and the indicated intent of adoption behavior. As stated

previously, current Facebook behavior indicated moderate to high activity. Conversely,

adoption behavior likelihood results were comparatively much lower on their respective

scale. Although each brand individually showed no significant relationship, for the

combined sample (disregarding brand), a significant negative and weak relationship

occurred between these variables. This relationship indicated that as current Facebook

behavior increases, brand adoption behavior likelihood decreases. The result may

suggest that indicated likely behavior does not strongly translate into actual fulfilled

behavior.

The significant differences between groups for all of these studied variables are

enough to justify separate regression examinations by brand, along with the combined

sample, for each of the hypotheses and research questions. The following sections will

examine these cases in more detail.

S Hypothesis 1: Self-brand congruity (both actual and ideal) will be positively related
to the likelihood of SNS brand adoptive behavior.

The goal of Hypothesis 1 was to determine if a positive relationship existed

between both actual and ideal self-brand congruity and adoption behavior likelihood.

This prediction was supported in the bivariate regression test with significant moderate

positive relationships between both actual and ideal self-brand congruity and adoption









behavior likelihood for both Starbucks and the combined brand sample. Dunkin' Donuts

showed only a significant positive moderate relationship for ideal congruity and adoption

behavior likelihood. Results can therefore vary, depending on brand, but overall, this

hypothesis was accepted. Higher perceived similarity between how a Facebook user

actually views himself and how he perceives the typical user of a brand indicates an

increased likelihood that the user will positively adopt the brand to his profile behavior in

some fashion. This positive relationship also applies with perceived similarity increases

between how a Facebook user would like to view himself and how he perceives the

typical user of a brand.

These results are in alignment with previous research performed by Sirgy (1997)

and Parker (2005) comparing ideal and actual self-perceptions to brand images and

personalities. The results also imply that attitudes are a key factor in the behavior of

sharing brand associations on SNSs. The correlation results for this hypothesis analysis

signify that adoption behavior likelihood is complementary with both actual and ideal

self-brand congruity cases, but not strong enough to denote complete independence in

predicting adoption behavior likelihood. The reasons for branded posting behavior thus

extend beyond simple self-brand congruity explanations, and suggest that other factors

not studied in this research are involved in the execution of such behavior. However,

with actual and ideal self-brand congruity acting as two of the variables significantly

related to adoption behavior, brands would benefit via word-of-mouth activity by aligning

themselves with the self-image perceptions of their target consumer market.

S Hypothesis 2: Brand-SNS compatibility will be positively related to the likelihood
of SNS brand adoption behavior.









Hypothesis 2 attempted to account for the possibility that some brands are not

considered relevant or appropriate to share on SNSs. The prediction of a positive effect

of brand-Facebook compatibility on adoption behavior likelihood was rejected. Contrary

to self-brand congruity, brand-Facebook compatibility did not have a significant

relationship with adoption behavior likelihood in this bivariate correlation model for either

brand individually or the combined sample. Although significant differences occurred of

perceived compatibility between the brands, bivariate correlation results suggest that

regardless of pre-conceived image perceptions, compatibility between a brand and

Facebook have no influence on decisions to carry out brand adoptive posting behaviors

on SNSs. Speculation suggests that SNS users show a lack of concern for the

appropriateness of their posting behavior-an area where additional research could

provide more insight. An added opportunity presents itself for further research linked

with the self-brand congruity implications of determining whether specific characteristics

or personalities of brands have more impact on these variables than the brands do as a

unit. Also, this result could have been a product of observing only one brand category.

Perhaps looking at results across brand categories would be more indicative of specific

effects on behavior between brands, as denoted in the previous limitations section.

The remainder of the statistical data analyses intended to identify interaction

effects of four motivations for using an SNS. Hypotheses 3 and 4 target the predicted

positive interaction effects of entertainment and information motivations respectively.

S Hypothesis 3: An entertainment SNS motivation will positively interact with both
actual and ideal self-brand congruity in predicting SNS adoption behavior
likelihood.

The multiple regression analysis for Hypothesis 3 presented the individual

entertainment motivation variable for Starbucks as the only variable significantly and









positively related to adoption behavior likelihood out of all three sample groups. The

more a person is motivated by entertainment in their SNS usage, the more likely they

are to adopt Starbucks into his SNS posting behavior. The Starbucks regression model

showed that there was a significant moderate relationship between the independent

variables, and the dependent variable is projectable onto the population. Entertainment

motivation individually is a complementary influence on adoption behavior likelihood, but

not strong enough to act independently. This inability to act independently could mean

that factors other than SNS usage motivations are concerned. It could also indicate that

the entertainment motivation touches on a more complex motivation or set of

motivations that could have stronger significant correlations with adoption behavior

likelihood. The possible involvement of other motivations could extend differently to

varying brands. The differences in significant relationships between sample groups

indicate that some brands experience higher adoptive behavior than others when a

person is motivated to use an SNS for entertainment purposes. This divergence could

be due to differences in brand personalities and entertainment qualities attributed to a

brand.

Furthermore, neither actual self-brand congruity nor ideal self-brand congruity

were significantly related to adoption behavior likelihood. Significant relationships were

also absent among the interaction effects between both these self-brand congruity

variables and entertainment motivation. Thus, neither a positive nor a negative

interaction occurred with self-brand congruity variables in predicting adoption behavior

likelihood by entertainment motivations. It is therefore reasonable to conclude that an

entertainment motivation has no impact via interaction with self-brand congruity on the









likelihood of adopting the brand in SNS posting behavior. This hypothesis was rejected

as a result. The analysis results indicated that it made no difference if a person felt any

kind of congruity with a brand, for a person to adopt Starbucks into his SNS posting

behavior for entertainment purposes.

S Hypothesis 4: An information SNS motivation will positively interact with both
actual and ideal self-brand congruity in predicting SNS adoption behavior
likelihood.

Hypothesis 4 performed an identical analysis as Hypothesis 3, but for the

information motivation. The multiple regression tests for Hypothesis 4 returned some

results that confirmed expectations, and some results that contradicted expectations.

First, the regression model for Dunkin' Donuts was not significant. Significant

relationships were not present for any of the individual variables inputted for this model

either. However, both the statistically significant Starbucks and combined sample

models disclosed the individual information motivation variable to have a significant

positive relationship with adoption behavior likelihood. Thus, the more users are

motivated to use an SNS for information purposes, the more likely they are to adopt a

brand into their SNS posting behavior.

Furthermore, in addition to the individual information motivation variable,

Starbucks also revealed actual self-brand congruity and its interaction with information

motivation as significantly related to adoption behavior likelihood. The effect of actual

congruity individually was positive, meaning that as perceived congruity between a

person's actual self and the typical user of Starbucks increased, so did his likelihood of

adopting Starbucks into his SNS posting behavior. However, the interaction effect of

actual self-brand congruity and information motivation was negative, contrary to

expectations, indicating that if a person using an SNS for information purposes also









perceives high congruity between his actual selves and the typical Starbucks user, he is

less likely to adopt Starbucks into his SNS posting behavior.

The data detailed that regardless of any other factors, those respondents who are

motivated to use an SNS for information motivations are more likely to adopt a brand

into their posting behavior. However, when information motivation interacts with actual

self-brand congruity, an SNS user becomes less likely to adopt a brand into his posting

behavior. Many explanations could be offered for this directional discrepancy between

the individual and interactional effects of information motivation. For example, Starbucks

is a widely known brand. Perhaps, when amplified by a sense of congruity with

themselves, whom they already know rather well, users do not feel as strong a need to

collect information regarding a brand they feel they already know fairly well, as opposed

to a brand they are not as familiar with. This is substantiated by the significantly higher

information motivation observed for Dunkin' Donuts than for Starbucks.

Moreover, the positive influence of information motivation individually could be

attributed to a rise in more interactive forum information-seeking, in which at least a

minimal level of information is often required of the information seeker to retrieve the

appropriate information. Subsequently, this informational response is likely from other

peers within an SNS user's network. As a result, it could be likely that this type of

information seeking on SNSs plays more of a conversational role than traditional

unidirectional information retrieval on the Internet.

Furthermore, although the actual congruity and its interaction effect with

information motivation were significant for Starbucks, the remaining interaction effects

for each sample were not significantly related to adoption behavior likelihood. Therefore,









an information motivation to use an SNS does not significantly interact, positively or

negatively, with the self-brand congruity perceived by that user in relation to branded

adoptive behavior likelihood. Motivations outside of those examined in this study could

be more suited as interaction variables, or other factors could be key influencers in this

model, such as brand familiarity. The differences per brand for this examination indicate

that these measures are significant factors unique for individual brands, meaning that

outside factors could explain the reasons for such different responses to these

variables.

Next, Research Questions 1 and 2 approached the same type of interaction

effects as in Hypothesis 3 and Hypothesis 4 for the final two SNS usage motivations,

social interaction, and convenience. These motivations were posed as research

questions because they were predicted to have negative interaction effects with

self-brand congruity in predicting adoption behavior likelihood.

* Research Question 1: Do social interaction SNS motivations negatively interact
with actual and ideal self-brand congruity in predicting SNS brand adoption
behavior likelihood?

The regression model equation for the combined sample was significant, showing

a moderate positive correlation between the independent variables and adoption

behavior likelihood. However, this combined sample did not return any of the inputted

variables to have significant relationships with adoption behavior likelihood. However,

many significant individual variables existed for each of the brands individually.

Starbucks revealed three significant relationships with adoption behavior

likelihood: 1) the independent social interaction motivation which had a positive effect,

2) actual self-brand congruity which also had a positive effect, and, 3) as expected, a

negative interaction effect between these two variables. These results mean that









individually, if a user perceives high congruity between his actual self and the typical

user of Starbucks, he is more likely to positively adopt Starbucks into his SNS posting

behavior. Likewise, if a user is motivated in his SNS usage by social interaction

purposes individually, he is more likely to positively adopt Starbucks into his SNS

posting behavior. However, if a user is both motivated to use an SNS for social

interaction purposes and feels high similarity between how he actually views himself

and the typical user of Starbucks, then he is less likely to adopt Starbucks into his SNS

posting behavior.

There could be a couple reasons for the tendency presented by these results. One

explanation could be that when a user is highly motivated by social interaction

purposes, whether or not he feels he is similar with the typical user of a brand plays a

small role compared to when social interaction is not a high motivational priority.

Another rationale is that perhaps he is more concerned about social implications of

expressing his actual congruity with the brand. This study was performed during an

economic recession where consumers showed a sense of pride in finding great deals as

opposed to over-spending on trendy items. This kind of social economic consciousness

could have played an influential role in the perceptions of brands as well as the

willingness to associate with those brands.

Dunkin' Donuts, on the other hand, did not return the individual social interaction

motivation variable as significantly related to adoption behavior likelihood. It did return

both actual self-brand congruity and the interaction of actual self-brand congruity and

social interaction motivation to have negative and positive significant relationships,

respectively, with the dependent variable though. Dunkin' Donuts also found ideal









self-brand congruity to be significantly related positively with adoption behavior

likelihood. These results found that the more a person perceives similarity between their

actual selves and the typical user of Dunkin' Donuts, the less likely he is to adopt

Dunkin' Donuts into their SNS posting behavior. This could suggest a sense of

embarrassment regarding the perceived similarity, which indicates brand attitude and

perhaps brand personality play important roles in this relationship. In concert with this

conclusion, the more they perceive congruity between how they would like to view

themselves and the typical user of Dunkin' Donuts, the more likely they are to adopt

Dunkin' Donuts into their SNS posting behavior, which possibly indicates a sense of

pride, self-confidence, or other qualities that would be natural products of positive brand

attitudes. Lastly, if a person felt strong congruity between how he views his actual self

and how he views the typical user of Dunkin' Donuts, and is also motivated in his SNS

usage by social interaction purposes, he is more likely to adopt Dunkin' Donuts into his

SNS posting behavior. This implies that if a person is highly motivated by social

interaction, perhaps he is less concerned with the social implications of expressing his

self-brand congruity in some fashion.

The remainder of the variables, including the interaction effects of social

interaction with ideal self-brand congruity for each sample, attested to a lack of

significant relationships with adoption behavior likelihood. Therefore, social interaction

motivation failed to confirm a negative interaction with both actual and ideal self-brand

congruity. Result differences between brands indicate that some brands are more

influenced by social interaction and self-brand congruity than others, and each brand









must be considered individually on qualities such as brand attitude and brand

personality.

* Research Question 2: Do convenience SNS motivations negatively interact with
actual and ideal self-brand congruity in predicting SNS brand adoption behavior
likelihood?

As with the previous motivation examinations, the multiple regression models for

Starbucks and the combined sample executed for Research Question 2 exposed linear

relationships between the independent variables and adoption behavior likelihood and,

can therefore be projected onto the population. Once again, the model for Dunkin'

Donuts was not significant, so it cannot be projected onto the population. Within the

Starbucks and combined samples, the independent convenience motivation variable

proved to have a significant positive relationship with adoption behavior likelihood.

Accordingly, the more a person is motivated by convenience in his SNS usage, the

more likely he is to adopt a brand into his SNS posting activity. This shift from an

expected negative effect to a positive one could be explained by ease of use associated

with individual SNSs because this factor would likely increase or decrease usage

outcomes of convenience motivations. Another possible explanation could be linked to a

person's SNS usage intensity. For example, if a person is already spending time on an

SNS, the convenience of already being logged in with a network and media capabilities

at his fingertips could positively influence the likelihood of adoption behavior.

In addition to the significant positive effect of convenience motivation, a significant

negative interaction effect on adoption behavior likelihood was observed by

convenience motivation and ideal self-brand congruity for the Starbucks sample. This

says that the more a person perceives similarities between how he would like to see

himself and the typical user of Starbucks and the more this person is motivated to use









an SNS for convenience purposes, the less likely he is to adopt Starbucks into his SNS

posting activity. One explanation for this could be that the person who views high

congruity between his ideal self and the typical user of Starbucks, is willing to extend

beyond simple convenience activity and is more likely to have other motivations than

convenience. This explanation would lead to a negative interaction with ideal self-brand

congruity's indication of adoption behavior likelihood.

Unlike the interaction between ideal convenience motivation and ideal self-brand

congruity for the Starbucks sample, the remainder interaction effects were not

significant. Therefore, while the convenience motivation may individually be significantly

related to adoption behavior likelihood, convenience motivation does not significantly

interact, positively or negatively, with actual and ideal self-brand congruity perceived by

that user in predicting branded adoptive behavior likelihood. The lack of significant

interaction effects hint that motivations outside of those examined in this study were

more suited as interaction variables. Differences between results for Dunkin' Donuts

and Starbucks indicate that individual brands are influenced by self-brand congruity and

convenience motivations uniquely and should be observed on a per brand basis for

future studies.

Implications Summary

Thus within the scope of this study, both actual and ideal self-brand congruity

proved to have significant positive relationships with adoption behavior likelihood

independently, while brand-Facebook compatibility did not. Together, these results

along with differences in findings per brand implicate a need for a more intense

examination of both brand attitude and unique brand characteristics or personalities and

their relationships with the variables in this study. Furthermore, actual self-brand









congruity had more impact on adoption behavior likelihood both individually and in

interaction with motivations.

None of the motivations significantly interacted with both the actual and ideal

aspects of a person's self-brand congruity in predicting the likelihood of whether or not a

user adopts a brand into his SNS activity. However, all four motivations showed promise

as individual variables to have a significant relationship with adoption behavior

likelihood on a per brand basis. Despite the fact that the four motivations used were

acceptable determinants of Internet usage and SNS usage as indicated by this study,

the results shown here also imply that these SNSs motivation categories leave room for

improvement in predicting branded SNS posting behavior. Improvements may involve a

slightly modified combination of motivations that are not quite as transferrable from

usage motivations of the Internet as a whole or new motivations that have evolved from

SNS usage. However, independent of interactions with self-brand congruity, each

motivation explored in this study demonstrated at least a moderate significant

relationship with branded adoption behavior likelihood in many cases, as observed in

this study.

In alignment with the increasing popularity of SNSs and interactivity between

people and media, the implications of this excavation can serve marketers as

inspirational tools to help navigate this realm in new ways. Marketers can use this

information to improve upon relationships with their target market and hopefully to

ultimately improve upon purchase behavior by more seamlessly integrating SNS

campaigns into a marketing mix. The results of this study substantiate previous notions

about the importance of knowing the target consumer on a personal level.









The significant relationship between self-brand congruity and branded SNS

behavior indicates an importance for brands when forming their SNS presence to align

themselves closely with their consumers' self-perceived qualities, both actual and ideal.

To make this alignment, first, a marketer needs to clearly define his target market. Part

of this process entails a marketer getting to know who his target market is, not just

demographically, but also on a personality and psychological level. Consumers'

entertainment, information, social interaction, and convenience tendencies should be

included in this acquaintanceship, since these motivations showed significant influences

on SNS adoption behavior unique per brand. Second, a marketer should align a brand's

perceived typical user image with the actual and ideal perceptions and assessments of

consumers in the group he just described as his target market.

According to the results of this study, a marketer positioning a brand this way will

increase the frequency of which his brand is shared in a way that is very emotionally

attached and which has the potential to efficiently reach a very large audience. As a

result, increasing this frequency could help develop consumers of a brand simply

through exposure. Furthermore, a key element in most marketing plans is

word-of-mouth communication. In lieu of this, the mention of a brand on a user's profile

can be especially valuable because that person is essentially attaching a trusted

endorsement not just in conversation on a singular level, but now on a platform that

reaches a more personal plural level.

Future Research

The insight obtained from this research could be just a small sample of a much

larger pool of information to be attained. Future research could include a variety of

outlooks for many questions are still left unexplained. First, this study focused only on









positive adoption behavior likelihood. It did not observe self-brand congruity in relation

to behavior of positive versus negative attitude or intent, which would be a natural next

step. The results of this study lend themselves to the likelihood that higher perceived

self-brand congruity results in more positive toned posting behavior, while lower

perceived self-brand congruity results in more negative toned posting behavior.

However, results opposite to this conjecture would pose a question regarding a standoff

between content attitude and simple exposure value of posting behavior as the more

beneficial component for a brand.

As the limitations section pointed out, privacy restrictions acted as a noteworthy

limitation. A Facebook user could view content only if he is friends with the individual

posting the content. This restriction limits the reach of such content. With access to

unrestricted profiles, accurate content analysis could be observed as subsequent

research. This kind of research would be uninhibited by the subjectivity of a participant

in his responses. Furthermore, privacy is a key issue to many who use SNSs and thus

may impact posting behavior that creates personal associations publicly. It could be

worth exploring the impact of such self-inflicted privacy restrictions, attitudes, and

preferences on SNS posting behavior in future research.

Also, this study could be expanded to observe relationships between the variables

in this study with purchase intent and purchase behavior. Do those who had a high

degree of actual and ideal self-brand congruity have an increased likelihood of purchase

intent? Do these people also have a higher tendency toward actual purchase behavior?

Is there a significant difference between the effect of actual congruity and ideal

congruity on purchase intent and purchase behavior? Do people with high self-brand


100









congruity indications or high adoption behavior on an SNS consume the brand with

higher frequency than those who do not post in reference to the brand or those who do

not feel a sense of self-brand congruity, actual or ideal? Does consumer loyalty depend

on whether a person posts regarding a brand first or become a consumer of the brand

first? These are just a few questions that could be explained further in subsequent

studies.

In addition, future studies could observe the interaction effects of consumers

versus non-consumers of a brand in question regarding predictive relationships

between self-brand congruity measures and adoption behavior likelihood. The current

study primarily considers consumers of a product category. However, non-consumers

are likely to know of brands they do not consume and still have brand attitudes about

them.

Finally, since only a select few of the motivation variables explored showed a

significant effect on adoption behavior likelihood, future research could attempt to clarify

more appropriate motivations for using SNSs and specifically for posting content on

SNSs. A subsequent influencing variable could be perceived genuineness. For

example, in terms of the word-of-mouth communication portrayed in branded posting

behavior, the effectiveness of this more personal form of communication could be

affected by the number of friends the poster is perceived to be delivering the message

to or the frequency with which such branded posts are made.

The research performed for this study successfully answered the questions it

posed by further clarifying factors that influenced (or did not influence) branded SNS

posting behavior. The suggested research topics would continue to provide a greater


101









knowledge of posting behavior and its value equation for marketers by taking the

implications revealed by this study and applying it to further examinations of this

subject.


102









APPENDIX A
INVITATION

This is an invitation to participate in a research study. The goal of the study is to learn
more about consumer's activity on social networking sites. The information acquired
will be important in expanding upon current research in advertising. Your participation
in this study is greatly appreciated. It should take no more than 15 minutes and should
require minimal effort, just reliable responses. Participant responses will only be used
for research and will remain completely confidential and anonymous. Please pass this
invitation to participate on to friends and family to complete as well. Participation is
greatly appreciated.

If your first name starts with letters A-L please click link #1:
http://www.surveymonkey.com/s.aspx?sm=GBfx61x_2f6NVMiuKrayum7A_3d_3d
If you first name starts with letters M-Z please click link #2:
http://www.surveymonkey.com/s.aspx?sm=mBVIKb57kB6ZRTpfDzPxJA_3d_3d

(If you manually enter the link into your navigation bar, please type the URL into your
browser exactly as it appears above, including the specific uppercase, lowercase,
symbols, and underscores as it is shown above or else it will not work)

Please complete the survey by May 26, 2009. Surveys must be completed in their
entirety in order to be valid data for this study. If you have any questions regarding the
study or have trouble accessing the website links, please feel free to contact me at any
time at riediger.2@ufl.edu. Thank you for your participation!


103









APPENDIX B
ONLINE SURVEY QUESTIONNAIRE

Welcome! One of the emerging methods of communication in today's society involves social
networking sites on the Internet. As a whole, these sites are an unchartered medium through
which consumers and brands may connect, yet very little research has been conducted on the
effects of such methods of consumer relationship development.

This survey includes a wide range of questions about you and your general feelings and activity
toward brands in the context of social networking sites. The questionnaire should take
approximately 15-20 minutes to complete.

Rest assured, you will not be asked to identify yourself individually within the survey and any
information you provide will remain strictly confidential.

There are no direct benefits, risks, or compensation to you for participating in the study. You
may discontinue or refuse to take part at any time and your responses will not be processed
unless you submit the survey upon completion.

By clicking the "Submit" button, you are indicating your voluntary consent to participate in this
research.

Please carefully read the instructions at the beginning of each section. Most of the questions
can be answered by clicking on the buttons) that best expresses your response.

Questions about the study should be directed to riediger.2@ufl.edu. Thank you very much for
helping with this important survey.

Submit

1. Age?

2. Do you drink coffee?
Yes No

a. If Yes, what kind of coffee do you drink most often?
(1) a purchased cup of regular coffee (caffeinated or decaffeinated roasted
coffee with or without some combination of cream and sugar)
(2) a purchased cup of specialty coffee (includes all of the designer coffee
options such as a Latte, Frappuccino, Espresso, etcetera)
(3) home brewed coffee

b. If Home Brewed Coffee was chosen for the previous question, what brand of
home brew do you drink most often?

(1) Starbucks
(2) Dunkin' Donuts
(3) Folgers
(4) Maxwell House
(5) Other (Please Specify)


104









3. Do you have an account with Facebook? If Yes, please indicate the date you joined the
community in the space provided (or estimate as accurately as possible).


(1) No
(2) Yes


Date:


Please answer the following questions as truthfully and accurately as you can. Clearly circle the
number that you think BEST describes you. The number '3' should indicate you neither agree
nor disagree.

