$3M:

 

 

 

49.

vvzfi'm'no

M
.ur

 

,M

S"
I

r *1-
237.1%

3
22??"2'
I

“—3,- .
4min

mm .
t"

t» .5
'39}. Hi: 3
.~ ‘9' 1“,!

IQ»: 3.4."
23.2% 1 ‘
. 22$ .

2

 

'1me

2J0 I

This is to certify that the
dissertation entitled

THE GROUP VACATION MARKET (COLLEGE STUDENTS):

A SOCIAL PSYCHOLOGICAL APPROACH

presented by

SEUNG HYUN KIM

has been accepted towards fulfillment
of the requirements for the

Ph. D. degree in Park, Recreation, and Tourism
Resources

 

 

I’M/4M

'Major Professor’s Signature
:27/«3/0 7

Date

MSU is an affinnative-action. equal-opportunity employer

2-..- -.-.-.—.-.—.—2—.—2---—--—.-.- -4— -—.— -

..-.-.-.-.---

 

 

LIBRARY

Michigan State

University

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DAIEDUE

DATE DUE

DATE DUE

 

 

$6 I: 42009

 

'JZI 2,4 2.8

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6I07 p:lC|RCIDateDue.indd-p.1

 

THE GROUP VACATION MARKET (COLLEGE STUDENTS):
A SOCIAL PSYCHOLOGICAL APPROACH

By

Seung Hyun Kim

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment ofthe requirement
for the degree of

DOCTOR OF PHILOSOPHY

Park, Recreation, and Tourism Resources

2007

ABSTRACT

THE GROUP VACATION MARKET (COLLEGE STUDENTS):
A SOCIAL PSYCHOLOGICAL APPROACH

By
Seung Hyun Kim

College students are a significant tourism industry market segment, and group
vacation trips are a common phenomenon among college students. Since college
students’ group behavior occurs in a social context involving mainly friends,
understanding of group vacation trips is an important element in the study oftourism
markets. However. few researchers have investigated the motivators involved in such
group vacation trip behavior. Two important questions were addressed in this study: I)
what are the characteristics of the college student travel market? and 2) what factors
motivate college students to commit to participating in group vacation trips? Via a
sample of US. college students enrolled at nine universities collected online and in-class,
a group vacation market profile of college students and the effects of individual and
social factors on their group vacation trip behavior were developed and investigated in
this study.

A theoretical model was adopted from the consumer and marketing literature
which integrated three social behavior theories: goal directed behavior (an extended
model from the theory of planned behavior). social identity theory, and a conceptual
framework based on the concept of we-intention.

Results ofthe descriptive analysis confirmed that group vacation trips are
common phenomena among college students and showed the important role of friends as

information sources and as trip companions. The theoretical model of group vacations

was tested and confirmed via Structural Equation Modeling (SEM). Confirmatory factor
analysis in SEM validated the measurement model and, revealed that social identity is a
second-order three-factor construct comprising cognitive, affective, and evaluative
components. All hypothesized relationships in the SEM model were positive and
statistically significant, except the link between subjective norms and desire to take a
vacation trip with a group of friends. Study results provide empirical support for the
mediating role of desire to take a vacation trip with a group of friends, and the effects of
social (i.e., social identity and subjective norms) and individual factors (i.e., perceived
behavioral control and attitude) on shared intention (i.e., we-intention) to take a vacation
trip. Social identity had the most significant effect on we-intention to take a vacation trip
with a group of friends. This study further explored the moderating role of group size on
relationships between social identity and outcome variables, indicating that small groups
have more social identity effect on group vacation trips.

When developing marketing programs targeting college students, tourism
marketers should emphasize group travelers’ needs and wants. When they develop
tourism-related products or services targeting college students, they need to consider
marketing strategies enhancing social identity, to motivate shared intention to take a
vacation trip with a group offriends. Marketers should build group-friendly environments
and/or group-based activities to meet group needs and enhance group cohesion.

Suggested directions for future study include: I) measuring the role ofsubjective
norm, 2) evaluating the generalizability of sample and external validity, 3) developing
more social "factors, 4) comparing between an on—going and an initial group, and 5)

investigating other moderator variables.

Copyright by
Seung Hyun Kim
2007

DEDICATION
This dissertation is dedicated to my wife, .laemin Cha, for her spiritual
companionship and support, and to our beloved daughters, Seohee Kim and soon-to-be-
delivered baby for giving me happiness and for inspiring and motivating me throughout

my doctoral journey.

ACKNOWLEDGMENTS

Thank you to many who helped and supported me on my doctoral journey.

My dissertation chair and advisor, Dr. Donald Holecek, afforded me fine direction
and much of his valuable time. Throughout my doctorate he encouraged me to develop
independent thinking and research skills, and his oral and written comments always were
perceptive, helpful and appropriate. My committee member, Dr. Bonnie Knutson, also
has my deep gratitude. Her kind support and attentiveness drew me along the necessary
path and over