"I use Facebook (because)... [coded Strongly Agree=1, Strongly Disagree=5]


4. To learn about unknown things
Strongly Disagree 5 4 3

5. It's a good way to do research
Strongly Disagree 5 4 3

6. To learn about useful things
Strongly Disagree 5 4 3

7. It's convenient to use
Strongly Disagree 5 4 3

8. I can get what I want for less effort
Strongly Disagree 5 4 3


2 1 Strongly Agree


2 1 Strongly Agree


2 1 Strongly Agree


2 1 Strongly Agree


2 1 Strongly Agree


9. I can use it anytime, anywhere
Strongly Disagree 5 4

10. To pass time
Strongly Disagree 5 4

11. I just like to surf the Internet
Strongly Disagree 5 4

12. It's enjoyable
Strongly Disagree 5 4

13. It's entertaining
Strongly Disagree 5 4

14. I wonder what other people saic
Strongly Disagree 5 4

15. To express myself freely
Strongly Disagree 5 4

16. To meet people with my interest
Strongly Disagree 5 4


3 2 1


Strongly Agree


3 2 1 Strongly Agree


3 2 1


3 2 1


Strongly Agree


Strongly Agree


3 2 1 Strongly Agree

j
3 2 1 Strongly Agree


3 2 1 Strongly Agree

ts
3 2 1 Strongly Agree


105









17. How familiar are you with Facebook?
Very Unfamiliar 5 4 3 2 1 Very Familiar

18. How acceptable is Facebook to you?
Very Unacceptable 5 4 3 2 1 Very Acceptable

19. How likely are you to visit Facebook the next time you use the Internet?
Very Unlikely 5 4 3 2 1 Very Likely

20. How often do you frequent Facebook?

(1) Multiple times per day
(2) Once per day
(3) Multiple times per week
(4) Once per week
(5) Multiple times per month
(6) Once per month
(7) Less than once per month

21. About how many hours and minutes would you say you actively spend on Facebook in a
typical week?

Facebook Activity Time: hrs min

22. Please rank the following Facebook activities from 1 to 3. (1 meaning the activity you spend
the most time on in your Facebook session, 3 being what you spend the least time doing.)

Adding content to your profile or other users' profiles
Browsing
Responding to/interacting with others

Please indicate your level of agreement or disagreement with the following:

23. Facebook builds a relationship with me
Strongly Disagree 5 4 3 2 1 Strongly Agree

24. I would like to visit Facebook again
Strongly Disagree 5 4 3 2 1 Strongly Agree

25. I am satisfied with Facebook's services
Strongly Disagree 5 4 3 2 1 Strongly Agree

26. I feel comfortable in surfing Facebook
Strongly Disagree 5 4 3 2 1 Strongly Agree

27. Facebook is a good place to spend my time
Strongly Disagree 5 4 3 2 1 Strongly Agree

28. I would rate Facebook as one of the best
Strongly Disagree 5 4 3 2 1 Strongly Agree


106









Your Facebook profile already displays:
29. Basic Information
(1)Full (2)Partial (3)Limited (4)None (5)1 am not familiar with this

30. Personal Information
(1)Full (2)Partial (3)Limited (4)None (5)1 am not familiar with this

31. Contact Information
(1)Full (2)Partial (3)Limited (4)None (5)1 am not familiar with this

32. Education and Work Information
(1)Full (2)Partial (3)Limited (4)None (5)1 am not familiar with this

33. Groups you've become a member of
(1)Many (2)A Few (3)None (4)1 am not familiar with this

34. Items you've become "a fan" of
(1)Many (2)A Few (3)None (4)1 am not familiar with this

35. Bumper Stickers from other Facebook members
(1)Many (2)A Few (3)None (4)1 am not familiar with this

36. Bumper Stickers added by yourself
(1)Many (2)A Few (3)None (4)1 am not familiar with this

37. Gifts from other Facebook members
(1)Many (2)A Few (3)None (4)1 am not familiar with this

38. Other Applications
(1)Many (2)A Few (3)None (4)1 am not familiar with this

39. How familiar are you with [brand]?
Very Familiar 1 2 3 4 5 Very Unfamiliar

[Brand] is...

40. Extremely Good 1 2 3 4 5 Extremely Bad

41. Extremely Pleasant 1 2 3 4 5 Extremely Unpleasant

42. Extremely Favorable 1 2 3 4 5 Extremely Unfavorable

43. Extremely Likable 1 2 3 4 5 Extremely Unlikable


107










[Brand]


Take a moment to think about [Brand]. Think about the kind of person who typically uses
the brand. Imagine this person in your mind and then describe this person using one or
two personal adjectives to describe the typical user of the brand.

44. Adjective #1
45. Adjective #2

Now indicate your agreement or disagreement to the following statements:

46. The typical user of Starbucks is consistent with how I see myself.
Strongly Agree 1 2 3 4 5 Strongly Disagree

47. The typical user of Starbucks is consistent with how I like to see myself.
Strongly Agree 1 2 3 4 5 Strongly Disagree

48. The Starbucks brand is compatible with Facebook.
Strongly Agree 1 2 3 4 5 Strongly Disagree


49. Would you add [Brand] to your page in any way, either visually or written verbally?
No, Definitely Not 1 2 3 4 5 Yes, Definitely

50. Consider the brand Starbucks on Facebook. Now, look through the following list and select
all that apply. Consider each option on individual terms, exclusive from the others.

I would:

(-1) write/post/display something negative about [Brand] in my profile or status
(+1) write/post/display something positive about [Brand] in my profile or status
(0) ignore the brand and take no action
(+1) become a fan of [Brand]
(+1) search out [Brand] profile and more info
(+1) add and display a bumper sticker of the [Brand] logo that you found
(0) deny a bumper sticker of the [Brand] logo that someone else sent you
(+1) add and display a bumper sticker of the [Brand] logo that someone else
sent you
(+1) send someone else a bumper sticker of the [Brand] logo
(+1) send a [Brand] logo gift to someone else
(0) deny a [Brand] logo gift that someone else sent you
(+1) accept and display a [Brand] logo gift that someone else sent you
(+1) search out and join a group with a positive attitude toward [Brand]
(-1) search out and join a group with a negative attitude toward [Brand]
(0) ignore an invite to join a group with a positive attitude toward [Brand]
(0) ignore an invite to join a group with a negative attitude toward [Brand]
(-1) accept an invite to join a group with a negative attitude toward [Brand]
(+1) accept an invite to join a group with a positive attitude toward [Brand]
(0) Other:


108









Demographic Questions:


51. Age_

52. Sex
(1) Male
S(2) Female

53. What is your highest completed level of education?
S(1) High School
S(2) Associate Degree
S(3) Associate Degree in progress
S(4) Undergraduate Degree
S(5) Undergraduate Degree in progress
S(6) Graduate Degree
S(7) Graduate Degree in progress
S(8) Doctorate Degree
S(9) Doctorate Degree in progress

54. Race/Ethnicity
S(1) Asian
S(2) Black/African American
S(3) Hispanic/Latino
S(4) Native American
S(5) White/Caucasian
S(6) Other


109









LIST OF REFERENCES


Aaker, D. A., Kumar, V., & Day, G. S. (2004). Marketing research (8th ed.). Hoboken,
NJ: John Wiley & Sons, Inc.

About Facebook. (2004, February 4). Retrieved February 8, 2009, from
www.facebook.com: http://www.facebook.com/facebook

Arndt, A. Z. (2001). New economy emotion: Engaging customer passion with e-crm.
Chichester. England: John Wiley & Sons Ltd.

Arrington, M. (2009, January 22). Facebook now nearly twice the size of MySpace
Worldwide. Retrieved August 18, 2009, from TechCrunch:
http://www.techcrunch.com/2009/01/22/facebook-now-nearly-twice-the-size-of-
myspace-worldwide/

Babbie, E. (2007). The practice of social research (11th ed.). Belmont, CA: Thomson
Wadsworth.

Barker, V. (2009). Older adolescents' motivations for use of SNS: The influence of
gender, group identity, and collective self-esteem. International Communication
Association (pp. 1-45). Montreal, Quebec, Canada: Retrieved from Communication &
Mass Media Complete database.

Bauman, Z. (2005). Liquid life. Cambridge, England. UK: Polity Press.

Beavis, C. (2009). Games within games. In R. Willett, M. Robinson, & J. Marsh (Eds.),
Play, creativity and digital cultures (pp. 15-35). New York: Routledge.

Blumler, J. G., & Katz, E. (1974). The uses of mass communications: Current
perspectives on gratifications research. Beverly Hills, CA: Sage Publications.

Boyd, D. (2006). G/localization: When global information and local interaction collide.
Retrieved March 23, 2009, from O'Reilly Emerging Technology Conference:
http://www.danah.org/papers/Etech2006.html

Boyd, D. (2008). Can social network sites enable political action? International Journal
of Media & Cultural Politics, 4 (2), 241-244.

Boyd, D. M., & Ellison, N. B. (2007). Social network sites: definition, history, and
scholarship. Journal of Computer-Mediated Communication, 13 (1), 210-230.

Chartrand, T., & Bargh, J. (1999). The chameleon effect: The perception-behavior link
and social interaction. Journal of Personality & Social Psychology, 76 (6), 893-910.


110









Company history. (2008). Retrieved August 18, 2009, from Linkedln:
http://press.linkedin.com/history

Company Timeline. (2009). Retrieved December 2, 2009, from Facebook:
http://www.facebook.com/press/product.php#/press/info.php?timeline

Consalvo, M. (2006). Console video games and global corporations: Creating a hybrid
culture. New Media and Society, 8 (1), pp. 117-137.

Consortium, C. C. (2006). About C3: Convergence. Retrieved April 27, 2009, from
Massachusetts Institute of Technology:
http://www.convergenceculture.org/aboutc3/index.html

Curry, L. A. (2004). Affect, decision making, and adolescent risk behavior. Gainesville,
FL: University of Florida.

Davies, J. (2009). Online connections, collaborations, chronicles and crossings. In R.
Willett, M. Robinson, & J. Marsh (Eds.), Play, creativity and digital cultures (pp. 108-
124). New York: Routledge.

Davis, J. J. (1997). Advertising research theory and practice. Upper Saddle River, NJ:
Prentice Hall.

Dong, Q., Urista, M. A., & Day, K. D. (2008). Explaining why young adults use MySpace
and Facebook through uses and gratification theory. Human Communication, 12 (2),
215-229.

Douglas, J. E. (1999). The congruency between the ideal network and television
network's brand personality: What makes the difference? Gainesville, FL: University of
Florida.

Dowdall, C. (2009). The texts of me and the texts of us. In R. Willett, M. Robinson, & J.
Marsh (Eds.), Play, creativity and digital cultures (pp. 73-91). New York: Routledge.

facebook.com Quantcast Audience Profile. (2010). Retrieved Feb 13, 2010, from
Quantcast: http://www.quantcast.com/facebook.com

Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An introduction
to theory and research. Reading, MA: Addison-Wesley.

Franken, R. (1994). Human motivation (3rd ed.). Pacific Grove, CA: Brooks/Cole.

Goodman, R. D., Maggio, J., Fedde, R. A., & Lyman, W. (2008). Growing up online.
[S.I.]: PBS Video http://purl.fcla.edu/UF/lib/kidsonline/.


111









Hall, A. (2009). College students' motives for using social network sites and their
relationships to users' personality traits. International Communication Association (pp.
1-38). Montreal, Quebec, Canada: Retrieved from Communication & Mass Media
Complete database.

Henson, C. A. (2007). An analysis of the coffee service industry in metro manila and the
buying behavior of its consumers. Quezon City, The Philippines: Loyola Schools
Review.

Hoffman, D. L., & Novak, T. P. (1996). Marketing in hypermedia computer-mediated
environments: Conceptual foundations. Journal of Marketing, 60 (July), 50-68.

Holzner, S. (2009). Facebook marketing: Leverage social media to grow your business.
Indianapolis, IN: Que Publishing.

Jenkins, H. (2006). Convergence culture: Where old and new media collide. New York:
New York University Press.

Kang, K. (1999). Interrelationship of ideal, actual, and undesired self/brand congruities
across high, moderate, and low vaue-expressive products by Strausbaugh's brand
personality measure. Gainesville, FL: University of Florida.

Kaye, B. K., & Johnson, T. J. (2001). A web for all reasons: Uses and gratifications of
internet resources for political information. Association for Education in Journalism and
Mass Communication Conference. Washington DC.

Keller, K. L. (2008). Strategic brand management: Building, measuring, and managing
brand equity (3rd ed.). Upper Saddle River, NJ, New Jersey: Pearson Prentice Hall.

Ko, H., Cho, C. -H., & Roberts, M. S. (2005). Internet uses and gratifications: A structual
equation model of interactive advertising. Journal of Advertising, 34 (2), 57-70.

Korgaonkar, P. K., & Wolin, L. D. (1999). A multivariate analysis of web usage. Journal
of Advertising Research, 39 (2), 53-68.

Lange, R., & Lampe, C. (2008). Feeding the privacy debate: An examination of
Facebook. International Communication Association (pp. 1-26). Montreal, Quebec,
Canada: Retrieved from Commnication & Mass Media Complete Database.

Levi-Strauss, C. (1974). The savage mind. London: Weidenfeld and Nicolson.

Li, D. (2008). Behaviors and motivations of social networking site users: A cross-gender
and cross-cultural comparison. Gainesville, FL: University of Florida.


112









Lin, C. A. (1999). Online-service adoption likelihood. Journal of Advertising Research,
39 (2), 79-89.

Lin, J.-S. (2008). Antecedents and consequences of cross-media usage [electronic
resource]: A study of a TV program's official website. Gainesville, FL: University of
Florida.

Lopez, K. A. (2008). Student speech rights [electronic resource]: First Amendment
implications for high school students on popular online social networks, MySpace and
Facebook. Gainesville, FL: University of Florida.

Lury, C. (1996). Consumer culture. Cambridge, England, UK: Polity Press.

MacKenzie, S. B., Lutz, R. J., & Park, W. P. (1989). An empirical examination of the
structural antecedents of attitude toward the ad in an advertising protesting context.
Journal of Marketing, 53, 48-65.

Mitchell, A. A., & Olson, J. C. (1981). Are product beliefs the only mediator of
advertising effect on brand attitudes? Journal of Marketing Research, 318-332.

Nixon, H. (2003). New research literacies for contemporary research into literacy and
new media? Reading Research Quarterly, 38 (3), pp. 407-413.

Papacharissi, Z., & Rubin, A. M. (2000). Predictors of Internet use. Journal of
Broadcasting and Electronic Media, 44 (2), 175-196.

Park, L., Jin, B., & Jin, S.-A. (2009). Motivations, impression management, and self-
disclosure in social network sites. International Communication Association (pp. 1-36).
Montreal, Quebec, Canada: Retrieved from Communication & Mass Media Complete
database.

Parker, B. T. (2005). This brand's for me: Brand personality and user imagery based
self-congruity. Gainesville, FL: University of Florida.

Quart, A. (2003). Branded: The buying and selling of teenagers. London: Arrow Books.

Rogers, C. R. (1959). A theory of therapy, personality, and interpersonal relationships,
as developed in the client-centered framework. (I. S. Koch, Ed.) Psychology: A study of
a science, pp. 184-256.

Rychlak, J. F. (1981). Introduction to personality and psychotherapy. Boston: Houghton
Mifflin Company.

Severin, W. J., & Tankard, J. W. (2001). Communication theories: Origins, methods,
and uses in the mass media. New York: Addison Wesley Longman.


113









Shaw, M. E., & Costanzo, P. R. (1970). Theories of social psychology. New York:
McGraw-Hill.

Shirky, C. (2009, May). Clay Shirky: How social media can make history I Video on
TED.com. Retrieved June 16, 2009, from TED: Ideas worth spreading:
http://www.ted.com/talks/clay_shirky_howcellphonestwitter-facebook-can-make-hist
ory.html

Sirgy, M. J., Grewal, D., Mangleburg, T. F., Park, J.-o., Chon, K.-S., Claiborne, C. B., et
al. (1997). Assessing the predictive validity of two methods of measuring self-image
congruence. Journal of the Academy of Marketing Science, 25 (3), 229-241.

Sirgy, M. (1986). Self-congruity: Toward a theory of personality and cybernetics. New
York: Praeger Publishers.

Statistics. (2009). Retrieved December 2, 2009, from Facebook:
http://www.facebook.com/facebook?v=app_7146470109&ref=pf#/press/info.php?statisti
cs

Steinbock, D. (2000). The Birth of Internet Marketing Communications. Westport, CT:
Quorum Books.

Tinkham, S., & Weaver-Lariscy, R. (1993). A diagnostic approach to assessing the
impact of negative political television commercials. Journal of Broadcasting & Electronic
Media, 37 (4), 377-398.

Trusov, M., Bucklin, R. E., & Pauwels, K. (2009). Effects of word-of-mouth versus
traditional marketing: Findings from an Internet social networking site. Journal of
Marketing, 73, 90-102.

Van Cleemput, K. (2009). Authenticity and subcultural style in adolescents' self-
presentation on social network sites. International Communication Association (pp. 1-
18). Montreal, Quebec, Canada: Retrieved from Communication & Mass Media
Complete Database.

Willett, R. (2009). Consumption, Production and Online Identities. In R. Willett, M.
Robinson, & J. Marsh (Eds.), Play, creativity and digital cultures (pp. 54-67). New York:
Routledge.

Wu, J. (2009). Facebook politics: An exploratory study of american youth's political
engagement during the 2008 presidential election. International Communication
Association (pp. 1-23). Montreal, Quebec, Canada: Retrieved from Communication &
Mass Media Complete Database.


114









BIOGRAPHICAL SKETCH

Stefanie Riediger was born in Ontario, Canada, where she lived for eight years

before moving to a suburb of Cleveland, Ohio. Though she is still a Canadian citizen,

her education has taken place in American schools. Stefanie completed her

undergraduate Bachelor of Science in Business Administration degree at The Ohio

State University, where she majored in business with a specialization in marketing and

minored in visual communication design. After her graduation in the spring of 2007, she

went on to earn her Master of Advertising degree from the College of Journalism and

Communications at the University of Florida in the summer of 2010. Stefanie also

completed two internships with Bizresearch, a search engine marketing firm in

Worthington, Ohio, and Saatchi & Saatchi X, the shopper marketing division of the

global Saatchi & Saatchi advertising agency in Fayetteville, Arkansas. She is currently

pursuing a career in advertising with worldwide agencies and clients.


115





PAGE 1

1 THE EFFECT OF BRANDCONSUMER CONGRUENCY ON BRAND ADOPTION BEHAVIOR IN SOCIAL NETWORKING SITES By S TEFANIE RIEDIGER 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 2010

PAGE 2

2 2010 Stefanie Riediger

PAGE 3

3 ACKNOWLEDGMENTS I would like to express a heartfelt thank you and much gratitude to everyone who has h elped me in achieving my Master of Advertising degree. First and foremost, I am greatly in debt to the extended invaluable and diligent hands on assistance of my thesis chair, Dr. John Sutherland. He agreed to take on advisor duties and pick up where another advisor left off when it was n o t required of him His unyielding patience and encouragement was monumental. His guidance pushed me to produce quality work and helped me believe in my abilities to do so. I would also like express my gratitude to Dr. Jorge Villegas, Dr. Jon Morris and Dr. Robyn Goodman for their time and guidance with regard to my study. Further thanks go to Jody Hedge for her administrative help. I would also like to thank my loving parents who have been supportive of me every step of the way, pushing me to do my best. Their continued love and support helped remind me of what was important when the go ing got tough and discouragement set in. I could always count on them to keep me focused on my goals and what needed to be done to achieve them. I would also like to extend my thanks to my sister, Allison, my brother in law, Kevin, and many wonderful friends who were sources of encouragement in this process. They were always there for me. Lastly, I would like to thank those from my past who helped me realize my capabilities as a student, a professional, and most importantly as a person. They made me see that I really can work through anything by taking things one step at a time.

PAGE 4

4 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 3 LIST OF TABLES ............................................................................................................ 6 LIST OF FIGURES .......................................................................................................... 7 ABSTRACT ..................................................................................................................... 8 CHAPTER 1 INTRODUCTION ...................................................................................................... 9 2 LITERATURE REVIEW .......................................................................................... 13 A Digital Re volution ................................................................................................ 13 Participatory Cultures .............................................................................................. 13 Uses and Gratifications ........................................................................................... 18 Online P rofiling ....................................................................................................... 20 The New Word of Mouth ......................................................................................... 23 Self Congruity Theory ............................................................................................. 25 Hypotheses ............................................................................................................. 28 3 METHOD ................................................................................................................ 34 Research Design .................................................................................................... 34 Participants ............................................................................................................. 35 Data Collection and Procedure ............................................................................... 36 Measures and Instrument ....................................................................................... 40 Independent Variable ....................................................................................... 42 Dependent Variable .......................................................................................... 45 Motivation Interaction Variables ....................................................................... 47 4 RESULTS ............................................................................................................... 48 Data Analysis .......................................................................................................... 48 Sample Profile .................................................................................................. 48 Facebook Tendencies/Activity .......................................................................... 49 Brand Attitudes, Congruity, and Compatibility .................................................. 50 Independent, Dependent, and Interaction Variables ........................................ 51 Dependent Variable: Adoptive Behavior ........................................................... 52 Hypothesis Testing ................................................................................................. 53 Interaction Effect of SNS Motivations ..................................................................... 57 Research Questions ............................................................................................... 61 Limitations ............................................................................................................... 66

PAGE 5

5 5 DISCUSSION AND CONCLUSIONS ...................................................................... 82 Discussion .............................................................................................................. 82 Implications ............................................................................................................. 83 Implications Summary ............................................................................................. 97 Future Research ..................................................................................................... 99 APPENDIX A INVITATION .......................................................................................................... 103 B ONLINE SURVEY QUESTIONNAIRE .................................................................. 104 LIST OF REFERENCES ............................................................................................. 110 BIOGRAPHICAL SKETCH .......................................................................................... 115

PAGE 6

6 LIST OF TABLES Table page 4 1 Sample profile summary statistics ...................................................................... 68 4 2 Current Facebook profile behavior (direct expression) ....................................... 70 4 3 Current Facebook profile behavior (indirect expression) .................................... 71 4 4 Brand attitudes, brand congruity, and compatibility ............................................ 72 4 5 Summary statistics for independent and motivation variables ............................ 73 4 6 Dependent variable component correlations ...................................................... 73 4 7 Dependent variable measures ............................................................................ 74 4 8 Correlations ........................................................................................................ 75 4 9 Self brand congruity correlations ........................................................................ 75 4 10 Compatibility correlations ................................................................................... 76 4 11 Multiple regression for entertainment motivation variable (H3) ........................... 77 4 12 Multiple regression for information motivation variable (H4) ............................... 78 4 13 Multiple regression for social interaction motivation variable (RQ1) ................... 79 4 14 Multiple regression for convenience motivation variable (RQ2) .......................... 80 4 15 Summary of multiple regression significant relationships ................................... 81

PAGE 7

7 LIST OF FIGURES Figure page 2 1 Overview of the study ......................................................................................... 33

PAGE 8

8 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 THE EFFECT OF BRANDCONSUMER CONGRUENCY ON BRAND ADOPTION BEHAVIOR IN SOCIAL NETWORKING SITES By Stefanie Riediger August 2010 Chair: John Sutherland Major: Advertising The emergence of social networking sites as a medium for communication has provided a wave of word of mouth opportunities for adverti sers. This study investigates the self congruity theory and social networking site (SNS) usage motivations as contributing factors of brand ed posting behavior on SNS s through regression analysis This research reveal ed that brand attitudes self brand congruity, and motivational effects are unique per brand, making context and relevancy of brand s important factors for marketing campaigns on SNSs Compatibility was not significant ly related to branded SNS posting behavior, but a significant positive relationship occurred between self brand congruity and the likelihood of adoptive posting behavior All four motivations tested also had significant relationships with adoption behavior likelihood acting independently but none of the motivations interacted significantly with self brand congruit y in predicting behavior Thus to increase likelihood of SNS word of mouth individuals must increasingly f eel as though the typical user of a brand mirrors their own self image Independent of self brand congruity SNS usages for entertainment, informati on, social interaction, and convenience purposes can also indicate increased adoption behavior depending on the brand in question.

PAGE 9

9 CHAPTER 1 INTRODUCTION T o further develop the most successful brand communication strategies, it is helpful to study the connections made by a consumer that can ultimately lead to branded behavior and purchase decisions. This study is designed to observe social networking site (S NS) posting behavior and the intricacies of both consumer perceptions and attitudes toward a brand to explain what factors contribute to the likelihood of branded adoption behavior on SNS. This objective was achieved by examining the effect s of self brand congruity and brandSNS compatibility on branded posting behavior within an SNS. Previous studies have examined congruity in terms of a persons actual self image in comparison with his/her ideal image ( Parker, 2005; Sirgy et al., 1997 ). Parker (2005) al so applied this comparison to brand personalities The most relevant portion of this research, however, involves the integration of these ideas by looking at the alignment between s elf image and branduser image. Measures of congruity between ones self im age and a branduser i mage are determined as a result of this integration ( Parker, 2005; Sirgy et al. 1997 ). The resulting self brand congruity influences consumer behavior as a function of the brand attitude elements of familiarity, attachment, and loyal ty (Parker, 2005). The self congruity model within the realm of consumer behavior will act as a pivotal variable, along with the perceived compatibility of a brand with a particular SNS as possible explanations for SNS posting behavior. T o facilitate th e explanation of why consumers interact with brands within SNS s, the uses and gratifications th eory was adapted to the model. The assumption involved in this theory is that media users are active users who take actions in search of a goal

PAGE 10

10 (Blumler & Katz, 1974). Furthermore, Ko, Cho, and Roberts (2005) argued that these people are aware of their needs, and select the appropriate media to gratify their needs ( p. 60 ). The following investigation of the translati on of consumer identification with a brand int o adoption behavior of a brand relies heavily on formulated insight based on this theorys foundational assumptions. The assumption to be explored is that the higher a persons self brand congruity (or the more a consumer identifies his/her actual and ideal self with the typical user of a brand), the more likely that person is to adopt the brand into his/her SNS posting behavior. In addition, the uses and gratifications theory is applied with the prediction that a persons motivation for using a n SNS could affect how that person perceives brands within th at context and ultimately how that person posts. Thus, the final element of this study which contributes a new perspective to the research on consumer attitudes and behaviors is the medium withi n which consumers are observed. The Internet has a growing and encouraging capacity for interaction that has intensified the uses and gratifications theo ry (Ko, Cho, & Roberts, 2005). N ot only has the Internet been a natural incubator for many types of relationships and interaction, but more specifically, SNSs have emerged as prominent stimulators of social activity (Boyd, 2006). Until the development of online communities, connectivity and interaction was limited to oneway or two way communication (Shirky, 2009). H owever, with the development of SNSs in recent years, this capacity has multiplied exponentially due to the power of mass speech given to the individual. Webs of oneway and many way communication are formed simultaneously creating an entire new world in which

PAGE 11

11 marketers can immerse themselves and capitalize on an opportunity for premium word of mouth communication value (Shirky, 2009). One outcome of the social networking age is the practice of online profiling. The analytical nature of the Internet makes a succinct, yet encompassing first impres sion a commanding deliverable. The resulting trend of profiling oneself on Internet communities continues to grow, as SNSs such as Facebook facilitate a diverse range of communications (Willett, 2009). As these types of sites become increasingly popular, new advantages of reaching consumers through this medium pres en t themselves to marketers. The self branding that millions of SNS members engag e in invites an opportunity to research a brands self brand congruity role in the process. This study aims to explain self brand congruitys role by answering some unexamined fundamental questions involving the consumers core drive for branded posting behavior on SNS s. Many studies have been executed regarding usage of the Internet but a relatively limited number of these have been specified to SNS s. Even fewer studies have related the self congruity theory to SNS behavior. T he following research expan d s upon knowledge regarding posting behavior on SNSs, particularly Facebook. These cognitive connections are important for marketers because their past habits are becoming obsolete as strategies transition from mostly push to mostly pull by the consumer. It is also important to gain a better understanding of the thought proces s behind brand adopti ve posting behavior because these ac tions carry wordof mouth trust that is greatly valued in todays marketing practices. The following research collected data from undergraduate students at the University of Florida and registered users of the SNS, Facebook. The data were

PAGE 12

12 collected by an online survey method using www.surveymonkey.com The questionnaire involved was designed to measure Facebook familiarity and attitudes, brand familiarit y and attitudes, self brand congruity brandFacebook compatibility, and motivations for using Facebook as possible influential factors of branded adoption behavior likelihood. Facebook was chosen as the SNS of interest in the hope of projecting key findings onto SNSs as a whole. T he two brands observed in th is study were Starbucks and Dunki n Donuts. The resulting data were analyzed with the help of the computer software formerly known as Predictive Analytics Software (PASW Sta tistics 18 or PASW ), and now known as IBM SPSS Statistics (SPSS) Correlations and relationships between variables were measured using regression analysis to test hypotheses and explore research questions that emerged from the review of the literature of c onsumer behavior, social psychology, brand management, and new media communications.

PAGE 13

13 CHAPTER 2 LITERATURE REVIEW The introduction of social networking sites (SNS s) to the Internet has brought with it unchartered territory. For those with mass communication interests, SNS s have introduced questions and possibilities in areas such as consumer behavior, social psychology, brand management, and new media communications. The present study aims to provide answers to a few of these questions by looking principally at motivations and self congruity as possible explanations for SNS behaviors. A Digital Revolution S ocial N etworking sites are creating a new revolution within the broader Internet revolution that has taken place during the last 20 years (Shirky, 2009) The Internet has conveniently provided the platform for social networks to develop. In the following excerpt, Shirky (2009) discussed the social networking implications of the Internet from a filmed presentation titled How Social Media Can Make History: The Internet is the first medium in history that has native support for groups and conversation at the same time. Whereas the phone gave us the oneto one pattern, and the television, radio, magazines, books gave us the one to many pattern. The Interne t gave us the many to many pattern. As all media gets digitized the Internet also becomes the mode of carriage for all other media. That means that every medium is right n ext door to every other medium. Internet .is increasingly more a s ite of coordination. (3:47) Participatory Cultures While increased coordination has certainly facilitated globalization and changed cognitive connections, people are primarily concerned with the cultures in which they participate (Boyd, 2006) Thus, the focus has evolved on participatory cultures. The Convergence Culture Consortium (2006) defines participatory culture as follows : Participatory culture describes the way consumers interact with media content, media producers and each other as they explore the resources

PAGE 14

14 available to them in the expanded media landscape. Consumers become active participants in shaping the creation, circulation and i nterpretation of media content. Such experiences deepen the consumers emotional investment in the media property, and expand their awareness of both content and brand. (p. n.p.) This increased awareness and active participation involves uploading content, adapting existing content, working collaboratively with others, and forming online connections as core activities (Willett, 2009) As participants in this respect, Internet users are considered to be active agents who display competency, personality and knowledge (Beavis, 2009) Social networking sites play an important role in this agency and participatory development, taking the Internet s coordination function mentioned by Shirky (2009) a step further. A n SNS is a type of online community and partici patory culture which typically follows a basic format consisting of a personal profile, a blog and/or message board, a private messaging utility, and a news feed that reports friends recent activities (Boyd & Ellison, 2007; Li, 2008) Boyd and Ellison (20 07) identif ied three defining features of SNSs They allow individuals to 1) construct a profile that is at least semi public, 2) formulate a list of other users with which they share a connection, and 3) navigat e and interact with other users profiles within that list (Boyd & Ellison, 2007) M any different SNSs are available these days for a variety of purposes The nature and nomenclature of connections via SNSs may vary from site to site, but they all generally serve the basic function of facilitating connect ions (Boyd & Ellison, 2007) For example, LinkedIn, founded in 2002 (Company History, 2008) is a n SNS with the specific purpose of making business and career connections. Some SNS s serve to unite those with certain health conditions such as Juvi nation which brings Type 1 d iabetics together to share stories, tips, and information. Furthermore, some sites such

PAGE 15

15 as Friendster which originally aimed to inspire more romantic connections were transformed by entertainers as a platform for their promot ions (Boyd & Ellison, 2007) Friendster ultimately ran into its demise by not embrac ing this new direction, but MyS pace was able to successfully capitalize on the networking opportunity in its wake (Boyd & Ellison, 2007) Moreover, Facebook was initially i ntended to serve a niche in the Harvard University only community before its popularity transitioned it into a global phenomenon. Since their launches, MySpace and Facebook have emerged as the leading communities with the highest number of members (Lopez, 2008) As of 2006, MySpace totaled 100 million users and has continued to linger around 100 million unique users in the period since (Arrington, 2009) Likewise, MySpaces highest rival, Facebook, grossed 18 million users in 2007 with an estimated 90% of all undergraduates at colleges and universities where Facebook is available are registered users of the site. Of those users, 60% log on daily (Lopez, 2008, p. 25) I n June 2009 Facebook tallied more than 200 million active users with more than 100 mill ion of those logging on at least once per day (Statistics, 2009) Most recently, less than one year later, Facebook reported more than 350 million active users in December 2009 with the average user spending 55 minutes of his/her day on the site (Statistic s, 2009) Facebook, initially launched in February 2004, is an SNS that aims to help people connect with others and the world around them (About Facebook, 2004) As with many other SNSs, Facebook allows users to share information and media such as photos, videos, and links to other online media. By providing these functions Facebook allows users to develop and maintain relationships with friends and family on a consistent

PAGE 16

16 basis (About Facebook, 2004) Typically, these personalized networks are mostly compr ised of connections previously made offline, m eaning that SNSs Facebook in particular, are less frequently used to make new connections (Boyd & Ellison, 2007) Facebook is unique from other SNSs, however, in that it originally began as an SNS solely for Harvard University students before expanding to other universities A person was required to have a verifiable college email address in order to register with the site. Facebook later expanded its registration requirement to all valid email addresses by S eptember 2006 (Company Timeline, 2009) According to the third party audience analysis site www.quantcast.com (2010) f acebook .com traffic is skewed slightly female (55%) and heavily Caucasian (75% ), with the highes t trafficking age group being 18 to 34 year olds (42% ). Furthermore, when compared with similar SNS s, Facebook has the highest percentage of traffic to have a college education, accounting for 53% of Facebook traffic (facebook.com Quantcast Audience Prof ile, 2010) The initial exclusivity to college students, the foundation of college educated users, and its topranked popularity among SNS s are all reasons Facebook was chosen for this study. For a sample of primarily undergraduate college students and rec ent graduates, Facebook is most likely to present respondents already familiar with the SNS With such social utilities as the ones provided by Facebook at hand, SNS s can be used for a variety of purposes. According to Boyd (2008) on the subject of political action and SNS behavior our society is status obsessed and narcissistic and typical social network site users are more invested in adding glitter to pages and SuperPoking their friends than engaging in any form of civ ically driven collectiv e action (p. 241).

PAGE 17

17 Research has indicated that approximately 40% of SNS users choose to include a self description, invitation to contact, hobbies and interests, and references to friends and relationships in their online profiles (Van Cleemput, 2009). Po sting photographs is a nother highly employed self identifying behavior observed. Lange and Lampe (2008) discovered that 88.9% of the participants they observed displayed an identifying personal photograph. With so much personal information available in one location, a nother study found, not surprisingly, that participants use Facebook to learn more about others, as the level of disclosure on Facebook may be higher and have more detail than disclosure in real lif e (Lange & Lampe, 2008, p. 15). Interchangea bility between online and offline communication and activity is also a common theme in S N S posting behavior. For example, participants tend to use Facebook to coordinate offline, real life activities (Lange & Lampe, 2008) Transposable o ffline worlds for o nline worlds are corroborated by the fact that most SNS users draw on SNSs to gather with friends when physical copresences is imposs ible or impractical . For many of the most active participants on social network sites, networked publics substitute for physical publics because physical publics are inaccessible, untenable, heavily regulated, or downright oppressive (Boyd, 2008, p. 242 ) These are a few of the SNS posting trends observed thus far. However, many questions have been left unanswered as new trends add and transform existing ones. A basic understanding from prebehavior explanations to post behavior observations would help provide detailed answers to these questions. Thus, in looking at a foundational measure, motivations and the uses and gratifications perspective can help attach meaning to these purposes for SNS behavior

PAGE 18

18 Uses and Gratifications Research has indicated that motivations can be of great consequence to SNS behavior (Beavis, 2009) The uses and gratifications theory states that varying reasons exist to use a mass medium (Blumler & Katz, 1974) This theory explores intrinsic needs and motivations that lead to an individuals choices to perform specific behaviors (Blumler & Katz, 1974) The behaviors chosen serve to sa tisfy those needs; needs which can vary greatly between Internet usage and for other mediums (Ko, Cho, & Roberts, 2005) T his theory assumes that media users are active users who take actions in search of a goal (Blumler & Katz, 1974) Furthermore, Ko, Cho, and Roberts (2005) argued that these people are aware of their needs, and select the appropriate media to gratify their needs (p. 60). For example, Wu (2009) stated that regarding political behavior on Facebook, e ndorsement of either an entertainment celebrity or a political figure indeed is first and foremost built upon strong affective affiliation, hence rendering a strong urge in the supporters to vocally cheer their heroes (p. 17) This urge to express support for their heroes r epresents the need in this case, and since the features of Facebook provide extended opportunity to share political ideas and other original political materials such as videos and transcripts of speeches (Wu, 2009), Facebook became the appropriate medium used to gratify this need. Thus, t he Internet is a growing and encouraging capacity for interaction that, as a result, has intensified the uses and gratifications theory (Ko, Cho, & Roberts, 2005) Korgaonkar and Wolin (1999) looked at 41 items related to Internet usage and concerns and found indications that consumers use the Internet for retrieving not only information, but also entertainment and escape. Furthermore, a previous study looked at the influence of Internet usage motivations on adoption b ehavior toward online services.

PAGE 19

19 This study showed, depending on the type of website (information, infortainment, shopping) that different motivators were key influencers on behavior (Lin 1999) Applying this same logic to SNS s would indica te that different users of SNS s could view SNS s in different capacities depending on their motivation for utility. Just as with various websites, SNS s can serve many functions within themselves Therefore, it should follow that the same concept of matching different motivational influencers to varying levels or types of behavior may be applicable within the more specific context of SNS s. As a result, some motivations may have a greater interaction ef fect on SNS behavior than other motivations In concert with this, the impact of motivations is further amplified by the interactivity of the Internet The Internet allows users to participate at an elevated level by giving them control, according to their preferences and needs, over the advertising messages they receive, the amount of information they give and/or receive, and the presentation method of such content (Hoffman & Novak, 1996) According to Papacharissi and Rubin (2000) a key component to the uses and gratification s theory is audience activity, which is heavily impacted by motivations. Ko, Cho, and Roberts (2005) extended this idea to say that the four main motivations of Internet usage are entertainment, information, social interaction, and convenience, and that each of these motivations were strongly related with distinct types of interaction. S everal other motivations for using SNSs have been explored, including time passing entertainment, relationship development, relationship maintenance, trendfollowing (Hall, 2009) impression management, self disclosure, perceived reciprocity (Park, Jin, & Jin, 2009) communication maintenance between friends, and social compensation

PAGE 20

20 (Barker, 2009) U pon review of these previously researched motives for using SNSs the majority categorize themselv es nicely within the four main motivations to use the Internet ( entertainment, information, social interaction, and convenience) set forth by Papacharissi and Rubin (2000) in a way that best suit s the purpose of this study. With SNS s mirroring many of the same functions as the Internet and the vast majority of researched motivations for using SNSs fitting into these four categories it seems appropriate to apply the motivation categories of entertainment, information, social interaction, and convenience to SNS behavior for the purpose of this study. Online Profiling The manifestation of behaviors and motivations on SNS s begins with the heart and soul of a social networking site: the construction of a profile page (Boyd & Ellison, 2007) This personal profile serves to represent the participant on a continual basis and acts like a personal hub for each users participation (Goodman et al. 2008) Through this profile creation, users provide understandings of themselves, about themselves and for themselves in relation to others (Dowdall, 2009, p. 78) including in relation to brands. This personal profile becomes a continuation of their offline social existence, which is extended into a crafted online identity. This morphing of offline and online life even extends to daily conversations. Conversations begun in offline situations are continued online and vice versa. The boundaries often become blurred ( Davies, 2009; Dowdall, 2009; Goodman et al., 2008) The idea of identity creation is not limited to basic profile creation, but also extends to other forms of posting behavior. Willett (2009) state d that by posting content, particularly mediarelated content, todays youth can be seen as creating identities. Although previous research has shown that identities implied by online profiles are

PAGE 21

21 strongly related to their matching offline identities (Van Cleemput, 2009) this inclination does not eliminate tendencies to play with it to some degree. Social networking sites which combine blogs, profiles and photo and videosharing can be viewed as cultural resources which are used by young people as a way of performing and perhaps playing with their identity (Willett, 2009, pp. 5556) In this way, users take resources available to them and modify, reapply and recontextualize them into a bricolage (Levi Strauss, 1974) that creates a new meaning and a unique identity (Willett, 2009) Material objects, products, and brands are often used as markers of identity (Lury, 1996) This idea is part of a bigger fan culture. T he Consortium (2006) labels this Brand Culture and defines these as communities of committed consumers of specific products and services (such as BMW drivers, iPod users and Coke collectors) (p. n.p.). In conjunction with this, Quart (2003) proclaim ed that consumer culture pegs teens as target consumers, but that the reverse is also true in that teens are branded objects themselves. As a result, a new landscape is being formed where products and services are no longer simply developed and presented to the consumer in a uni directional flow Instead, these fan bases are now providing feedback productions of media, text, and interpretations (Jenkins, 2006) Products and the content that accompanies them are now being modified uniquely by individuals who identify with them. Th e se personal connections and personal influencers have more impact than previously considered. Beavis (2009) found in her research on convergence that with respect to convergences of fan, brand and style culture [such as on social networking sites], brand and fan culture had less of a role to play but style culture, and [students] friendships and self concepts mattered considerably (p. 32)

PAGE 22

22 This trend is particularly apparent among younger users. In a PBS documentary about youth today on SNSs a high school boy claim ed, You need to have the Internet on to talk to your friends [because] everybody uses it. Its like a currency. If you [do n ot] use it youre going to be at the loss (Goodman et al, 2008, 6:14 ) The online world has become a familiar parallel universe to many people, and as such social networking sites are also increasingly the place where kids hash out their conflicts (Good man et al. 2008, 14:46), once again blurring the lines between online and offline life. One student interviewed about her Facebook profile replied, Pretty much everyone has one. Its like a section of the Internet that is your own. Like, you can make it your personality exactly (13:13) One expert in the document, Pascoe, explained : i n a way, the social networking sites are a digital representation of what we think of as adolescence. So what teens are doing are going around and trying on these different identities. Im a goth, or Im a surfer or Im a punk rocker, or Im a this or that, and the Internet s allowed them to display that identity in a very dramatic and very distinct way (18:26) This kind of play is adopted by adults as well particularly through connectivity Connectivity is increasingly part of adults working and social lives; like the young, they use the Internet as a way of exploring new ways of connecting socially (Davies, 2009, p. 110) Pascoe su ggested that an opportunity for play with regard to identity is one component of profiling (Goodman et al. 2008) Thus, profiling can be described as more of a means oriented practice as opposed to a goal oriented practice that can be applied by SNS users of all ages (Bauman, 2005) As a means oriented practice, a user is likely to utilize a very liquid structure, in which a profile can be modified as often and as much as a user prefers This means that a profile is continually evolving, just as the

PAGE 23

23 actual user. In this case, content remains only so long as the user still feels it is congruent with his/her own current identity to some degree (Dowdall, 2009) The New Wordof Mouth Social networking sites have also impacted wordof mouth practices. As a result of the many to many pattern discussed previously (Shirky, 2009) and the practice of identity creation using products and brands as markers of such identity, SNSs are a prime medium for advanced wordof mouth opportunities, which can be appealing to mark eters. Word of mouth is already a highly valued asset for marketers. I n the SNS environment, reach is significantly large snowball effects thrive from multi directional conversations and electronic recording of activity translates into word of mouth that is more traceable and measureable than ever before (Trusov, Bucklin, & Pauwels, 2009) A ttaining t his wordof mouth advantage presents a powerful opportunity. Due to the augmented coordination capacity of the Internet and online identity creation trends, as mentioned earlier, users can now act as producers and consumers (Shirky, 2009). In this way, traditional advertising efforts are being restructured to a pull strategy, which means that companies are increasingly placing consumers in control. Lester Wun derman, considered by many to be the creator of todays direct marketing, thought direct mail to be similar to a game of shooting a target in the sense that the mail is the bullet that hits the target (selected consumers) every time (Steinbock, 2000). The Internet presents itself to this utility as uniquely opportunistic due to its adoption rate, ubiquity, and constant access to and ability to distribute virtually limitless content (Arndt, 2001). The realm of consumer relationships has consequently been alt ered by the moderation of the Internet R egarding the birth of the Internet as an advertising medium, Wunderman progressively stated, I wanted something different. I

PAGE 24

24 wanted a medium where the advertiser would become the target and the consumer would become the shooter. Thats whats happening. It is a profoundly different marketplace (Steinbock, 2000, p. 127). This vision has matured with the emergence of major SNS s such as MySpace and Facebook in the period since Wundermans statement in 2000. The interaction changes accelerated by SNS s have redefined how people engage with brands by allowing the relationships to be more personal and individualized as if the company is simply another person to befriend (Boyd, 2006). Wunderman stated that with the rise of interactivity, more companies would have to become relationship marketers and win new clients one at a time. In the process, computers would substitute human interactivity (Steinbock, 2000, p. 129) Social networking sites facilitate this i ndividualized relationship marketing by attai ning clients and consumers one friend or fan at a time, creating more customized and focused relationships with consumers who freely elect to do so. The creator and CEO of Facebook, Mark Zuckerberg observed this trend by saying, [In] the last hundred years the way to advertise was to get into the mass media and push your content . In the next hundred years information [will not] just be pushed out to people, it will be shared among the millions o f connections people have. Advertising will change. You will need to get into these connections (Holzner, 2009, p. 1 ) Discovering the factors behind SNS branded posting behavior will bring marketers one step closer to being in the heart of these connections. A s previously stated, the consumer is therefore increasingly in control, but the many to many communication pattern is also allowing brands to mold their products and marketing efforts to better fit their consumers according to this consumer feedback (Shirky, 2009) Previously marketers simply strove to influence and promote

PAGE 25

25 word of mouth about their brands. Consumers are showing through participation that they are willing to talk. They are willing to adopt brands, ideas, and concepts as their own, as defining characteristics of themselves (Boyd & Ellison, 2007) With the development of SNS s, marketers can still promote branded wordof mouth, but now have the power of the many to many discussions that characterize SNS s behind them and an increased abil ity to compactly observe and quantify results (Trusov, Bucklin, & Pauwels, 2009) Boyd (2008) argued on behalf of social activists that given the typical friend overlap in most networks, many within those networks hear the same thing over and over until they believe it to be true [which gives] the impression that activists have spread a message further than they have ( p. 243) This not only applies to social a ctivists, but also branded wordof mouth. In addition to this increased effectiveness, con sumers voluntarily open doors on SNSs for brands to share valuable content and promotions by electing to be friends or by becoming a fan of the brand. While marketers goals are still to get people to talk about their brand, the difference is that on a n SNS those people have a place highly conducive to discussions about that brand, and brands can even join the conversation. Just as different national identities have been mixed in the hybrid [of online and offline worlds], so too the realms of business and culture are converging in novel ways (Consalvo, 2006, p. 120) These transition s make understanding consumer connections, motivations and behavior even more relevant. Self Congruity Theory Two theories, the uses and gratifications theory and the self congruity theory, help explai n what factors contribute to an individuals branded activity on SNS s. As

PAGE 26

26 previously discussed, the uses and gratifications theory provides a focus on how motivating factors are likely to influence the usage of a medium to gratify specific needs (Blumler & Katz, 1974) Thus, in alignment with this theory, it should follow that a persons motivation for using an SNS is likely to impact his usage behavior. To supplement this approach in a socially influenced setting, image concepts and perceptions can be very influential on public behavior decisions (Shaw & Costanzo, 1970) Thus, a self congruity theory perspective could contribute insight into posting behavior. The s elf congruity theory applies self congruity effects on self expr ession to the attitude theory by suggesting favorable predispositions toward brands that enhance or confirm perceptions of their own self image (Sirgy 1986) Essentially it is a social cognition theory due to the fact that an awareness of self serves a fundamental role and application in life experience (Rychlak, 1981) The a ttitude theory explains consumers favorability (likes and dislikes) in connection with consumer needs, which ultimately drives consumption behavior (Sirgy, 1986) Subsequently, sel f congruity is the match or mismatch perception associated with a consumers comparison between self image and other stimuli such as brand image, product image, or company image (Sirgy, 1986) Moreover, Parker (2005) explained the implications of this in t erms of brand attitudes: Self brand congruity, the comparison of self to brand, affects brand attitudes particularly when the social signaling value of a brand is high (i.e., used in a public situation) and when symbolic, self expressive motivations are involved. Positive brand attitudes should result as the similarity between brand image and the consumers self image increases (p. 4) Self brand congruity as used by Parker (2005) and Sirgy ( 1986) is the congruity match between self image and brandimage. Sirgy et al. (1997) used the phrasing typical user of a brand to measure brandimage instead of simple and direct references to ones

PAGE 27

27 image of a brand itself because personal images of a product reflect the stereotype of the generalized users of that product and are determined by a host of factors such as advertising, price, and other marketing and psychological associations (p. 229). In his research Sirgy (1986) examined self image and its comparison to the actual perceptions of characteristics of the I or me, as Rogers (1959) defined self, as well as the perception of the ideal self, which refers to how a person would like to view himself Self image is a subset of self based on self awareness that can change in different social roles (S irgy, 1986) This construct is a key indicator for understanding consumer behavior (Parker, 2005) Rogers (1959) state d that individuals are motivated by a basic and fundamental actualizing tendency that one hopes will at the very least maintain, if no t enhance a persons self image. Furthermore, Franken (1994) indicates the structural basis on how the concept of self becomes actualized: There is great deal of research which shows that the self concept is, perhaps, the basis for all motivated behavior. It is the self concept that gives rise to possible selves, and it is possible selves that create the motivation for behavior ( p. 443) Thus, both the uses and gratification s theory and self congruity theory provide rounded reasoning for a resulting conclusion that the more a person thinks that a brand is congruent with his own self image s, the more he will be motivated to behave in ways that re inforce those self image s on SNS s for various gratification purposes. If a persons natural tendency is to self actualize according to his motivations as Rogers (1959) claimed it seems probable that such a person would display these behaviors publicly via his own profiles and through social interaction on SNS s

PAGE 28

28 Hypotheses This study expanded upon previous theoretical models involving antecedents of media usage, particularly SNSs and congruity between an individual and a brand within such a medium. The constructs here are dependent upon familiarity and loyalty with the brand and the medium. F amiliarity and loyalty are then used to shed insight on a persons motivation and likelihood to integrate a brand on his SNS profile as a personally identifying charact eristic. The operational definitions in this case are important to clarify due to the colloquially interchangeable use of terms such as online communities social networking utilities, online social networks, and social networking sites. For the purposes of this study social networking sites ( SNS s) will be the term used. Accordingly, the concepts to be measured in this theoretical development include self brand congruity brandSNS compatibility, brand attitude, SNS usage motivations, and finally the resulting brand adoption behavior to ones SNS activity A brand, according to the American Marketing Association is a name, term, sign, symbol, or design, or a combination of them, intended to identify the goods and services of one seller or group of sellers and to differentiate them from those of competition (Keller, 2008, p. 2 ) However, Keller (2008) widen ed this definition to a more universal industry concept by stating that a brand is also something that has actually created a certain amount of awareness, reputation, prominence, and so on in the marketplace (Keller, 2008, p. 2 ) Though he limit ed this brand development to the marketplace, people in general are increasingly branding themselves through the pheno menon of online profiling; creating the awareness, reputation, and prominence that Keller (2008) refer red to. It is this consumption need of self expression that often drives consumers to

PAGE 29

29 purchase brands (e.g., Gucci and Rolex) that communicate a particul ar image or social role (Parker, 2005, p.10 ) Thus, they are creating a brand for themselves often through a selected compilation of other brands in the marketplace. Some insight into this self expression practice can be provided by the self brand congrui ty concept. Self brand congruity is based upon attitudes regarding oneself and a brand or rather, the typical user of a brand. Attitudes are typically researched in marketing and advertising because they have shown to perceptively predict consumer behavi or (Mitchell & Olson, 1981) The core influences of behavior are the cognitive building blocks, or beliefs, which are considered to be fundamental in developing attitudes (Fishbein & Ajzen, 1975) Thus, an individuals beliefs are likely to be highly translated into his introspective attitudes and their attitudes toward brands. Since attitudes have been shown to mold behavior in past consumer behavior research, it is not unreasonable to presume that these beliefs and attitudes can be translated into measur eable, personally identifying actions (Parker, 2005) Self brand congruity is defined by this perceived alignment of attitudes toward a brand and the belie fs that make up their identity. For the purposes of this study, the resulting behavior likelihood is defined as the likely or unlikely adoption of a brand to be presented in any fashion on an individuals Facebook activity or profile page, referred to as brand adoption behavior likelihood in this study This behavior is proposed to be indicative of a persons actual and ideal self brand congruity with the particular brand. H ypothesis 1: S elf brand congruity (both actual and ideal) will be positive ly related to the likelihood of SNS brand adoption behavior

PAGE 30

30 It is possible, however, that self brand congruity may be present, yet not be exhibited on a n SNS because the brand or product category may not be considered suitable for a public pers ona so openly broadcasted. For example, a product may not be deemed as relevant, applicable, appropriate, or natural for Facebook topics. Thus, the individual may not consider a brand or product category compatible with Facebook. This possible lack of compatibility must be addressed in an effort to make this research more reliable. Thus, the more an individual consi ders a brand as being relatable and suitable for Facebook, the more likely that person will be to display such accordance. H ypothesis 2: Brand SNS compatibility will be positive ly related to the likelihood of SNS brand adoption behavior As previously stated, in order for a brand to be selected by the individual for such self expressive prof iling, it is important that he have a favorable consumer attitude toward it. The same could also be true for the rev erse situation. If a person has a strong unfavorable attitude toward a brand, he may consider that anti congruity to be an integral part of his personal characteristic set as well. With this assumption in mind, the current study will assume d this association to be true and executed evaluations solely in pos itive terms. Parker (2005) declared, In general, consumers tend to favor brands and products that satisfy their needs and wants better than competitive choices ( p. 10) This statement implies a working application of the uses and gratifications theory. The four primary motivations for Internet usage laid out by Kaye and Johnson (2001) and Papacharissi and Rubin (2000) we re information, convenience, entertainment, and social interaction. As previously discussed, t hese four fundamental Internet usage

PAGE 31

31 motiv ations can be applied more specifically to the use of SNSs to derive insight on their effect on behavior in this new context. Within these four motivations, previous studies have indicated that media with higher interactive value or usage value (such as e ntertainment or information), tend to induce higher motivation in consumers to use them (Ko, Cho, & Roberts, 2005) C oncurrent with this finding, the entertainment and information motivations should inspire a positive influence on brand adoption behavior l ikelihood. H ypothesis 3: An entertainment SNS motivation will positively interact with both actual and ideal self brand congruity in predicting SNS adoption behavior likelihood. H ypothesis 4: A n information SNS motivation will positively interact with both actual and ideal self brand congruity in predicting SNS adoption behavior likelihood. Furthermore, social interaction is highly impacted by perceived social consequences A ssociations are therefore carefully expressed. One study found that when users express their political affiliations, they are typically prompted to justify their choices (Wu, 2009) Such accountability and s ocial judg ment would likely inflict a sense of reserve and prudence in freely admitting to associations unless the individual feel s very strongly about them. Moreover, if a person is using an SNS out of convenience motivations, it seems unlikely that the person would exhibit the effort of associating himself with a brand in extended behavior or exhibiting critical evaluation involved in justifying such associations (Wu, 2009) Therefore, both of these motivations can be expected to stimulate a negative influence on the addition of a brand to a persons SNS posting behavior Thus t he following research questions arise : R esearch Q uestion 1: Do social interaction SNS motivations negatively interact with actual and ideal self brand congruity in predicting SNS brand adoption behavior likelihood?

PAGE 32

32 R esearch Q uestion 2: Do convenience SNS motivations negatively interact with actual and ideal self brand congruity in predicting SNS brand adoption behavior likelihood?

PAGE 33

33 Figure 21 Overview of the study Self Brand Congruity Brand SNS Compatibility Adoption Behavior Likelihood SNS Usage Motivations Entertainment (H3) Information (H4) Social Interaction (RQ1) Convenience (RQ2) Interaction Effect (H3, H4, RQ1, RQ2) (H1) (H2)

PAGE 34

34 CHAPTER 3 METHOD The purpose of this study was to explain branded posting behavior on SNS s. Such answers would ideally provide brand management answers for how to excel using social media as a pivotal resource. The goal wa s to gain insight from active SNS participants and consumers that could unlock new approaches to brand management. This study execute d this goal primarily by looking at self brand congruity measures but also by obs erving participants motivation( s) for posting on a n SNS as an interacting effect. The data for this research were collected using an online survey with self administered questionnaires. The variables measured in reference to the leading SNS, Facebook, included 1) familiarity with Facebook and the brand, 2) attitudes toward Facebook and the brand, 3) current Facebook behavior, 4) image perceptions regarding self brand congruity as well as brandSNS compatibility, and 5) their resulting adoption behavior likelihood. Research Design This study employed an online survey to fulfill i ts research purpose. The questionnaires for this survey were available for completion on www.surveymonkey.com from May 15, 2009 through May 26, 2009 ( Appendix B). Sample participants received an invitation to complete a questionnaire, which directed them t o the appropriate link ( Appendix A ) In completing the questionnaire, each participant was asked to evaluate himself an individual brand, and Facebook on qualities such as familiarity, image perception, self congruity and compatibility between the brands and Facebook. These constructs were assessed using Likert scales with five levels. Variables were typically coded with

PAGE 35

35 the most positive response equaling one, and more negative responses were assigned incrementally increasing values. The data for each q uestionnair e were coded and recorded into a SPSS (formerly PASW) database. The data were then analyzed for correlations with actions taken in reference to the brand on Facebook through regression analysis. Descriptive and frequency analyses were also executed to supplement these observations. P articipants The sampling method employed in this case was a nonprobability purposive convenience sample. Purposive sampling depends on the researcher to select participants within their convenience and judgment who m ost appropriately represent the targeted sample (Babbie, 2007) To obtain a valid sample of SNS users within the scope and constraints of this study, purposive sampling was combined with the snowball sampling method, where a researcher collects data on the few members of the target population he or she can locate, then asks those individuals to provide the information needed to locate other members of that population whom they happen to know (Babbie, 2007, p. 185) This combination was especially appropriate for this study because it asked the population, Facebook users, to participate in a way that mirrored the kind of SNS behavior in which the study was interested. Thus, to fulfill the purpose set out for this res earch, a total of 151 respondents participated in the study Of these 151 participants, 74 responses qualified for inclusion in analysis of this survey. These participants were registered users of Facebook and coffee drinkers who were at least 18 years of age.

PAGE 36

36 Data Collection and P rocedure Surveys are used for studies with descriptive, explanatory, and exploratory purposes where the units of study are typically individual s as opposed to groups or interactions (Babbie, 2007) Babbie (2007) also stated that survey research is probably the best method available to the social researcher who is interested in collecting original data for describing a population too large to observe directly, and it can be an excellent vehicle for measuring attitudes and ori ent ations in a large population (p. 244). Privacy limitations of SNS s and the vast population of individual SNS users eliminate the possibility of observing the entire population. As a result, survey research that relies on a representative sample to charact erize the population was the most effective method to provide the attitudinal and behavioral insight desired in this study. This method was utilized to gain exploratory insight on participants due to the succinctness of both the method and the medium. F ocu s groups are typically most appropriate for exploratory inquiries where li mited research on the specific subject in question has been done previously and inductive reasoning is required. However, they carry a risk of t he chameleon affect occurring. The cha meleon affect is the nonconscious mimicry of the postures, mannerisms, facial expressions, and other behaviors of one's interaction partners, such that one's behavior passively and unintentionally changes to match that of others in one's current social environment (Chartrand & Bargh, 1999, p. 893) Instead, an online survey allow ed for much of the same questions to be asked as in a focus group or personal interview but with distinct and definitive answers that mirro red the SNS medium in question. T hus minimal, if any, valuable insight was lost. The objective of this exploratory study was to obtain measurable and meaningful

PAGE 37

37 information from participants by identifying motivational, attitudinal and behavioral patterns. To provide sound explanation for the sampling method chosen, it is important to clarify why Facebook was the social networking site chosen for this research. As explained earlier, SNS s often target specific groups of users. Therefore, in selecting a ma nageable representation of SNS s, the scope of this research was narrowed down to Facebook for a variety of reasons. First, Facebook is an SNS with one of the highest number of members and the most registered activity by facilitating interaction and sharing of news and media between individuals and groups (Lopez, 2008) Beyond this top ranked popularity, Facebooks users are fair ly diverse demographically, but are highly saturated among college students due to its initial exclusivity as a utility solely for college students (About Facebook, 2004) Facebooks qualities position it as most aligned with the available convenience sample population, but also most likely to provide respondents who are familiar with the specific social utility. The sample chosen mir rors the users of Facebook. The first execution of this sampl ing was done via an invitation requesting Facebook users participation in the study which was posted directly on Facebook. Facebook viewers of the invitation were encouraged to repost it and pass it along to others within their network This invitation explained what the study regard ed, as well as what participation meant for them in ter ms of time, effort, and confidentiality. A second purposive convenience sample was used in conjunction with the first sample. The sample population for this group wa s college undergraduate students between the ages of 18 and 25, enrolled in Journalism and Communication classes at

PAGE 38

38 the University of Florida for the 20082009 academic year The same invitation to participate was distributed to students in three classes within the University of Floridas Mass Communication department. Students participated voluntarily and, at the discretion of each classs professor, the classes were given the opportunity to earn course credit for the completion of the questionnaire. T o keep the questionnaire within a reasonable length, one questionnaire was dedicated to eac h of the two brands used. Participants were directed, according to last name to one of the two questionnaires to complete. Similar to the carefully chosen SNS for this research, the brands observed in the questionnaires were also well considered. Without establishing a level of familiarity with a brand, answers to subsequent questions regarding that brand would carry little meaning, if answers were possible at all. Thus, familiarity was an important founding construct of this study, and finding brands wit h which participants were highly familiar improved the research validity. As a result, two brands were chosen: Starbucks, a well known brand internationally, and Dunkin Donuts, a well known coffee provider in the U nited S tates but with a slightly different consumer base. The product category of coffee distributors was chosen for its prominence among college students, as it remains a staple in many college students lives. It was also ubiquitous enough that regardless of coffee consumption levels, the sample recruited would likely have attitudes and beliefs regarding the brands within the category. In fact, in a study about the differences between inhome coffee drinking and instore consumers, 46% of respondents ordered noncoffee products, which implies a dditional reasons for developing attitudes toward

PAGE 39

39 coffee shops than the mere liking of coffee (Henson, 2007) Furthermore, according to Parkers (2005) pretests, Starbucks was among the top 25 publicly consumed brands. A pre test was executed to determine the most appropriate brands for the study. Brands tested in the pretest included two national companies Starbucks and Dunkin Donuts, and three local companies in Gainesville, Florida, Barnies Coffee & Tea Co., Maudes Classic Ca f, and Lollicup Coffe e & Tea. The pretest was very brief and included questions evaluating familiarity, attitudes, self brand congruity and brand compatibility with Facebook. Starbucks and Dunkin Donuts proved to be the most appropriate brands for this study. T hey both had similar high levels of familiarity, yet attitude, congruity and compatibility answers showed that they did not have the same group of consumers. Further analysis for the actual study included a review of each brands set of responses and entering the data into PASW Frequency analysis and simple summ ary statistical tests were applied to each brand group individually as well as the combined sample data, to provide a descriptive sample profile and a summary of existing Facebook activit y. Subsequently, several comparisons of means analyses were executed to establish significant differences between the two brands regarding demographics, behaviors, and psychographics. Among the variables ass essed in these comparisons were independent variables of brand at titude, actual and ideal self brand congruity and brandFacebook compatibility. Next, bivariate correlation and multiple regression tests were performed to determine whether to accept or reject each h ypothesis, as well as to explore possible answers to th e research questions. From this synthesized information, the statistical tests were evaluated on whether the data

PAGE 40

40 provided sufficient a nswers regarding the correlation between each independent or interaction variable (explained in more detail in the following sections) and the dependent variable adoption behavior likelihood on Facebook. The results of this research are detailed in Chapter 4 Measures and Instrument The materials required for this study included a questionnaire (see Appendix B) a computer with Internet access, and an invitation to complete the questionnaire (se e A ppendix A ). The invitation was brief, describing the general context of the study and participation, as well as providing the link where the students could find and complete the questionnaire. As previously stated, this study included two sets of questionnaires: one set of questions regarding Starbucks and the same set of questions regarding Dunkin Donuts. This split was done to shorten the length of the questionnaire with the intention of increas ing completion rates. Furthermore, within the questionnaires, it was important to set background context with a list of questions regarding the participants behavior tendencies in their current Facebook usage. The q uestionnaire also target ed topics about attitudes congruity compatibility, and adoption behavior Some questions ask ed participants to role play in a given situation; other questions simply ask ed participants to what degree they agree or disagree with an item Each questi onnaire consisted of eight sections. These sections aimed to establish 1) participant qualification; 2) motivations for using Facebook ; 3) familiarity with, usage level of, and attitudes toward Fac ebook; 4) existi ng user habits on Facebook; 5) familiarity and attitudes toward the brand; 6) self brand congruity and brandFacebook compatibility; 7) adoption behavior likelihood of the brand; and 8) a demographic profile.

PAGE 41

41 More specifically, the survey questionnaire began with a series of qualification and backg round questions regarding Facebook. To qualify to complete the entire questionnaire, participants were required to be coffee drinkers, and also have a registered Facebook account. Question 3 asked for the date of initial membership with Facebook to discov er the possible level of expertise and give more meaning to activity on these sites on the basis of membership length. Questions 17 through 28 focused on the participants level of familiarity with, usage of, and attitudes toward Facebook. This section consisted of 5 point semantic differential questions based on questionnaires by Douglas (1999) and Kang (1999) Of these differential questions Questions 19 through 22 ascertained how active participants are on the site. Question 21 indicated participants approximate total activity time on Facebook per week. Although this question was separated into two variables for hours and minutes in data collection, the two were converted into a single hourly measurement and reported in hours. Q uestions 29 through 38 establish ed the users current habits on Facebook and operationalize the current Facebook behavior variable. In doing so, the results of the 10 questions for each participant were summed to create a current Facebook behavior index where a maximum score of 44 indicates the least behavior and familiarity and a minimum score of 10 indicates the most intense behavior practices. This measure helped to give proper significance to results of the research. For example, if a participant starts out at a high lev el of activity, it shows that this individual is willing to display information about himself. The same idea applies to the other extreme.

PAGE 42

42 Independent V ariable Two official independent variables were established in this study. These were 1) self brand cong ruity (H ypothesis 1) where both actual and ideal self brand congruity were observed, and 2) brandFacebook compatibility (H ypothesis 2). However, an added interest was observed in brand attitude as a corroborating independent variable. This observation was implemented to confirm the theoretical progression from brand attitude to self brand congruity and finally, adoption behavior as laid out by previous research. B rand attitude and differences between brand groups were explor ed for a preliminary assessment of the appropriateness of measures with alignment to previous research and an assessment of different brands influence s o n self brand congruity and brandFacebook compatibility. Confirming a lack of significant differences between brand groups lend s more justification to examining the variables of self brand congruity and brandFacebook compatibility outlined for H ypothesis 1 and H ypothesis 2. On the other hand, confirmed significant differences between brand groups would in dicate a need to keep brand influence in mind when considering the results of these tests. Furthermore, to rule out significant differences between brands would mean that both brand samples c ould be evaluated together as one combined sample. This would add to the overall sample size and reliability of the research. T o give further depth to the evaluation of the independent brand groups familiarity with the brand was assessed briefly, and brand attitude was considered to a further extent To establish fami liarity with the brand and measure brand attitude, the chosen brands Starbucks and Dunkin Donuts, were introduced in each parallel questionnaire by asking semantic diff erential questions (Q uestions 39 through 43 ). Familiarity was addressed with a single semantic differential on a 5 point scale from 1 ( very familiar )

PAGE 43

43 to 5 ( very unfamiliar ). Furthermore, attitude toward the brand was also measured via four semantic differential questions (40 through 43) rating the following four qualities for the brand: 1) good /bad, 2) pleasant /unpleasant 3) favorable/unfavorable, and 4) likab le/unlikable T hese questions were also adopted from questionnaires by Papacharissi and Rubin (2000) Douglas (1999) Kang (1999) and Parker (2005) This method of evaluatio n is commonly used to measure brand attitude (MacKenzie, Lutz, & Park, 1989) T he results of the four questions were averaged to create an index for operationalization of the brand attitude variable with a reliable Cronbachs alpha calculated in previous r esearch as .83 Next, the independent variable of self brand congruity was addressed in the latter half of the questionnaire. This section linked the individual, the environment, and the brands by asking questions regarding the participants image percept ions and level of congruity with the brands within the context of Facebook. The initial questions for this section (questions 44 through 48 ) were adopted from the global self congruity scales method proposed by Sirgy (1997) and further executed in Parkers self congruity research (2005) This method first determined brand user imagery descriptors and subsequently measured self brand congruity based on the descriptors cataloged. The respondent was first asked to provide two adjectives to describe the typical user of the brand in question. Next, the respondent was asked questions to establish the level of congruity he felt between his own actual and ideal self image s and the descriptors he just used to describe the typical user of the brand as follows: Take a moment to think about [Brand x]. Think about the kind of person w ho typically uses the brand. Imagine this person in your mind and then d escribe this person using one or two personal adjectives to describe the typical u ser of the brand. Now indicate your agreement or disagreement to

PAGE 44

44 the following statements: The typical user of [Brand x] is consistent with how I see myself (actual self congruity); The typical user of [Brand x] is consistent with how I like to s ee myself (ideal self congruity); The [Brand x] brand is compatible with Facebook (brandFacebook compatibility). As labeled above, the first of these questions is intended to measure actual self congruity, while the second question is intended to measure ideal self congruity. The congruity indicator scores collected from both of these questions were averaged to create a solitary self brand congruity index to simplify analysis and discussion, provided no significant difference occurred between the actual and ideal congruity responses If significant differences occurred, howev er, actual and ideal congruity were treated separately as individual indicators of self brand congruity This method measures self brand congruity directly, as opposed to performing separate evaluations of self image and brand (or product user) image (Sirgy et al., 1997) The method proposed by Sirgy et al. (1997) also took a holistic and global perspective on the measurement by allowing participant s to conjure up their own perception descriptors for typical brand users instead of presenting them with predetermined image perceptions to evaluate congruity (Sirgy et al., 1997) High or low self brand congruity was determined by the level to which the p articipant indicated he perceived a match (or mismatch) between his own self image (actual and ideal) and the perceived image of the typical user of the brand. In this study, high self brand congruit y was indicated with scores closer to 1, while scores clo ser to 5 indicated low self brand congruity. BrandFacebook compatibility, another independent variable, was subsequently measured as part of this set by asking the participant to indicate his level of agreement or disagreement that the brand is compatible with Facebook. This measurement used a solitary 5 point Likert scale ranging from 1 ( strongly agree ) to 5 ( strongly

PAGE 45

45 disagree ). The identical questions were asked for both Starbucks and Dunkin Donuts on their respective questionnaires for each of these independent variables. Dependent V ariable The dependent variable in this study was brand adoption behavior likelihood on a Facebook users posting activity This variable was examined in the final portion of the questionnaire with two questions Questions 49 and 5 0 asked the respondent to role play in various Facebook situations. First, in Q uestion 49, the respondent was simply asked whether he would add the brand to his profile to which he must answer his level of agreement on a 5 point scale which was coded from 1 ( n o, d efinitely n ot ) to 5 (y es, definitely ). Thus, a higher number indicated a higher likelihood of adoption behavior. Since research is limited on specific posting behavior intention this first dependent variable question was based on measures of intended voting behavior. For example, the following question was utilized to determine candidate preference in a study that examined the impact of negative political television commercials, If you were voting in this election, a fter seeing this commercial, would you vote for the sponsoring candidate? Respondents were then asked to indicate their selection on a scale ranging from no definitely not ( 3) to yes, definitely (+3) (Tinkham & Weaver Lariscy, 1993) As reinforcemen t to this measurement, the next question asked about more detailed posting behavior decisions. A list of 18 possible brand adoption behaviors with positive neutral, and negative brand attitudes were presented to the respondent, along with an Other opt ion Ten of these items were denoted as positive actions regarding the brand, five were considered neutral, and three were considered to be negative actions. Neutral and negative actions were included to provide a more well rounded

PAGE 46

46 pool of possible actions reflective of typical Facebook behavior. This list of 18 behaviors w as chosen because these behaviors provided a possible range typical of Facebook users, which gave the opportunity for a user to publicly express or display a brand in his Facebook activity. The participant was asked to check all the actions he was likely to take. Since this study aim ed to identify positive effects on behavior, the 10 positive action items were the focus for data analysis. To operationalize this dependent variable, a value of 1 was assigned if the participant indicated he would likely take that action, and a value of 0 was assigned if the action was not selected. Responses to the 10 positive items were tallied to create an aggregate positive action scor e. This aggregate score from 0 to 10 was added to the score from the previous question to create a total adoption behavior likelihood index. This index had possible scores from 1 to 15 as a result. In this case, the higher the adoption behavior lik elihood index score, the more likely a person was to take action regarding the brand in his Facebook activity. Question 50 was originally intended to modif y the model developed by Curry (2004) which measured decisionmaking bases in which each answer was assigned a positive, negative, or zero value. These values were aggregated to create four composite scores for each of the items measured in Currys research (Curry, 2004) This aggregated scoring was adapted to the current study with a few changes. The m ain modification made to this model was that essentially one item was examined in this section of the questionnaire and multiple response options were not mutually exclusive. Thus, answer options were increased and one composite score was created as a result of this modification. This model was chosen because users could perform more than one of the proposed behaviors simultaneously any of which could potentially

PAGE 47

47 express varying attitudes toward the indicated brand on a positive negative continuum As with Currys model, positive and negative aggregate scoring could be applied in future research to determine a correlation between negative brand attitude, self brand congruity and behaviors with negative attitudes, or to observe inaction tendencies as opposed to posting with positive or negative attitudes toward a brand. Motivation I nteraction V ariable s The interaction variables in this research explored the interac ting effect s of various motivatio nal reasons for using Facebook. A set of 13 qu estions addressed the four main motivations set forth by Papacharissi and Rubin (2000) These motivations were information (Q uestions 4 through 6) convenience (Q uestions 7 through 9) entertainment (Q uestions 10 through 13) and social interaction (Q uesti ons 14 through 16) Each of the four motivations had three questions dedicated to its measurement except for entertainment which included four questions in the set. The participant was asked to indicate his level of agreement with these 13 statements on a 5 point scale from 1 ( strongly agree ) to 5 ( strongly dis agree ) Results for each question within a motivation set were averaged to create an index score for that motivation. As a result, lower motivation indexes indicated that a participant was more motivated by that usage purpose. This model indicated sufficient reliability with a collective Cronbachs alpha score of .78 (Ko, Cho, & Roberts, 2005) M otivational indices were t hen multiplied by the self brand congruity indices to create an interaction variable for each motivation. Self brand congruity indices individual motivation indices, and interaction variables were all entered into multiple regression analyses for each of the four motivations examined in H ypothesis 3, H ypothesis 4, R esearch Q uestion 1, and R esearch Q uestion 2

PAGE 48

48 CHAPTER 4 RESULTS Data Analysis Sample Profile This study recruited 151 participants for the online survey. However, due to the nature of the study a nd the qualifying measures required to produce valid data, 77 respondents were eliminated from the final analysis of results Respondents were removed for two reasons : 1) respondents voluntary incompletion of the survey and 2) respondents not drinking co ffee and /or not hav ing a registered Facebook account. Therefore, many respondents were removed because they did not fit one or both of these qualifications, and thus were unable to answer the questions relevant to the core purpose of this study. Taking these eliminations into account, the final valid sample i ncludes 74 respondents. Of these 74 participants, 41 took the questionnaire with questions regarding Starbucks and 33 took a questionnaire with questions regarding Dunkin Donuts. Table 41 displays the summary description of the Starbucks, Dunkin Donu ts and total sample. The Starbucks and Dunkin Donuts groups were not significantly different (Table 4 1). Mostly female, participants ranged in age fro m 18 to 41 with the majority 18 to 24. These statistics are in alignment with Facebooks original target market a ge range of 18 to 24. Just over 80% of the participants were in the process of earning an undergraduate degree or had already completed a degree. More than half of the respondents were Caucasian.

PAGE 49

49 Facebook Tendencies/Activity Means for familiarity acceptability and likelihood to visit Facebook were high for both groups (Table 4 1), with no significant difference between groups. Likewise, with no difference between groups, 85.2% of the respondents use Facebook at least on a daily basis (once per day plus multiple times per day). Participants a ctivity time on Facebook averaged 6.09 hours per week. Participants level of public expression on Facebook was measured for direct (Table 42) and indirect expression of information (Table 4 3 ). The direct expression informational categories typically completed on Facebook included Basic Information, Personal Information, Contact Information, and Educational and Work Information. Among these direct expression behaviors, t he most co mpleted and detailed section was the Basic Information section with 25% of the respondents stating it to be fully completed, whereas respondents typically indicated they provide very limited information in the Personal, Contact and Work and Education Sections. This would imply a general sense of privacy and inaction with respect to directly providing descriptive details about themselves. This also supports the tendency for members to primarily use SNS s to maintain current relationships (with those who would likely already have such i nformation such as contact, work, education and personal information) rather than to search out new relationships (Boyd & Ellison, 2007) The indirect expression of information was represented by a display of associations with other items that they either searched for or were prompted to add by other Facebook users. These items included Groups, Fan items, Bumper Stickers, Gifts, and Other items. Groups came out as the most common behavior by far for each of the three samples with 91.9% of the combined sample noting that they had elected to

PAGE 50

50 become a member of at least a few groups if not many groups. Gifts sent fro m other Facebook users, Other Items an d Fan Items also presented a majority of users who reported they showed a few of these items on their profile. Bumper stickers found by the participant s themselves exhibited the least common behavior with a majority choosing to display none in that category. Overall Facebook behavior was calculated by summing each respondents current profile behavior with a maximum possible score of 44 (indicative of lesser activity) and a low possible score of 1 0 (indicative of greater activity) (Table 43 ). This me asurement resulted in total current activity scores between 23.0 and 24.0 for both Starbucks and Dunkin Donuts individually, as well as for the combined sample. On the scale in this study from most active (10) to least active (44) these scores fell on the more active side of the median, demonstrating moderate to high behavioral tendencies for each group. N o significant differences occurred between groups for total current Facebook behavior or any of the individual behaviors Brand Attitudes, Congruity and Compatibility Consistent with previous use of the brand attitude measures and suggested acceptable levels, between .7 and 1.0 (Davis, 1997) the items used for the b rand a ttitude i ndex produced a Cronbachs alpha of .95 overall, 92 for the Starbucks group and .97 for the Dunkin Donuts group (Table 4 4 ). While the data showed no significant difference of familiarity between the brands, the attitude toward Starbucks w as significantly more positive than the attitude toward Dunkin Do nuts. S elf brand congruity components and brand Facebook compatibility also returned significant differences between Starbucks and Dunkin Donuts (Table 44).

PAGE 51

51 Although no significant differences occurred between actual and ideal self brand congruity rating s for Dunkin Donuts, Starbucks s actual self brand congruity ratings were significantly greater than Starbucks s ideal self brand congruity ratings. Since at least one of the two observed brands revealed significant differences between these self brand congruity components, both actual and ideal self brand congruity measures will be used as separate variables in each sample group for the remainder of this studys statistical analysis. In addition to significant differences between actual and ideal self br and congruity, significant differences were also present between groups. Starbuc ks s respondents rated Starbucks as significantly more congruent with their actual and ideal selves than Dunkin Donuts respondents rated their actual and ideal selves. Lastly Starbucks wa s perceived to be significantly more compatible with Facebook than Dunkin Donuts Since participants also had significantly more positive attitudes toward Starbucks than Dunkin Donuts, this could be indicative that if a person feels negativ e associations toward a brand, he feel s more strongly that the brand is not appropriate for SNS discussion or favorable in making positive social impressions and vice versa. Independent, Dependent, and Interaction Variables Table 4 5 presents a summary of this studys key variables, means and reliability measures. F or multiple item indicators, Cronbach alphas were reported, and the Pearson correlation coefficient was reported for the dual components (actual and ideal) of self brand c ongruity Compatibility was a single item measure. While most of the alphas met the acceptable minimum (.70) (Davis, 1997) s everal did not (Convenience and Social Interac tion). However, given the sample sizes link to unstable correlations, it was decided that it was better to include the scales rather than delete them keeping

PAGE 52

52 the low alphas in mind when the data were analyzed. Significant differences between brands were observed for each of the independent variables and for one motivational variable (Information). The other three motivations showed no significant difference between brands. Dependent Variable: Adoptive Behavior The dependent variable, adoption behavior likelihood consisted of two measures. The first was a single item scale asking the respondent to indicate if he would either visually or verbally add the brand to his Facebook ( 1 = n o, d efinitely n ot to 5 = yes, d efinitely) ( see Likely of Adding in Table 4 7 ). The second measu re, which was added to the first measure to create the calculated dependent variable, consisted of the sum of all the positive actions a participant indicated he was likely to take regarding the brand on Facebook. These two behavioral measures were significantly and moderately strongly correlated for the combi ned sample and for each brand individually (Table 46) indicating these were appropriate measures Table 47 shows descriptive statistics and frequencies for each of the two measures used to calculate the dependent variable, adoption behavior likelihood. Total positive actions were significantly higher for Starbucks than for Dunkin Donuts In a more detailed look at each of the positive actions two action items displayed significantly higher results for Starbucks than for Dunkin Donuts These action items included sendi ng a brand logo gift to someone and adding a bumper sticker of the brand logo that was sent by another user The three actions most frequently indicated for the combined group were to accept and display a brand logo gift that someone sent you, write/post/display something positive about a brand in your profile or status, and become a fan of the

PAGE 53

53 brand on Facebook. These items also appear within the most frequent responses for both the Starbucks and Dunkin Donuts sample groups. The action least frequently indicated was to actively search out and join a group with a positive attitude toward the brand. This item was followed closely by the act of searching for the brands Facebook profile and more information regarding the brand. The results in Table 47 sho w that participants were least likely to fulfill action items requiring a participant to actively go out of his way to seek out the brand or media feature. Hypothesis Testing This study explored possible relationships among a number of variables within the concepts of consumer behavior, social psychology, brand management and new media communications. The constructs at the heart of this study which have been briefly explored thus far included self brand congruity brandSNS compatibility, and the four SN S motivation variables of entert ainment, information, social interaction, and convenience For the scope of this study, the best way to uncover answers to the hypotheses between these variables was by executing regression analyses to determine relationshi ps. Hypothesis 1: Self brand congruity (both actual and ideal) will be positively related to the likelihood of SNS brand adoptive behavior. First, a bivariate correlation test was performed examining indexes for actual and ideal self brand congruity current Facebook behavior, brand attitude, and adoption behavior likelihood ( Table 48 ) Adoption Behavior Likelihood is inversely coded from the remaining variables N egative correlation values are therefore indicative of positive relationships.

PAGE 54

54 This b ivariate correlation examination highlighted a few key observations. First, current Facebook behavior showed no significant correl ation, and thus no relationship with either self brand congruity or brand attitude. Furthermore, though a significant correlat ion was observed between current behavior and adoption behavior likelihood for the combined sample evidence showed only a weak positive relationship (R = 27). These apparent discrepancies suggest that positive attitudes and high ly perceived self brand congruity may be highly correlated with likely adoption behaviors indicate d by participants, but this in turn, may not necessaril y translate strongly into fulfilled behavior. Second, brand at titude displayed a significant moderate relationship with actual self brand congruity, ideal self brand congruity and adoption behavior likelihood for all three sample groups This result further confirms previous research and exemplifies these variables as good measures. Third since brand attitude and adoption behavior likelihood had a significantly moderate relationship, it reasonably follows that actual and ideal self brand congruity would also illustrate a significant moderate relationship with the dependent variable. This was in fact the case for the combined sample and Starbucks. Dunkin Donuts, however, showed only ideal self brand congruity to have a significant moderate relationship with adoption behavior likelihood. Th e important overall take away from these observations was that as brand atti tude and both actual and ideal self brand congruity increased positively, adoption behavior likelihood also increased revealing a significant and positive moderate relationship With these confirmations in mind, H ypothesis 1 further examined the component s of self brand congruity using bivariate correlations (Table 4 9 ). For the purpose of this

PAGE 55

55 test, actual and ideal self brand congruity measures were recoded such that 1 = s trongly d isagree and 5 = s trongly a gree, in order to have positive correlations depict a positive relationship and vice versa. Also, since preliminary statistical tests showed significant differences between groups for both actual and ideal self brand congruity bivariate correlation tests were run for both the Starbucks and Dunkin Donuts samples in addition to the combined sample. Three separate correlations were examined with regard to both actual and ideal self brand congruity The first set looked at participants intent to add the brand to their Facebook pr ofile either verbally or visually. This measurement showed significant moderately positive correlations with ideal self brand congruity for Starbucks, Dunkin Donuts and the combined sample. B ut for actual self brand congruity this was only the case for t he Starbucks and combined sample. The second set observed the correlation with the sum of positive actions which participants indicated they would likely take Both actual and ideal self brand congruity revealed significant, s lightly weaker moderate correlations than in the previous observation set for both Starbucks and the combined sample. However Dunkin Donuts was not significant at the p < .05 level for either actual or ideal self brand congruity this time The third set calculated dependent variabl e for this study adoption behavior likelihood was compared with the self brand congruity components As previously stated, adoption behavior likelihood was calculated by adding the score of the user s intent to add the brand to his Facebook activity (min = 1, max = 5) (Q uestion 49 of the questionnaire) and the sum of positive actions (min = 0, max = 10) which the participant

PAGE 56

56 indicated he would take regarding the brand on Facebook ( Q uestion 50 of the questionnaire) The bivariate correlation results were si milar to the first set regarding intent to add the brand to Facebook activity. Both actual and ideal self brand congruity correlations with adoption behavior likelihood were significantly positive with moderate strength for both Starbucks and the combined sample. Dunkin Donuts however, returned ideal self brand congruity as significantly and moderately correlated with ideal self brand congruity, but not with actual self brand congruity. These moderately positive correlations indicate that as a participant perceives increased self brand congruity (both actual and ideal), his adopti on behavior likelihood also increases. Starbuck s s correlation strengths were consistently greater for actual self brand congruity, whereas Dunkin Donuts showed the opposite to be true, though only two of the Dunkin Donuts correlations were significant. Overall, t his test showed that both actual and ideal self brand congruity were positively related to adoption behavior likelihood Hypothesis 1 wa s accepted as a result Hypothesi s 2: Brand SNS compatibility will be positively related to the likelihood of SNS brand adoption behavior. Just as in the analysis for self brand congruity in Hypothesis 1, brand Facebook compatibility for Hypothesis 2 was also recoded such that increased compatibility was associated with increasing numerical values ( 1 = s trongly d isagree and 5 = s trongly a gree ) to keep positive correlations depicting a positive relationship. Hypothesis 2 was explored with the same process as the latter part of Hyp othesis 1 with bivariate correlations As with Hypothesis 1, comparisons with brandFacebook compatibility were observed with both intent to add the brand to ones Facebook profile and positive action items indicated by the participant that they would likely take, as well

PAGE 57

57 as the calculated adoption behavior likelihood (Table 410). Since t tests showed significant differences between groups for brandFacebook compatibility, each brand was observed separately in addition to the combined sample. Observations of the intent to add the brand to their Facebook activity or profile showed a lack of significance for weakly correlated Starbucks, Dunkin Donuts, and combined samples. These same results were similarly repeated showing no significance for the sum of posi tive adoptive actions and also for adoption behavior likelihood. Thus, compatibility did not prove to be correlated, positively or negatively, with brand adoption behavior likelihood, which denoted that a relationship between these two variables was not pr esent Unlike self brand congruity results in Hypothesis 1, brandFacebook compatibility did not demonstrate a significant positive relationship with adoption behavior likelihood, resulting in the rejection of Hypothesis 2. Interaction Effect of SNS Motiva tions The secondary aspect explored in this study was the interaction effect of four usage motivations for SNSs including entertainment, information, social interaction, and convenience The dependent variable for the following two hypotheses and two research questions was the calculated overall ado ption behavior likelihood addressed previously in Hypotheses 1 and 2. Since the independent and interaction variables we re inversely coded from the dependent variable, a negative coefficient indicat ed a positive relationship, and vice versa, for each of the four motivational effects explored in H ypothesis 3, H ypothesis 4, R esearch Q uestion 1, and R esearch Q uestion 2. D ifferences between Starbuc ks and Dunkin Donuts s samples for the individual entertainment, social interaction, and convenience motivations were not significant As a result, a greater focus was placed on the combined sample for these examinations than

PAGE 58

58 for the information motivation T he differences between groups for the information motivation were significant meaning that a greater weight needed to be assigned to each brand individually, especially for the interaction with self brand congruity measures Since significant differences occurred between groups for both actual and ideal self brand congruity brand groups w ere still given individual attention for not only the information motivation, but also the entertainment, social interaction, and convenience motivations. Hypothesis 3: An entertainment SNS motivation will positively interact with both actual and ideal self brand congruity in predicting SNS adoption behavior likelihood. H ypothesis 3 addressed interaction effects between an entertainment motivation and self brand congr uity measures with a multiple regression model ( Table 4 11) The independent variables for this analysis were actual and ideal self brand congruity entertainment motivation, and the interaction effect between the entertainment motivation and both actual a nd ideal self brand congruity The combined sample equation was statistically significant [ F( 5,68) = 4.06*, *p < .01]. This model produced an R value of .48 indicating a moderate correlation between actual and ideal self brand congruity entertainment motivation, both their interaction effect s, and the dependent variable, adoption behavi or likelihood. R squared was .23, meaning that 23 % of variance in these five variables is explained by adoption behavior likelihood. Likewise, the Starbucks equation was also statistically significant [ F( 5 3 5) = 5.12*, *p < .01 ] with a moderate correlation of .65 and 42% of variance explained by the dependent variable. Since both of these equations were significant, a linear relationship occurred between the variables for each sample group and these two equations can be projected onto the population. In contrast to the Starbucks and

PAGE 59

59 combined samples, the Dunkin Donuts equation was not statistically significant and thus, no linear relationship wa s present for this sample. Although both the Starbucks and combined sample equations proved significant, independently the only significant relationship with adoption behavior likelihood was for entertainment motivation for the Starbucks sample (t = 2.96, p < .05 ). However, this means that f or each sample group both the individual self brand congruity variables and the interaction effects between entertainment motivation and actual and ideal self brand congruity were not significant T hus the two interactions between the entertainment motivation and both actual and ideal congruity have no unique contribution to predicting a Y value for the regression equation. As a result, Hypothesis 3 is rejected. Hypothesis 4: An information SNS Motivation will p ositively interact with both actual and ideal self brand congruity in predicting SNS adoption behavior likelihood. The same multiple regression tests from H ypothesis 3 were applied in the exploration of Hypothesis 4 with entertainment motivation replaced with information motivation. The independent variables for this analysis were actual and ideal self brand congruity, information motivation, and the two interaction effects between information motivation and the two self brand congruity components (Table 412 ). Significant differences between brands for both the individual information motivation variable and the self brand congruity measures provided reason to carefully consider each brand on an individual basis. For this hypothesis, Dunkin Donuts returned a regression equation that was not significant. In addition, this examination produced no significant individual relationships with adoption behavior likelihood for the Dunkin Donuts sample. However, as with the

PAGE 60

60 previous motivation, both Starbucks [ F(5,3 5) = 7.66*, *p <.01] and the combined sample [ F(5,68) = 6.13*, *p <.01] produced significant linear regression equations. Thus, linear relationships are present for these sample groups and these regression equations can be projected onto the population. St arbucks showed a strong correlation (R = .72) between the inputted self brand congruity, motivation, and interaction variables, and the dependent variable, with 52% of variance explained by adoption behavior likelihood. Likewise, the combined sample reveal ed a moderate correlation between these variables (R = .56), explaining 31% of the variance present. The individual information motivation variable was significant ly related to adoption behavior likelihood for both the combined sample (t = 3.12, p < .01) and the Starbucks sample (t = 3.04, p < .01). In fact, information motivation was the only variable with a significant relationship with the dependent variable for the combined sample. T he negative coefficients for the information variable for these sample groups indicate a positive effect on adoption behavior likelihood. Furthermore, while Dunkin Donuts produced no significant individual relationships Starbucks returned actual self brand congruity (t = 2.87, p < .05) and the interaction variable between actual congruity and information motivation (t = 2.18, p < .05) to also have significant relationships with adoption behavior likelihood. Although not significant for the combined sample, these variables were the most important variables for the m odel with Beta values of 1.43 and 1.35, respectively. Partially consistent with predictions for these variables, the negative coefficient for actual congruity implied that actual self brand congruity is positively related to adoption behavior likelihood.

PAGE 61

61 However, the positive coefficient for the interaction between actual congruity and information motivation indicated a negative effect. The other half of the self brand congruity measures (ideal congrui ty) showed no significan t correlations with adoption behavior likelihood individually nor by interacting with information motivation. Thus, these variables had no unique contribution to predicting a Y value for the models regression equation. As a result, a posi tive information motivation interaction with bot h actual and ideal self brand congruity cannot be statistically confirmed and H ypothesis 4 is rejected Supplementary to this, the differences between Starbucks and Dunkin Donuts results signify that the effect of information motivation and the interacti on effect between information motivation and self brand congruity vary per brand, so each brand needs to be considered on an individual basis in future studies. Research Questions The research questions that naturally follow these hypotheses are simply a continuation of the interaction effects of the final two SNS usage motivations, social interaction, and convenience Since these variables were expected to have negative impacts on behavior toward a brand, they were posed as research questions to be explored not to provide definitive statistical answers. However, multiple regression analyses were performed for both the social interaction motivation and convenience motivation in the same manner as in H ypothesis 3 and H ypothesis 4, but with expected negative influences rather than positive influences. Research Question 1: Do social interaction SNS motivations negatively interact with actual and ideal self brand congruity in predicting SNS brand adoption behavior likelihood?

PAGE 62

62 Research Question 1 en tered actual and ideal self brand congruity, social interaction motivation, and the two interactions between each self brand congruity component and social interaction motivation to have po tential significant relationships with adoption behavior likelihood (Table 413). Social interaction motivations were expected to negatively impact the effect of both actual and ideal self brand congruity on adoption behavior likelihood. Significant differences between brands for the self brand congruity measures provided reason to carefully consider each brand on an individual basis. As with the combined samples of H ypothesis 3 and H ypothesis 4 the combined sample returned a significant linear regression equation [ F(5,68) = 5.28*, *p <.01] but no individual significant individual relationships meaning that they had no unique contribution to predicting a Y value for the regression equation. However, Starbucks and Dunkin Donuts showed varying results. Starbucks returned both a significant regression equation [ F(5,35) = 9 .29*, *p < .01], as well as individual significant correlations between independent variables and adoption behavior likelihood. The regression equation indicated a strong correlation between the independent and dependent variables (R = .76), accounting for 57% of the variance present. The s ignificant relationships with adoption behavior likelihood for this sample were with actual self brand congruity (t = 3.39, p < .01), the independent social interaction motivation (t = 3.66, p < .01), and the interactio n between social interaction motivation and actual congruity (t = 2.82, p < .01). Consistent with expectations, t he positive coefficient for the interaction between ideal self brand congruity and social interaction motivation indicates a negative influence on adoption behavior likelihood.

PAGE 63

63 Dunkin Donuts also returned significant individual variables, including both actual congruity (t = 2.11, p < .05), ideal congruity (t = 2.05, p < .05), and the interaction effect between social interaction motivation and actual congruity (t = 2.01, p < .05). The negative coefficient for the interaction between actual self brand congruity and social interaction motivation reveals a positive influence on adoption behavior likelihood, which con tradict s the predicted negati ve impact and contrasts the Starbucks result. However, the overall regression equation for the Dunkin Donuts sample was not significant. Thus no linear relationship existed between the inputted variables and adoption behavior likelihood and the equation cannot be projected onto the population. While the Starbucks and combined sample models were statistically significant, individual variables complicated this analysis. For both brands, social interaction did significantly interact ed with actual self bran d congruity. However, besides the fact that the directional influences of these two results contradicted each other, the second self brand congruity component, ideal congruity, revealed that it did not have a significant interaction effect with social interaction motivation for either brand. Consequently, the data do not confirm that social interaction motivation negatively interacts with both self brand congruity components in predicting adoption behavior likelihood However, the significant interaction ef fects between actual self brand congruity and social interaction motivation give reason to explore this avenue further. Actual self brand congruity may be a significant factor in this model, whether it is independently or interacting with social interaction motivation. Research Question 2: Do convenience SNS motivations negatively interact with actual and ideal self brand congruity in predicting SNS brand adoption behavior likelihood?

PAGE 64

64 The final motivation observed was convenience motivation and its interaction effect with actual and ideal self brand congruity on adoption behavior likeliness (Table 4 1 4 ). Variable effects were observed individually per brand due to significant differences observed between groups for self brand congruity measures. As w ith the previous motivations, the equations for both Starbucks [ F( 5 35) = 7.35*, *p < .01 ] and the combined sample [ F( 5 68) = 5.45 *, *p < .01 ] were statistically significant allowing the linear relationship to be projected onto the population T he regressi on equation for Dunkin Donuts was not significant and therefore cannot be projected onto the population. Starbucks showed a strong correlation (R = 72 ) accounting for 51% of variance. The combined sample followed wi th a moderate correlation (R = .54 ) with 2 9 % of variance accounted for by adoption behavior likelihood. A lthough Dunkin Donuts did not return any of the independent or interaction variables to be significant, Starbucks and the combined sample showed discrepancies on this matter S imilar to the information motivation results both Starbucks (t = 3.97, p < .01) and the combined sample (t = 2.81, p < .05) returned the independent convenience motivation as significantly correlated with adoption behavior likelihood. This was the only significa nt relationship for the combined sample, but t he data for the Starbucks sample illustrated that in addition to the independent convenience motivation, its interaction effect with ideal self brand congruity also had a significant impact on adoption behavior likelihood (t = 2.04, p < .05) Negative coefficients for both the combined sample and Starbucks independent convenience motivation variable implied a positive effect on adoption behavior

PAGE 65

65 likelihood contrary to expectations On the other hand, the interaction effect between convenience motivation and ideal congruity for the Starbucks sample displayed a positive coefficient, meaning that convenience motivation negatively interacted with ideal self brand congruity in predicting adoption behavior likel ihood. Despite the fact that this component of the interaction variable set was consistent with predictions for this research question, it was the only interaction effect for any of the sample groups that was significant The lack of significance for this motivation s interactions with actual cong ruity for Starbucks, and with both actual and ideal congruity for Dunkin Donuts and the combined sample, indicate that they have no unique contribution to predicting a Y value in the regression model equation. Hence, this multiple regression analysis revealed that while the convenience motivation may individually have a significant relationship with adoption behavior likelihood, no significant negative interaction effect between convenience motivation and self bra nd congruity components on adoption behavior likelihood was confirmed. This was true of all four motivational interaction variables examined in this study. Also similar to the previous three motivations examined, t h e differences between Starbucks and Dunk in Donuts s results indicate d that the effect of convenience motivation and its interaction effect with self brand congruity measures varied according to brand. A summary of the significant relationships with adoption behavior likelihood for each motivations multiple regression analysis made in H ypothesis 3, H ypothesis 4, R esearch Q uestion 1, and R esearch Q uestion 2 for the combined sample ( labeled as C.S. ), Starbucks ( labeled as S ), and Dunkin Donuts ( labeled as D.D. ) indicates that s ocial interaction had the highest number of significant relationships followed by

PAGE 66

66 information, convenience, and lastly, entertainment (Table 415) A look at each sample indicates that Starbucks returned the most variables significant ly r elated to adoption behavior likelihood, including all four independent motivation variables. Actual self brand congruity proved to have a significant relationship with adoption behavior likelihood more often than ideal self brand congruity Interaction eff ects showed this same trend. Limitations This study is limited by several fa ctors. First, relatively low variances accounted for by the models act as a noteworthy hindrance in this study, as do the low reliabilities of the individual motiva tion and compat ibility variables. Second, the sample includes students from the College of Journalism and Communication at the University of Florida. These students, being already clearly interested in communication, may be more interested and involved with SNSs and higher posting behavior than students in other disciplines. Third the issue of online privacy could have been an influential factor as indicated by many responses about current profile habits, but wa s too broad to explore in the scope of this study with due diligence. Fourth, due to the nature of the study where qualifying components had to be met to fulfill the questionnaire to completion, the resulting sample size was lower than preferred which lead to questionable reliability measures Fifth for the port ion of the sample that was extended on Facebook, a more reliable sample would have been beneficial from a more national and global audience. I f this were the case, bran ds chosen for observation would need to be represented in those selected geographical a reas. Sixth due to limiting the length of the questionnaire, brands were chosen to be restricted to one product category and participants were presented with only one of the

PAGE 67

67 two brands. A broader brand collection and a direct comparison between brands presented at once would have been helpful in achieving a broader perspective of branded behavior. The questionnaire itself included several limitations. In addition to the items mentioned previously, if it had been possible to return to the questionnaire and modify it, changes would have included discarding the coffee drinker qualification to increase sample size, since many noncoffee drinkers also have developed atti tudes towar d well known brands and their typical users. A modified questionnaire also would have included more questions to measure compatibility to obtain a reliability score for the variable. Finally, the question regarding posting behavior was limited by an unequal number of items for positive, negative, and neutral responses. As a result, only positive actions were sufficiently observed. F uture research observations regarding simple action versus inaction and positive action versus negative action could provide added valuable insight on this subject. Such research modifications would likely introduce new variables in the effect of brand on self brand congruity measures and adoptive SNS posting behavior that were beyond the scope and capabilities of this study

PAGE 68

68 Table 4 1 Sample p rofile s ummary s tatistics Starbucks (n=41) Dunkin Donuts (n=33) Total s ample (n=74) Sig. Mean S. D. Mean S. D. Mean S.D. Age 22.15 3.69 23.12 4.21 22.58 3.93 NS* Age g roup # % # % # % NS** 18 24 35 85.4 25 75.8 60 81.1 25 34 5 12.2 7 21.2 12 16.2 35+ 1 2.4 1 3.0 2 2.7 Gender # % # % # % NS** Male 15 36.6 11 33.3 26 35.1 Female 26 63.4 22 66.7 48 64.9 Education # % # % # % NS** High School 10 24.4 2 6.1 12 16.2 AA in progress 1 2.4 --1 1.4 AA --3 9.1 3 4.1 Bachelor in progress 22 53.7 17 51.5 39 52.7 Bachelor 5 12.2 5 15.2 10 13.5 Graduate 2 4.9 6 18.2 8 10.8 PhD in progress 1 2.4 --1 1.4 Ethnicity # % # % # % NS** Asian 1 2.4 2 6.1 3 4.1 Black/African A mer ican 4 9.8 4 12.1 8 10.8 Hispanic/L atino 9 22.0 3 9.1 12 16.2 White / Caucasian 26 63.4 24 72.7 50 67.6 Other 1 2.4 --1 1.4

PAGE 69

69 Table 4 1. Continued Starbucks (n=41) Dunkin Donuts (n=33) Total s ample (n=74) Sig. Mean S. D. Mean S. D. Mean S.D. Facebook f amiliarity a 1.51 .60 1.48 .83 1.50 .70 NS* Facebook a cceptability a 1.76 .70 194 .79 1.84 .74 NS* Facebook l ikely to v isit a 1.61 .89 1.30 .47 1.47 .74 NS* Facebook f requency # % # % # % NS** Multiple times/day 29 70.7 25 75.8 54 73.0 Once per day 4 9.8 5 15.2 9 12.2 Multiple times/week 4 9.8 3 9.1 7 9.5 Once per week 2 4.9 0 0.0 2 2.7 One per month 1 2.4 0 0.0 1 1.4 LT once per month. 1 2.4 0 0.0 1 1.4 Facebook current b ehavior # % # % # % NS** Browsing 15 36.6 10 30.3 25 33.8 Interacting 22 53.7 17 51.5 39 52.7 Adding profile info 4 9.8 6 18.2 10 13.5 Mean S. D. Mean S. D. Mean S.D. Facebook total activity t ime b 6.69 10.58 5.35 4.43 6.09 8.39 NS* *One way ANOVA **Chi square test aFacebook f amiliarity. Facebook acceptability and Facebook l ikely to v isit were measured on scale of 1 to 5 where 1 = v ery f amiliar, v ery acceptable, and definitely w ill v isit bFacebook total activity t ime is reported in hours, e.g, 6.69 = 6 hours and .69 of an hour (41.4 minutes)

PAGE 70

70 Table 4 2 Cu rrent Facebook p rofile b ehavior (d irect e xpression ) Starbucks Dunkin Donuts To tal s ample Sig. Facebook p rofile b asic # % # % # % NS Full 8 19.5 11 33.3 19 25.7 Almost f ull 15 36.6 12 36.4 27 36.5 Very l imited 16 39.0 10 30.3 26 35.1 None 2 4.9 --2 2.7 Facebook p rofile p ersonal # % # % # % NS Full 2 4.9 4 12.1 6 8.1 Almost f ull 12 29.3 7 21.2 19 25.7 Very l imited 23 56.1 21 63.6 44 59.5 None 4 9.8 1 3.0 5 6.8 Facebook p rofile contact i nfo # % # % # % NS Full --1 3.0 1 1.4 Almost f ull 8 19.5 7 21.2 15 20.3 Very l imited 27 65.9 21 63.6 48 64.9 None 6 14.6 4 12.1 10 13.5 Facebook profile education/w ork # % # % # % NS Full 8 19.5 8 24.2 16 21.6 Almost f ull 13 31.7 11 33.3 24 32.4 Very l imited 18 43.9 12 36.4 30 40.5 None 2 4.9 2 6.1 4 5.4

PAGE 71

71 Table 4 3 Current Facebook p rofile b ehavior (i ndirect e xpression ) Starbucks Dunkin Donuts Total s ample Sig. Facebook profile g roups # % # % # % NS Many 15 36.6 10 30.3 25 33.8 A f ew 23 56.1 20 60.6 43 58.1 None 3 7.3 3 9.1 6 8.1 Facebook profile fan i tems # % # % # % NS Many 4 9.8 5 15.2 9 12.2 A f ew 18 43.9 17 51.5 35 47.3 None 18 43.9 11 33.3 29 39.2 Not f amiliar 1 2.4 --1 1.4 Facebook profile bumpers from o thers # % # % # % NS Many 7 17.1 7 21.2 14 18.9 A f ew 18 43.9 8 24.2 26 35.1 None 15 36.6 18 54.5 33 44.6 Not f amiliar 1 2.4 --1 1.4 Facebook profile b umpers by s elf # % # % # % NS Many 2 4.9 3 9.1 5 6.8 A f ew 10 24.4 3 9.1 13 17.6 None 28 68.3 27 81.8 55 74.3 Not f amiliar 1 2.4 --1 1.4 Facebook p rofile g ifts from o thers # % # % # % NS Many 1 2.4 6 18.2 7 9.5 A f ew 29 70.7 9 27.3 38 51.4 None 11 26.8 18 54.5 29 39.2 Facebook p rofile o ther # % # % # % NS Many 2 4.9 3 9.1 5 6.8 A f ew 26 63.4 18 54.5 44 59.5 None 13 31.7 12 36.4 25 33.8 Mean S. D. Mean S. D. Mean S.D. Sig. Total current Facebook a ctivity a 23.83 4.33 23.30 3.95 23.60 4.14 NS a The sum of all Facebook actions from Tables 4 2 and 4 3 where max = 44 and min = 10

PAGE 72

72 Table 4 4 Brand attitudes, brand congruity, and c ompatibility Starbucks (n=41) Dunkin Donuts (n=33) Total s ample (n=74) Mean S. D. Mean S. D. Mean S.D. Sig. Brand f amiliarity 1.76 .97 2.09 .91 1.91 .95 NS Brand a ttitude i ndex a 1.98 .59 2.35 .80 2.15 .71 p<.05 Good b ad 2.00 .71 2.33 .96 2.15 .84 -Pleasant u npleasant 1.95 .74 2.39 .70 2.15 .75 -Favorable u nfavorable 2.20 .78 2.52 .94 2.34 .86 -Likeable u nlikeable 2.00 .71 2.42 .97 2.19 .85 -Alpha .92 .97 .95 -Actual self b rand c ongruity b 2.56 1.05 3.18 1.16 2.84 1.14 p<.05* Ideal self brand c ongruity c 2.76 1.11 3.30 1.21 3.00 1.18 p<.05** Pearson c orrelation .83 .81 .83 -Compatibility d 2.46 .67 2.82 .95 2.62 .82 p<.05 a Attitude toward Starbucks was si gnificantly more positive than a ttitude toward Dunkin Donuts (t = 2.29; df = 72; p < .05) bStarbucks actual s elf rating was signif icantly greater than Starbucks ideal s elf rating (t = 1.95, df = 40; p < .05) (paired t tes t) cDunkin Donuts actual s elf rating was not significantly differe nt from its ideal s elf rating (paired t test) *Actual self congruity was significantly greater for Starbucks than Dunkin Donuts (t = 2.39, df = 72; p < .05). **Ideal self congruity was significantly greater for Starbucks than Dunkin Donuts (t = 2.00, df = 72; p < .05). dCompatibility between Starbucks and Facebook was significantly higher than compatibility between Dunkin Donuts and Facebook (t = 1.88; df = 72; p < .05)

PAGE 73

73 Table 4 5 Summary s tatistics for independent and motivation v ariables Starbucks Dunkin Donuts Total s ample Sig. Independent v ariables Mean Reliability Mean Reliability Mean Reliability Brand a ttitude 1.98 2.35 2.15 p < .05 a Actual c ongruity 2.56 R = .83 3.18 R = .81 2.84 R = .83 p < .05 b Ideal c ongruity 2.76 3.30 3.00 p < .05 c Compatibility 2.46 NA 1 2.82 NA 1 2.62 NA 1 p < .05 d Motivation v ariables Mean Reliability Mean Reliability Mean Reliability Sig. Entertainment 1.90 1.89 1.90 NS Convenience 2.17 2.20 2.19 NS Social i nteraction 2.59 2.46 2.54 NS Information 3.63 3.27 3.47 p < .05 e a Attitude toward Starbucks was significantly more positive than a ttitude toward Dunkin Donuts (t = 2.29; df = 72; p < .05) b Actual self congruity was significantly greater for Starbucks than Dunkin Donuts (t = 2.39, df = 72; p < .05). c Ideal self congruity was significantly greater for Starbucks than Dunkin Donuts (t = 2.00, df = 72; p < .05). dCompatibility was measured by a single 15 item where 1 = Very Compatible. C ompatibility between Starbucks and Facebook was significantly higher than compat ibility between Dunkin Donuts and Facebook (t = 1.88; df = 72; p < .05) eInformation motivation was significantly higher for Dunkin Donuts participants than for Starbucks participants (t = 1.74; df = 72; p < .05) Tabl e 4 6 Dependent variable compone nt c orrelations Dependent v ariable Starbucks Dunkin Donuts Total s ample R Sig. R Sig. R Sig. Intent to add brand to Facebook with adoptive b ehavior .71 p < .05 .72 p < .05 .72 p < .05 Brand attitude with combined a dopt ion behavior l ikelihood .71 p < .05 .38 p < .05 .56 p < .05

PAGE 74

74 Table 4 7 Dependent variable m easures Starbucks Dunkin Donuts Total s ample Mean Alpha Mean Alpha Mean Alpha Sig. Likely of adding (single semantic scale: 5 = D efinitely ) 2.20 NA 1.82 NA 2.03 NA NS Positive adoptive b ehavior (n=10) # % # % # % Sig. Accept and display a brand logo gift that someone sent you 16 39.0 8 24.2 24 32.4 NS Write/post/display something positive about brand in profile or status 13 31.7 7 21.2 20 27.0 NS Become a fan of the brand on Facebook 9 22.0 9 22.0 18 24.3 NS Send a brand logo gift to someone 10 24.4 3 9.1 13 17.6 p<.05 Add/display Facebook bumper sticker of brand logo someone sent you 10 24.4 3 9.1 13 17.6 p<.05 Accept an invitation to join a group with positive attitude toward brand 8 19.5 4 12.1 12 16.2 NS Send someone else a bumper sticker of the brand logo 7 17.1 2 6.1 9 12.2 NS Add/display Facebook bumper sticker of brand logo you found 5 12.2 1 3.0 6 8.1 NS Search out brand Facebook profile and more info 4 9.8 1 3.0 5 6.8 NS Search out and join group with positive attitude toward the brand 2 4.9 1 3.0 3 4.1 NS Mean S.D. Mean S.D. Mean S.D. Sig. Total positive b ehaviors 2.05 2.29 1.18 1.69 1.66 2.08 p<.05* *Total positive behaviors regarding Starbucks were significantly higher than total positive behaviors regarding Dunkin Donuts (t = 1.87; df = 72; p < .05)

PAGE 75

75 Table 4 8 Correlations Self brand c ongruity Current b ehavior Brand a ttitude Adoption behavior l ikelihood 1 Actual Ideal Com bined sample Actual c ongruity 1.00 .83** .07 .58** .43** Ideal c ongruity .83** 1.00 .12 .57** .39** Current b ehavior .07 .12 1.00 .13 .27* Brand a ttitude .58** .57** .13 1.00 .56** Adoption b ehavior l ikelihood 1 .43** .39** .27* .56** 1.00 Starbucks Actual c ongruity 1.00 .83** .03 .60** .47** Ideal c ongruity .83** 1.00 .06 .48** .36* Current b ehavior .03 .06 1.00 .17 .28 Brand a ttitude .60** .48** .17 1.00 .71** Adoption b ehavior l ikelihood .47** .36* .28 .71** 1.00 Dunkin Donuts Actual c ongruity 1.00 .81** .25 .50** .29 Ideal c ongruity .81** 1.00 .24 .60** .37* Current b ehavior .25 .24 1.00 .14 .32 Brand a ttitude .50** .60** .14 1.00 .38* Adoption b ehavior l ikelihood .29 .37* .32 .38* 1.00 1 Negative correlation values indicate positive relationships due to inverse coding **. Correlation is significant at the 0.01 level (2tailed) *. Correlation is significant at the 0.05 level (2 tailed) Table 4 9 Self brand congruity c orrelations Hypothesis 1 Starbucks Dunkin Donuts Total s ample Actual s elf brand congruity and intent to add brand to Facebook .49** .33 .44** Ideal s elf brand congruity and intent to add b rand to Facebook .39* .37* .41** Actual s elf brand congruity and positive a ction i tems .42** .24 .38** Ideal s elf brand congruity and positive action i tems .31* .33 .35** Actual s elf brand congruity and a doption behavior l ikelihood .47** .29 .43** Ideal s elf brand congruity and adoption b ehavior l ikelihood .36* .37* .39** **. Correlation is significant at the 0.01 level (2 tailed) *. Correlation is significant at the 0.05 level (2 tailed)

PAGE 76

76 Table 4 10 Compatibility c orrelations Hypothesis 2 Starbucks Dunkin Donuts Total s ample R Sig. R Sig. R Sig. Compatibility and intent to add brand to Facebook .26 .10 .20 .27 .07 .57 Compatibility and positive action i tems .21 .19 .28 .12 .04 .76 Combined intention and adoption behavior l ikelihood .24 .13 .26 .14 .05 .68

PAGE 77

77 Table 411. Multiple regression for entertainment motivation variable (H3)

PAGE 78

78 Table 412. Multiple regression for information motivation variable (H4)

PAGE 79

79 Table 413. Multiple regression for social interaction motivation variable (RQ1)

PAGE 80

80 Table 414. Multiple regression for convenience motivation variable (RQ2)

PAGE 81

81 Table 4 15 Summary of multiple r egression s ignificant r elationships Motivations Self brand c ongruity Individual m otivation Interaction e ffect Actual Ideal Actual Ideal C.S. S D.D. C.S. S D.D. C.S. S D.D. C.S. S D.D. C.S. S D.D. Entertainment X Social i nteraction X X X X X X Information X X X X Convenience X X X

PAGE 82

82 CHAPTER 5 DISCUSSION AND CONCL USIONS Discussion The main purpose of this research was to determine if an opportunity occurs for greater customer relationship management and to clarify a few of the fundamental principles that can lead to successful relationships with consumers via SNS s. To do so, this study examined the relationship of actual and ideal self brand congruity as well as brandSNS compatibility with consumer likelihood of publicly associating themselves with a brand on Facebook. In conjunction with these variables, motivational interactions with each self brand congruity measure were examined to attempt to further explain what impacts SNS posting behavior. The two brands utilized to exemplify the se potential relati onships were Starbucks and Dunkin Donuts within the SNS context of Facebook Two theories were applied in shaping this research. In previous research, self congruity theory has been used to explore brand attitude. Since brand attitude is often used in predicting branded behavior, the linkages between these three constructs led to an investigation in the present study of self brand congruity s relationships with branded behavior within the SNS environment. This investigation was fulfilled by observing both actual and ideal self brand congruity perceived by participants and then determining the presence or absence of relationships with brand adoption behavior likelihood on Facebook. In the process, brand attitude was also measured as a source of extra security. This examination presented results in alignment with previous research findings demonstrating that with an increase in brand attitude, both actual and ideal self brand congruity also increase.

PAGE 83

83 Second, the uses and gratifications theory inspired an inspection of the interaction effects of SNS usage motiv ations on posting behavior. Interaction effects were studied for four motivation categories : entertainment, information, social interaction, and convenience These components were employed to observe each motivations interaction with each self brand congruity element in predicting adoption behavior likelihood. These theoretical foundations provided structure for advancing previously explained relationships in a complex and relatively young social media environment. The following theoretical implications detail the conclusions derived from the statistical tests completed in C hapter 4 Collectively they help cl arify which factors contribute d to brand adoption behavior likelihood, as well as give reasons for elimination of those that do not contribute to this behavior as originally expected. A discussion of the implications of these findings for marketers and advertisers in the future follows to provide guidance for the next steps to be taken regarding this topic. Implications This research was fueled by an inquiry into which factors contributed to SNS posting behavior which indicate s personal adoption of a bran d. The primary constructs observed in the present research were self brand congruity brandFacebook compatibility, and motivations for using SNS. B oth self brand congruity (actual and ideal) and brandFacebook compatibility were examined in Hypothesis 1 and Hypothesis 2, respectively using bivariate correlation analyses with regard to adoption behavior likelihood. The four motivational interaction effects were each assigned their own examination. Hypotheses 3 and 4 targeted the entertainment and informati on motivations respectively. Research Q uestions 1 and 2 addressed the predicted

PAGE 84

84 negative interaction effects of the social interaction and convenience motivations respectively. To achieve the purpose of this research, a combination of frequency analyses, descriptive summaries, t tests, bivariate correlations and multiple regression analyses were implemented. Statistical compilations o f these tests illustrate existent and nonexistent linear relationships between the variables A precautionary assessment of each sample group was performed to make sure the brand was not overlooked as an extraneous or spurious variable. Variables were evaluated for significant differences between the Starbucks sample and the Dunkin Donuts sample. These tests showed no significant differences for any of the demographic variables or variables addressing Facebook familiarity, attitudes, and activity. In addition, familiarity with Starbucks was not significantly different from familiarity with Dunkin Donuts which both averaged between familiar and very familiar in responses. Thus, both Starbucks and Dunkin Donuts participants mirrored each other in this sample demographically, but also in that both groups can be assumed to be equally familiar with both F acebook and their respective brand This established familiarity improved the reliability of the subsequent data. Moreover, the lack of significant differences for these variables indicated that Facebook activity and attitudes w ere not contingent on varying brands Current Facebook behavior indicated that participants demonstrated moderate to high levels of expression behavior overall. However, attitudes toward Starbucks were significantly more positive than those toward Dunkin Donuts. P revious research es tablishing self brand congruity as a predictor of brand attitudes wa s further validated by this studys high correlations between these constructs. Both actual and ideal self brand congruity were significant ly

PAGE 85

85 greater for Starbucks than for Dunkin Donuts In this case, the results illustrate that Starbucks participants perceived more congruity between both the ir actual and ideal selves and typical Starbucks users than Dunkin Donuts participants perceived between both the ir actual and ideal selves and the typical Dunkin Donuts user Within this Starbucks significantly greater sample, participants perceived significantly higher congruency between their actual selves and the typical user of Starbucks than between the typical user of Starbucks and how they w ould like to see them selves (ideal congruity). However, no similar significant differences occurred between actual and ideal self brand congruity for the Dunkin Donuts sample. Significant differences between groups signified as with attitudes toward dif ferent brands, that individual brands have unique levels of self brand congruity which applies to both actual and ideal congruity This finding suggest s that self brand congruity measures are more reliant on brand attitudes and perhaps brand personality, than originally considered In turn, brands that manage to achieve a high level of actual and ideal self brand congruity with their current or target consumers, and also achieve highly positive brand attitudes among these consumers could observe increased adoptive behavior and SNS wordof mouth benefits over brands that fail to achieve these goals. This implication is supported further by the results of Hypothesis 1 in the following section. The implication that follows is a need to investigate se lf brand congruity components on a per brand basis in future research. Also, based on the findings of this research, marketers should attempt to relate their typical user image with the actual and ideal self images of their current and target consumers for increased wordof mouth benefits from SNS activity.

PAGE 86

86 Similar to the outcomes of self brand congruity compatibility also varies between brands depicted by significant ly higher perceived Starbucks compatibility with Facebook than perceived Dunkin Donuts co mpatibility with Facebook. This difference could ultimately mean that some brands are considered more appropriate or acceptable to talk about and display openly on an SNS than others and that compatibility could be a significant factor in the execution of branded Facebook behaviors. Motivational interactions were also checked for significant differences between brands. Among the four motivation variables, entertainment, social i nteraction, and conveniences yielded no significant differences between brands. Thus, participants were assumed to be equally likely or unlikely to identify these SNS usage motivations for themselves regardless of the brands presented to them. Information was t he only motivation to exhibit a significantly higher information motivation for Dunkin Donuts than for Starbucks. Thus, some brands elicit greater information motivation for their SNS usage than other brands This could mean for future research purposes t hat those pursuing informational purposes are more likely to actively search for brands and carry out associated branded behavior. Lastly, the dependent variable also showed that adoptive behavior of Starbucks was significantly higher than adoption of Dunk in Donuts into SNS behavior Correlation measurement s showed significant strong correlations between the components used to compute the variable. S ignificant moderate and strong correlations were also revealed between the computed adoption behavior likeli hood and brand attitudes These correlation tests substantiated the dependent variable components as good measurement s, although the relationship between brand attitude and adoption behavior

PAGE 87

87 likelihood was significantly stronger for Starbucks than Dunkin Donuts. A key observation from frequency statistics of each individual proposed action was that participants were least likely to fulfill actions which required them to exert increased or initiated effort in completing the behavior. One inconsistency to n ote, however, was the correlations between current Facebook behavior reported and the indicated intent of adoption behavior. As stated previously, current Facebook behavior indicated moderate to high activity. Conversely, adoption behavior likelihood resul ts were comparatively much lower on their respective scale Although each brand individually showed no significant relationship, for the combined sample (disregarding brand) a significant negative and weak relationship occurred between these variables Th is relationship indicated that as current Facebook behavior increases, brand adoption behavior likelihood decreases. The result may suggest that indicated likely behavior does not strongly translate into actual fulfilled behavior. T he significant differences between groups for all of these studied variables are enough to justify separate regression examinations by brand, along with the combined sample, for each of the hy potheses and research questions The following sections will examine these cases in more detail. Hypothesis 1: Self brand congruity (both actual and ideal) will be positively related to the likelihood of SNS brand adoptive behavior. The goal of H ypothesis 1 was to determine if a positive relationship existed between both actual and ideal self brand congruity and adoption behavior likelihood. This prediction was supported in the bivariate regression test with significant moderate positive relationships between both actual and ideal self brand congruity and adoption

PAGE 88

88 behavior likelihood for both Starbucks and the combined brand sample Dunkin Donuts showed only a significant positive moderate relationship for ideal congruity and adoption behavior likelihood. R esults can therefore vary depending on brand, but overall, this hypothesis was accepted. H igher perceived similarity between how a Facebook user actually views himself and how he perceives the typical user of a brand indicates an increased likelihood that the user will positively adopt the brand to his profile behavior in some fashion. This positive relationship also applies with perceived similarity increases between how a Facebook user would like to view himself and how he perceives the typical user of a brand. These results are in alignm ent with previous research performed by Sirgy (1997) and Parker (2005) comparing ideal and actual self perceptions to brand images and personalities. The results also imply that attitudes are a key factor in the behavior of sharing brand associations on SN Ss The correlation results for this hypothesis analysis signify that adoption behavior likelihood is complementary with both actual and ideal self brand congruity cases, but not strong enough to denote complete independence in predicting adoption behavior likelihood. T he reasons for branded posting behavior thus extend beyond simple self brand congruity explanations, and suggest that other factors not studied in this research are involved in the execution of such behavior. However, with actual and ideal se lf brand congruity acting as two of the variables significant ly related to adoption behavior, brands would benefit via wordof mouth activity by aligning themselves with the self image perceptions of their target consumer market. Hypothesis 2: Brand SNS compatibility will be positively related to the likelihood of SNS brand adoption behavior.

PAGE 89

89 H ypothesis 2 attempted to account for the possibility that some brands are n o t considered relevant or appropriate to share on SNSs The prediction of a p ositive effect of brand Facebook compatibility on adoption behavior likelihood was rejected. Contrary to self brand congruity brandFacebook compatibility did not have a significant relationship with adoption behavior likelihood in this bivariate correlat ion model for either brand individually or the combined sample. Although significant differences occurred of perceived compatibility between the brands, bivariate correlation results suggest that regardless of preconceived image perceptions, compatibility between a brand and Facebook have no influence on decisions to carry out brand adoptive posting behaviors on SNSs. Speculation suggests that SNS users show a lack of concern for the appropriateness of their posting behavior an area where additional resear ch could provide more insight. A n added opportunity presents itself for further research linked with the self brand congruity implications of determining whether specific characteristics or personalities of brands have more impact on these variables than t he brands do as a unit Also this result could have been a product of observing only one brand category. Perhaps looking at results across brand categories would be more indicative of specific effects on behavior between brands, as denoted in the previous limitations section. The remainder of the statistical data analyses intended to identify interaction effects of four motivations for using an SNS. Hypotheses 3 and 4 target the predicted positive interaction effects of entertainment and information motiv ations respectively. Hypothesis 3: An entertainment SNS motivation will positively interact with both actual and ideal self brand congruity in predicting SNS adoption behavior likelihood. The multiple regression analysis for H ypothesis 3 presented the ind ividual entertainment motivation variable for Starbucks as the only variable significantly and

PAGE 90

90 positively related to adoption behavior likelihood out of all three sample groups. T he more a person is motivated by entertainment in their SNS usage, the more likely they are to adopt Starbucks into his SNS posting behavior. The Starbucks regression model showed that there was a significant moderate relationship between the independent v ariables and the dependent variable is projectable onto the population. E ntertainment motivation individually is a complementary influence on adoption behavior likelihood, but not strong enough to act independently. This inability to act independently cou ld mean that factors other than SNS usage motivations are concerned. It could also indicate that the entertainment motivation touches on a more complex motivation or set of motivations that could have stronger significant correlations with adoption behavior likelihood The possible involvement of other motivations could extend differently to varying brands. The differences in significant relationships between sample groups indicate that some brands experience higher adopti ve behavior than others when a pers on is motivated to use an SNS for entertainment purposes. This divergence could be due to differences in brand personalities and entertainment qualities attributed to a brand. Furthermore, neither actual self brand congruity nor ideal self brand congruity were significant ly related to ado ption behavior likelihood. Significant relationships were also absent among the interaction effects between both these self brand congruity variables and entertainment motivation. Thus, neither a positive n or a negative int eraction occurred with self brand congruity variables in predicting adopti on behavior likelihood by entertainment motivations It is therefore reasonable to conclude that an entertainment motivation has no impact via interaction with self brand congruity on the

PAGE 91

91 likelihood of adopt ing the brand in SNS posting behavior. This hypothesis was rejected as a result. The analysis results indicated that it made no difference if a person felt any kind of congruity with a brand, for a person to adopt Starbucks into his SNS posting behavior for entertainment purposes. Hypothesis 4 : An information SNS motivation will positively interact with both actual and ideal self brand congruity in predicting SNS adoption behavior likelihood. H ypothesis 4 performed an identical a nalysis as Hypothesis 3, but for the information motivation. The multiple regression tests for H ypothesis 4 returned some results that confirmed expectations, and some results that contradicted expectations. First, the regression model for Dunkin Donuts w as not significant. Significant relationships were not present for any of the individual variables inputted for this model either However, both the statistically significant Starbucks and combined sample models disclosed the individual information motivat ion variable to have a significant positive relationship with adoption behavior likelihood. T hus, t he more users are motivated to use an SNS for information purposes, the more likely they are to adopt a brand into their SNS posting behavior. Furthermore, i n addition to the individual information motivation variable, Starbucks also revealed actual self brand congruity and its interaction with inform ation motivation as significantly related to adoption behavior likelihood. The effect of actual congruity indiv idually was positive, meaning that as perceived congruity between a persons actual self and the typical user of Starbucks increased, so did his likelihood of adopting Starbucks into his SNS posting behavior. However, the interaction effect of actual self brand congruity and infor mation motivation was negative, contrary to expectations indicating that if a person using an SNS for information purposes also

PAGE 92

92 perceives high congruity between his actual selves and the typical Starbucks user, he is less likely t o adopt Starbucks into his SNS posting behavior. The data detailed that regardless of any other factors, those respondents who are motivated to use an SNS for information motivations are more likely to adopt a brand into their posting behavior. However, w hen information motivation interacts with actual self brand congruity, an SNS user becomes less likely to adopt a brand into his posting behavior. Many explanations could be offered for this directional discrepancy between the individual and interactional effects of information motivation. For example, Starbucks is a widely known brand. P erhaps when amplified by a sense of congruity with themselves, whom they already know rather well, users do n o t feel as strong a need to collect information regarding a br and they feel they already know fairly well, as opposed to a brand they are n o t as familiar with. This is substantiated by the significantly higher information motivation observed for Dunkin Donuts than for Starbucks. Moreover the positive influence of information motivation individually could be attributed to a rise in more interactive forum informationseeking, in which at least a minimal level of information is often required of the information seeker to retrieve the appropriate information. Subsequently, this informational response is likely from other peers within an SNS users network. As a result, it could be likely that this type of information seeking on SNSs plays more of a conversational role than traditional unidirecti onal information retrieval on the Internet Furthermore, although the actual congruity and its interaction effect with information motivation were significant for Starbucks, the remaining interaction effects for each sample were not significant ly related t o adoption behavior likelihood. Therefore,

PAGE 93

93 an information motivation to use an SNS does not significantly interact, positively or negatively, with the self brand congruity perceived by that user in relation to branded adoptive behavior likelihood. M otivati ons outside of those examined in this study could be more suited as interaction variables, or other factors could be key influencers in this model such as brand familiarity. The differences per brand for this examination indicate that these measures are s ignificant factors unique for individual brands, meaning that outside factors could explain the reasons for such different responses to these variables. Next, R esearch Q uestions 1 and 2 approached the same type of interaction effect s as in H ypothesis 3 and H ypothesis 4 for the final two SNS usage motivations, social interaction, and convenience. These motivations were posed as research questions because they were predicted to have negative interaction effects with self brand congruity in predicting adoption behavior likelihood. Research Question 1: Do social interaction SNS motivations negatively interact with actual and ideal self brand congruity in predicting SNS brand adoption behavior likelihood? The regression model equation for the combined sample was significant showing a moderate positive correlation between the independent variables and adoption behavior likelihood. However, this combined sample did not return any of the inputted variables to have significant relationships with adoption behavior li kelihood. H owever, many significant individual variables existed for each of the brands individually. Starbucks revealed three significant relationships with adoption behavior likelihood : 1) the independent social interaction motivation which had a positiv e effect 2) actual self brand congruity which also had a positive effect and 3) as expected, a negative interaction effect between these two variables. Th ese result s mean that

PAGE 94

94 individually if a user perceives high congruity between his actual self and the typical user of Starbucks, he is more likely to positively adopt Starbucks into his SNS posting behavior. Likewise, if a user is motivated in his SNS usage by social interaction purposes individually he is more likely to positively adopt Starbucks int o his SNS posting behavior. However, if a user is both motivated to use an SNS for social interaction purposes and feel s high similarity between how he actually view s himself and the typical user of Starbucks, then he is less likely to adopt Starbucks into his SNS posting behavior. There could be a couple reasons for the tendency presented by these results. One explanation could be that when a user is highly motivated by social interaction purposes, whether or not he fee l s he is similar with the typical user of a brand plays a small role compared to when social interaction is no t a high motivational priority Another rationale is that perhaps he is more concerned about social implications of expressing his actual congruit y with the brand This study was performed during an economic recession where consumers showed a sense of pride in finding great deal s as opposed to over spending on trendy items. This kind of social economic consciousness could have played an influential role in the perceptions of brands as well as the willingness to associate with those brands. Dunkin Donuts on the other hand, did not return the individual social interaction motivation variable as significant ly related to adoption behavior likelihood. It did return both actual self brand congruity and the interaction of actual self brand congruity and social interaction motivation to have negative and positive significant relationships, respectively with the dependent variable though. Dunkin Donuts als o found ideal

PAGE 95

95 self brand congruity to be significant ly related positive ly with adoption behavior likelihood. These results found that the more a person perceives similarity between their actual selves and the typical user of Dunkin Donuts, the less likely he is to adopt Dunkin Donuts into their SNS posting behavior. This could suggest a sense of embarrassment regarding the perceived similarity, which indicates brand attitude and perhaps brand personality play important roles in this relationship. In concert with this conclusion, the more they perceive congruity between how they would like to view themselves and the typical user of Dunkin Donuts, the more likely they are to adopt Dunkin Donuts into their SNS posting behavior, which possibly i ndicates a se nse of pride, self confidence, or other qualities that would be natural products of positive brand attitudes. Lastly, if a person felt strong congruity between how he view s his actual self and how he view s the typical user of Dunkin Donuts and is also motivated in his SNS usage by social interaction purposes, he is more likely to adopt Dunkin Donuts into his SNS posting behavior. This implies that if a person is highly motivated by social interaction, perhaps he is less concerned with the social implic ations of expressing his self brand congruity in some fashion. The remainder of the variables, including th e interaction effects of social interaction with ideal self brand congruity for each sample, attested to a lack of significant relationship s with adoption behavior likelihood. Therefore, social interaction motivation failed to confirm a negative interact ion with both actual and ide al self brand congruity. Result differences between brands indicate that some brands are more influenced by social interact ion and self brand congruity than others, and each brand

PAGE 96

96 must be considered individually on qualities such as brand attitude and brand personality. Research Question 2: Do convenience SNS motivations negatively interact with actual and ideal self brand congruity in predicting SNS brand adoption behavior likelihood? As with the previous motivation examinations, t he multiple regression models for Starbucks and the combined sample executed for R esearch Q uestion 2 exposed linear relationships between the independent variables and adoption behavior likelihood and, can therefore b e projected onto the population. Once again, the model for Dunkin Donuts was not significant, so it cannot be projected onto the population. Within the Starbucks and combined samples, the independent convenience motivation variable proved to have a significant positive relationship with adoption behavior likelihood. Accordingly, the more a person is motivated by convenience in his SNS usage, the more likely he is to adopt a brand into his SNS posting activity. This shift from an expected negative effect to a positive one could be explained by ease of use associated with individual SNSs because this factor would likely increase or decrease usage outcomes of convenience motivations. A nother possible explanation could be linked to a persons SNS usage intensity. For example, if a person is already spending time on an SNS, the convenience of already being logged in with a network and media capabilities at his fingertips could positively infl uence the likelihood of adoption behavior. In addition to the significant positive effect of convenience motivation, a significant negative interaction effect on adoption behavior likelihood was observed by convenience motivation and ideal self brand cong ruity for the Starbucks sample. This says that the more a person perceives similarities between how he would like to see himself and the typical user of Starbucks and the more this person is motivated to use

PAGE 97

97 an SNS for convenience purposes, the less likely he is to adopt Starbucks into his SNS posting activity. One explanation for this could be that the person who view s high congruity between his ideal self and the typical user of Starbucks, is willing to extend beyond simple convenience activity and is mor e likely to have other motivations than convenience. This explanation would lead to a negative interaction with ideal self brand congruitys indication of adoption behavior likelihood. Unlike the interaction between ideal convenience motivation and ideal s elf brand congruity for the Starbucks sample, the remainder interaction effects were not significant. Therefore, while the convenience motivation may individually be significant ly related to adoption behavior likelihood, convenience motivation does not sig nificantly interact, positively or negatively, with actual and ideal self brand congruity perceived by that user in predicting branded adoptive behavior likelihood. T he lack of significant interaction effects hint that motivations outside of those examined in this study were more suited as interaction variables. Differences between results for Dunkin Donuts and Starbucks indicate that individual brands are influenced by self brand congruity and convenience motivations uniquely and should be observed on a per brand basis for future studies. Implications Summary Thus within the scope of this study, both actual and ideal self brand congruity proved to have significant positive relationships with adoption behavior likelihood independently while brandF acebook compatibility did not. Together these results along with differences in findings per brand implicate a need for a more intense examination of both brand attitude and unique brand characteristics or personalities and their relationships with the variables in this study. Furthermore, actual self brand

PAGE 98

98 congruity had more impact on adoption behavior likelihood both indivi dually and in interaction with motivations. N one of the motivation s significantly interacted with both the actual and ideal aspects of a persons self brand congruity in predicting the likelihood of whether or not a user adopts a brand into his SNS activity However, all four motivations showed promise as individual variables to have a significant relationship with adoption behavior likelihood on a per brand basis. Despite the fact that the four motivations used were acceptable determinants of Internet usa ge and SNS usage as indicated by this study the results shown here also imply that these SNS s motivation categories leave room for improvement in predicting branded SNS posting behavior. Improvements may involve a slightly modified combination of motivati ons that are n o t quite as transferrable from usage motivations of the Internet as a whole or new motivations that have evolved from SNS usage However, independent of interactions with self brand congruity each motivation explored in this study demonstrat ed a t least a moderate significant relationship with branded adoption behavior likelihood in many cases, as observed in this study In alignment with the increasing popularity of SNS s and interactivity between people and medi a the implications of this excavation can serve marketers as inspiration al tools to help navigate this realm in new ways Marketers can use this information to improve upon relationships with their target market and hopefully to ultimately improve upon pur chase behavior by more seamlessly integrating SNS campaigns into a marketing mix. The results of this study substantiate previous notions about the importance of knowing the target consumer on a personal level.

PAGE 99

99 The significant relationship between self br and congruity and branded SNS behavior indicates an importance for brands when forming their SNS presence to align themselves closely with their consumers self perceived qualities both actual and ideal To make this alignment, first, a marketer needs to clearly define his target market. Part of this process entails a marketer getting to know who his target market is, not just demographically, but also on a personality and psychological level Consumers entertainment, information social interaction and convenience tendencies should be included in this acquaintanceship, since these motivations showed significant influences on SNS adoption behavior unique per brand. Second, a marketer should align a brands perceived typical user image with the actual and ideal p erceptions and assessments of consumer s in the group he just described as his target market. According to the results of this study, a marketer positioning a brand this way will increase the frequency of which his brand is shared in a way that is very emotionally attached and which has the potential to efficiently reach a very large audience. As a result, increasing this frequency could help develop consumers of a brand simply through exposure. Furthermore, a key element in most marketing plans is word of mouth communication In lieu of this, the mention of a brand on a users profile can be especially valuable because that person is essentially attaching a trusted endorsement not just in conversation on a singular level, but now on a platf orm that reaches a more personal plural level. Future Research The insight obtained from this research could be just a small sample of a much larger pool of information to be attained. Future research could include a variety of outlooks for many questions are still left unexplained. First, this study focused only on

PAGE 100

100 positive adoption behavior likelihood. It did not observe self brand congruity in relation to behavior of positive versus negative attitude or intent, which would be a natural next step. The res ults of this study lend themselves to the likelihood that higher perceived self brand congruity results in more positive toned posting behavior while lower perceived self brand congruity results in more negative toned posting behavior However, results op posite to this conjecture would pose a question regarding a standoff between content attitude and simple exposure value of posting behavior as the more beneficial component for a brand. As the limitations section pointed out, privacy restrictions acted as a noteworthy limitation. A Facebook user could view content only if he is friends with the individual posting the content. This restriction limits the reach of such content. W ith access to unrestricted profiles, accurate content analysis could be observed as subsequent research. This kind of research would be uninhibited by the subjectivity of a participant in his responses. Furthermore, privacy is a key issue to many who use SNS s and thu s may impact posting behavior that create s personal associations publicly. It c ould be worth exploring the impact of such self inflicted privacy restrictions, attitudes, and preferences on SNS posting behavior in future research. Also, this study could be expanded to observe relationships between the variables in this study with purchase intent and purchase behavior. Do those who had a high degree of actual and ideal self brand congruity have an increased likelihood of purchase intent? Do these people als o have a higher tendency toward actual purchase behavior? Is there a significant difference between the effect of actual congruity and ideal congruity on purchase intent and purchase behavior? Do people with high self brand

PAGE 101

101 congruity indications or high adoption behavior on an SNS consume the brand with higher frequency than those who do not post in reference to the brand or those who do not feel a sense of self brand congruity actual or ideal ? Does consumer loyalty depend on whether a person posts regarding a brand first or become a consumer of the brand first? These are just a few questions that could be explained further in subsequent stud ies In addition, future studies could observe the interaction effects of consumers versus nonconsumers of a brand in question regarding predictive relationships between self brand congruity measures and adoption behavior likelihood. The current study primarily considers consumers of a product category. However, nonconsumers are likely to know of brands they do not consume and still have brand attitudes about them. Finally since only a select few of the motivation variables explored showed a significant effect on adoption behavior likelihood, future research could attempt to clarify more appropriate motivations for using SNS s and specifically for posting content on SNS s. A subsequent influencing variable could be perceived genuineness. For example, in terms of the wordof mouth communication portrayed in branded posting behavior, the effectiveness of this more personal form of communication could be affected by the number of friends the poster is perceiv ed to be delivering the message to or the frequency with which such branded posts are made. The research performed for this study successfully answered the questions it posed by further clarifying factors that influenced (or did not influence) branded SNS posting behavior. Th e suggested research topics would continue to provide a greater

PAGE 102

102 knowledge of posting behavior and its value equation for marketers by taking the implic ations revealed by this study and applying it to further examinations of this subject

PAGE 103

103 APPENDIX A INVITATION This is an invitation to participate in a research study. The goal of the study is to learn more about consumers activity on social networking sites. The information acquired will be important in expanding upon current research in advertising. Your par ticipation in this study is greatly appreciated. It should take no more than 15 minutes and should require minimal effort, just reliable responses. Participant responses will only be used for research and will remain completely confidential and anonymous Please pass this invitation to participate on to friends and family to complete as well. P articipation is greatly appreciated. If your first name starts with letters A L please click link #1: http://www.surveymonkey.com/s.aspx?sm=GBfx6lx_2f6NVMiuKrayum7A_3d_3d If you first name starts with letters M Z please click link #2: http://www.surveymonkey.com/s.aspx?sm=mBVIKb57kB6ZRTpfDzPxJA_3d_3d (If you manually enter the link into your navigation bar, please type the URL into your browser exactly as it appears above, including the specific uppercase, lowercase, symbols, and underscores as it is shown above or else it will not work) Please complete the survey by May 26, 2009. Surveys must be completed in their entirety in order to be valid data f or this study. If you have any questions regarding the study or have trouble accessing the website links, please feel free to contact me at any time at riediger.2@ufl.edu. Thank you for your participation!

PAGE 104

104 APPENDIX B ONLINE SURVEY QUESTI ONNAIRE Welcome! One of the emerging methods of communication in todays society involves social networking sites on the Internet As a whole, these sites are an unchartered medium through which consumers and brands may connect, yet very little research has been conducted on the effects of such methods of consumer relationship development. This survey includes a wide range of questions about you and your general feelings and activity toward brands in the context of social networking sites. The questionnaire should take approximately 1520 minutes to complete. Rest assured, you will not be asked to identify yourself individually within the survey and any information you provide will remain strictly confidential. The re are no direct benefits, risks, or compensation to you for participating in the study. You may discontinue or refuse to take part at any time and your responses will not be processed unless you submit the survey upon completion. By clicking the Submit button, you are indicating your voluntary consent to participate in this research. Please carefully read the instructions at the beginning of each section. Most of the questions can be answered by clicking on the button(s) that best expresses your response. Questions about the study should be directed to riediger.2@ufl.edu. Thank you very much for helping with this important survey. Submit____ 1. Age?____ 2. Do you drink coffee? ___Yes ___No a. If Yes, what kind of coffee do you drink most often? ___(1) a purchased cup of regular coffee (caffeinated or decaffeinated roasted coffee with or without some combination of cream and sugar) ___(2) a purchased cup of specialty coffee (includes all of the designer coffee options such as a Latte, Frappuccino, E spresso, etcetera) ___(3) home brewed coffee b. If Home Brewed Coffee was chosen for the previous question, what brand of home brew do you drink most often? ___(1) Starbucks ___(2) Dunkin Donuts ___(3) Folgers ___(4) Maxwell House ___(5) Other (Please Specify)_____________________

PAGE 105

105 3. Do you have an account with Facebook? If Yes, please indicate the date you joined the community in the space provided (or estimate as accurately as possible). ___(1) No ___(2) Yes Date : ___________ Please answer the following questions as truthfully and accurately as you can. Clearly circle the number that you think BEST describes you. The number 3 should indicate you neither agree nor disagree. I use Facebook (because) [coded Strongly Agree=1, Strongly Disagree= 5] 4. To learn about unknown things Strongly Disagree 5 4 3 2 1 Strongly Agree 5. Its a good way to do research Strongly Disagree 5 4 3 2 1 Strongly Agree 6. To learn about useful things Strongly Disagree 5 4 3 2 1 Strongly Agree 7. Its convenient to use Strongly Disagree 5 4 3 2 1 Strongly Agree 8. I can get what I want for less effort Strongly Disagree 5 4 3 2 1 Strongly Agree 9. I can use it anytime, anywhere Strongly Disagree 5 4 3 2 1 Strongly Agree 10. To pass time Strongly Disagree 5 4 3 2 1 Strongly Agree 11. I just like to surf the Internet Strongly Disagree 5 4 3 2 1 Strongly Agree 12. Its enjoyable Strongly Disagree 5 4 3 2 1 Strongly Agree 13. Its entertaining Strongly Disagree 5 4 3 2 1 Strongly Agree 14. I wonder what other people said Strongly Disagree 5 4 3 2 1 Strongly Agree 15. To express myself freely Strongly Disagree 5 4 3 2 1 Strongly Agree 16. To meet people with my interests Strongly Disagree 5 4 3 2 1 Strongly Agree

PAGE 106

106 17. How familiar are you with Facebook? Very Unfamiliar 5 4 3 2 1 Very Familiar 18. How acceptable is Facebook to you? Very Unacceptable 5 4 3 2 1 Very Acceptable 19. How likely are you to visit Facebook the next time you use the Internet ? Very Unlikely 5 4 3 2 1 Very Likely 20. How often do you frequent Facebook? _____(1) Multiple times per day _____(2) Once per day _____(3) Multiple times per week _____(4) Once per week _____(5) Multiple times per month _____(6) Once per month _____(7) Less than once per month 21. About how many hours and minutes would you say you actively s pend on Facebook in a typical week? Facebook Activity Time: ____hrs ____min 22. Please rank the following Facebook activities from 1 to 3. (1 meaning the activity you spend the most time on in your Facebook session, 3 being what you spend the least time doing.) _____Adding content to your profile or other users profiles _____Browsing _____Responding to/interacting with others Please indicate your level of agreement or disagreement with the following: 23. Facebook builds a relationship with me Strongly Disagree 5 4 3 2 1 Strongly Agree 24. I would like to visit Facebook again Strongly Disagree 5 4 3 2 1 Strongly Agree 25. I am satisfied with Facebooks services Strongly Disagree 5 4 3 2 1 Strongly Agree 26. I feel comfortable in surfing Facebook Strongly Disagree 5 4 3 2 1 Strongly Agree 27. Facebook is a good place to spend my time Strongly Disagree 5 4 3 2 1 Strongly Agree 28. I would rate Facebook as one of the best Strongly Disagree 5 4 3 2 1 Strongly Agree

PAGE 107

107 Your Facebook profile already displays: 29. Basic Information (1)Full (2)Partial (3)Limited (4)None (5)I am not familiar with this 30. Personal Information (1)Full (2)Partial (3)Limited (4)None (5)I am not familiar with this 31. Contact Information (1)Full (2)Partial (3)Limited (4)None (5)I am not familiar with this 32. Education and Work Information (1)Full (2)Partial (3)Limited (4)None (5)I am not familiar with this 33. Groups youve become a member of (1)Many (2)A Few (3)None (4)I am not familiar with this 34. Items youve become a fan of (1)Many (2)A Few (3)None (4)I am not familiar with this 35. Bumper Stickers from other Facebook members (1)Many (2)A Few (3)None (4)I am not familiar with this 36. Bumper Stickers added by yourself (1)Many (2)A Few (3)None (4)I am not familiar with this 37. Gifts f rom other Facebook members (1)Many (2)A Few (3)None (4)I am not familiar with this 38. Other Applications (1)Many (2)A Few (3)None (4)I am not familiar with this 39. How familiar are you with [brand]? Very Familiar 1 2 3 4 5 Very Unf amiliar [Brand] is 40. Extremely Good 1 2 3 4 5 Extremely Bad 41. Extremely Pleasant 1 2 3 4 5 Extremely Unpleasant 42. Extremely Favorable 1 2 3 4 5 Extremely Unfavorable 43. Extremely Lik able 1 2 3 4 5 Extremely Unlikable

PAGE 108

108 44. 45. 46. 47. 48. 49. Would you add [Brand] to your page in any way, either visually or written verbally? No, Definitely Not 1 2 3 4 5 Yes, Definitely 50. Consider the brand Starbucks on Facebook. Now, look through the following list and select all that apply. Consider each option on individual terms, exclusive from the others. I would: _____( 1) write/post/display something negative about [Brand] in my profile or status _____(+1) write/post/display something positive about [Brand] in my profile or status _____(0) ignore the brand and take no action _____(+1) become a fan of [Brand] _____(+1) search out [Brand] profile and more info _____(+1) add and display a bumper sticker of the [Brand] logo that you found _____(0) deny a bumper sticker of the [Brand] logo that someone else sent you _____(+1) add and display a bumper sticker of the [Brand] logo that someone else sent you _____(+1) send someone else a bumper sticker of the [Brand] logo _____(+1) send a [Brand] logo gift to someone else _____(0) deny a [Brand] logo gift that someone else sent you _____(+1) accept and display a [Brand] logo gift that someone else sent you _____(+1) search out and join a group with a positive attitude toward [Brand] _____( 1) search out and join a group with a negative att itude toward [Brand] _____(0) ignore an invite to join a group with a positive attitude toward [Brand] _____(0) ignore an invite to join a group with a negative attitude toward [Brand] _____( 1) accept an invite to join a group with a negative attitude tow ard [Brand] _____(+1) accept an invite to join a group with a positive attitude toward [Brand] _____(0) Other:_____________________________________________________ [Brand] Take a moment to think about [Brand]. Think about the kind of person who typically uses the brand. Imagine this person in your mind and then describe this person using one or two personal adjectives to describe the typical user of the brand. 44. Adjective #1 ___________________ 45. Adjective #2 ___________________ Now indicate your agreement or disagr eement to the following statements: 46. The typical user of Starbucks is consistent with how I see myself. Strongly Agree 1 2 3 4 5 Strongly Disagree 47. The typical user of Starbucks is consistent with how I like to see mysel f. Strongly Agree 1 2 3 4 5 Strongly Disagree 48. The Starbucks brand is compatible with Facebook. Strongly Agree 1 2 3 4 5 Strongly Disagree

PAGE 109

109 Demographic Questions: 51. Age _____ 52. Sex _____ (1) Male _____ (2) Female 53. What is your highest completed level of education? _____ (1) High School _____ (2) Associate Degree _____ (3) Associate Degree in progress _____ (4) Undergraduate Degree _____ (5) Undergraduate Degree in progress _____ (6) Graduate Degree _____ (7) Graduate Degr ee in progress _____ (8) Doctorate Degree _____ (9) Doctorate Degree in progress 54. Race/Ethnicity _____ (1) Asian _____ (2) Black/African American _____ (3) Hispanic/Latino _____ (4) Native American _____ (5) White/Caucasian _____ (6) Other__________________

PAGE 110

110 LIST OF REFERENCES Aaker, D. A., Kumar, V., & Day, G. S. (2004). Marketing research (8th ed.). Hoboken, NJ: John Wiley & Sons, Inc. About Facebook (2004, February 4). Retrieved February 8, 2009, from www.facebook.com: http://www.facebook.com/facebook Arndt, A. Z. (2001). New economy emotion: Engaging customer passion with ecrm. Chichester. England: John Wiley & Sons Ltd. Arrington, M. (2009, January 22). Facebook now nearly twice the size of MySpace Worldwide Retrieve d August 18, 2009, from TechCrunch: http://www.techcrunch.com/2009/01/22/facebook now nearly twice thesize of myspace worldwide/ Babbie, E. (2007). The practice of social research (11th ed.). Belmont, CA: Thomson Wadsworth. Barker, V. (2009). Older adoles cents' motivations for use of SNS: The influence of gender, group identity, and collective self esteem. International Communication Association (pp. 145). Montreal, Quebec, Canada: Retrieved from Communication & Mass Media Complete database. Bauman, Z. (2005). Liquid life. Cambridge, England. UK: Polity Press. Beavis, C. (2009). Games within games. In R. Willett, M. Robinson, & J. Marsh (Eds.), Play, creativity and digital cultures (pp. 1535). New York: Routledge. Blumler, J. G., & Katz, E. (1974). The us es of mass communications: Current perspectives on gratifications research. Beverly Hills, CA: Sage Publications. Boyd, D. (2006). G/localization: When global information and local interaction collide. Retrieved March 23, 2009, from O'Reilly Emerging Techn ology Conference: http://www.danah.org/papers/Etech2006.html Boyd, D. (2008). Can social network sites enable political action? International Journal of Media & Cultural Politics 4 (2), 241244. Boyd, D. M., & Ellison, N. B. (2007). Social network sites: definition, history, and scholarship. Journal of Computer Mediated Communication 13 (1), 210230. Chartrand, T., & Bargh, J. (1999). The chameleon effect: The perceptionbehavior link and social interaction. Journal of Personality & Social Psychology 76 (6), 893910.

PAGE 111

111 Company history (2008). Retrieved August 18, 2009, from LinkedIn: http://press.linkedin.com/history Company Timeline. (2009). Retrieved December 2, 2009, from Facebook: http://www.facebook.com/press/product.php#/press/info.php?timeline Consalvo, M. (2006). Console video games and global corporations: Creating a hybrid culture. New Media and Society 8 (1), pp. 117137. Consortium, C. C. (2006). About C3: Convergence. Retrieved April 27, 2009, from Massachusetts Institute of Technology: http://www.convergenceculture.org/aboutc3/index.html Curry, L. A. (2004). Affect, decision making, and adolescent risk behavior. Gainesville, FL: University of Florida. Davies, J. (2009). Online connections, collaborations, chronicles and crossings. In R. Willett, M. Robinson, & J. Marsh (Eds.), Play, creativity and digital cultures (pp. 108124). New York: Routledge. Davis, J. J. (1997). Advertising research theory and practice. Upper Saddle River, NJ: Prentice Hall. Dong, Q., Urista, M. A., & Day, K. D. ( 2008). Explaining why young adults use MySpace and Facebook through uses and gratification theory. Human Communication 12 (2), 215229. Douglas, J. E. (1999). The congruency between the ideal network and television network's brand personality: What makes the difference? Gainesville, FL: University of Florida. Dowdall, C. (2009). The texts of me and the texts of us. In R. Willett, M. Robinson, & J. Marsh (Eds.), Play, creativity and digital cultures (pp. 7391). New York: Routledge. facebook.com Quantcas t Audience Profile. (2010). Retrieved Feb 13, 2010, from Quantcast: http://www.quantcast.com/facebook.com Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An introduction to theory and research. Reading, MA: AddisonWesley. Fran ken, R. (1994). Human motivation (3rd ed.). Pacific Grove, CA: Brooks/Cole. Goodman, R. D., Maggio, J., Fedde, R. A., & Lyman, W. (2008). Growing up online [S.I.]: PBS Video http://purl.fcla.edu/UF/lib/kidsonline/.

PAGE 112

112 Hall, A. (2009). College students' motives for using social network sites and their relationships to users' personality traits. International Communication Association (pp. 1 38). Montreal, Quebec, Canada: Retrieved from Communication & Mass Media Complete database. Henson, C. A. (2007). An analysis of the coffee service industry in metro manila and the buying behavior of its consumers. Quezon City, The Philippines: Loyola Schools Review. Hoffman, D. L., & Novak, T. P. (1996). Marketing in hypermedia comput er mediated environments: Conceptual foundations. Journal of Marketing 60 (July), 50 68. Holzner, S. (2009). Facebook marketing: Leverage social media to grow your business. Indianapolis, IN: Que Publishing. Jenkins, H. (2006). Convergence culture: Where old and new media collide. New York: New York University Press. Kang, K. (1999). Interrelationship of ideal, actual, and undesired self/brand congruities across high, moderate, and low vaueexpressive products by Strausbaugh's brand personality measure. G ainesville, FL: University of Florida. Kaye, B. K., & Johnson, T. J. (2001). A web for all reasons: Uses and gratifications of internet resources for political information. Association for Education in Journalism and Mass Communication Conference. Washingt on DC. Keller, K. L. (2008). Strategic brand management: Building, measuring, and managing brand equity (3rd ed.). Upper Saddle River, NJ, New Jersey: Pearson Prentice Hall. Ko, H., Cho, C. H., & Roberts, M. S. (2005). Internet uses and gratifications: A structual equation model of interactive advertising. Journal of Advertising 34 (2), 5770. Korgaonkar, P. K., & Wolin, L. D. (1999). A multivariate analysis of web usage. Journal of Advertising Research 39 (2), 53 68. Lange, R., & Lampe, C. (2008). Fee ding the privacy debate: An examination of Facebook. International Communication Association (pp. 1 26). Montreal, Quebec, Canada: Retrieved from Commnication & Mass Media Complete Database. Levi Strauss, C. (1974). The savage mind. London: Weidenfeld and Nicolson. Li, D. (2008). Behaviors and motivations of social networking site users: A cross gender and cross cultural comparison. Gainesville, FL: University of Florida.

PAGE 113

113 Lin, C. A. (1999). Onlineservice adoption likelihood. Journal of Advertising Research 39 (2), 79 89. Lin, J. S. (2008). Antecedents and consequences of cross media usage [electronic resource] : A study of a TV program's official website. Gainesville, FL: University of Florida. Lopez, K. A. (2008). Student speech rights [electronic resour ce]: First Amendment implications for high school students on popular online social networks, MySpace and Facebook. Gainesville, FL: University of Florida. Lury, C. (1996). Consumer culture. Cambridge, England, UK: Polity Press. MacKenzie, S. B., Lutz, R. J., & Park, W. P. (1989). An empirical examination of the structural antecedents of attitude toward the ad in an advertising pretesting context. Journal of Marketing 53 48 65. Mitchell, A. A., & Olson, J. C. (1981). Are product beliefs the only mediator of advertising effect on brand attitudes? Journal of Marketing Research 318 332. Nixon, H. (2003). New research literacies for contemporary research into literacy and new media? Reading Research Quarterly 38 (3), pp. 407413. Papacharissi, Z., & Rubin, A. M. (2000). Predictors of Internet use. Journal of Broadcasting and Electronic Media 44 (2), 175 196. Park, L., Jin, B., & Jin, S.A. (2009). Motivations, impression management, and self disclosure in social network sites. International Communication Association (pp. 136). Montreal, Quebec, Canada: Retrieved from Communication & Mass Media Complete database. Parker, B. T. (2005). This brand's for me: Brand personality and user imagery based self congruity. Gainesville, FL: University of Florida. Quar t, A. (2003). Branded: The buying and selling of teenagers. London: Arrow Books. Rogers, C. R. (1959). A theory of therapy, personality, and interpersonal relationships, as developed in the client centered framework. (I. S. Koch, Ed.) Psychology: A study of a science pp. 184256. Rychlak, J. F. (1981). Introduction to personality and psychotherapy. Boston: Houghton Mifflin Company. Severin, W. J., & Tankard, J. W. (2001). Communication theories: Origins, methods, and uses in the mass media. New York: Addi son Wesley Longman.

PAGE 114

114 Shaw, M. E., & Costanzo, P. R. (1970). Theories of social psychology. New York: McGraw Hill. Shirky, C. (2009, May). Clay Shirky: How social media can make history | Video on TED.com. Retrieved June 16, 2009, from TED: Ideas worth spreading: http://www.ted.com/talks/clay_shirky_how_cellphones_twitter_facebook_can_make_hist ory.html Sirgy, M. J., Grewal, D., Mangleburg, T. F., Park, J.o., Chon, K.S., Claiborne, C. B., et al. (1997). As sessing the predictive validity of two methods of measuring self image congruence. Journal of the Academy of Marketing Science 25 (3), 229241. Sirgy, M. (1986). Self congruity: Toward a theory of personality and cybernetics. New York: Praeger Publishers Statistics (2009). Retrieved December 2, 2009, from Facebook: http://www.facebook.com/facebook?v=app_7146470109&ref=pf#/press/info.php?statisti cs Steinbock, D. (2000). The Birth of Internet Marketing Communications. Westport, CT: Quorum Books. Tinkham, S., & Weaver Lariscy, R. (1993). A diagnostic approach to assessing the impact of negative politcal television commercials. Journal of Broadcasting & Electronic Media 37 (4), 377 398. Trusov, M., Bucklin, R. E., & Pauwels, K. (2009). Effects of word of m outh versus traditional marketing: Findings from an Internet social networking site. Journal of Marketing 73 90 102. Van Cleemput, K. (2009). Authenticity and subcultural style in adolescents self presentation on social network sites. International Com munication Association (pp. 118). Montreal, Quebec, Canada: Retrieved from Communication & Mass Media Complete Database. Willett, R. (2009). Consumption, Production and Online Identities. In R. Willett, M. Robinson, & J. Marsh (Eds.), Play, creativity and digital cultures (pp. 5467). New York: Routledge. Wu, J. (2009). Facebook politics: An exploratory study of american youth's political engagement during the 2008 presidential election. International Communication Association (pp. 123). Montreal, Quebec, Canada: Retrieved from Communication & Mass Media Complete Database.

PAGE 115

115 BIOGRAPHICAL SKETCH Stefanie Riediger was born in Ontario, Canada, where she lived for eight years before moving to a suburb of Cleveland, Ohio. Though she is still a Canadian citizen her education has taken place in American schools. Stefanie completed her undergraduate Bachelor of Science in Business Administration degree at The Ohio State University w here she majored in b us iness with a specialization in marketing and minore d in visual communication design. After her graduation in the spring of 2007, she went on to earn her M aster of A dvertising degree from the College of Journalism and Communications at the University of Florida in th e summer of 20 1 0 Stefanie also complete d two internships with Bizresearch, a search engine marketing firm in Worthington, Ohio, and Saatchi & Saatchi X the shopper marketing division of the global Saatchi & Saatchi advertising agency in Fayetteville, Arkansas. S he is currently pursuing a career in advertising with worldwide agencies and clients.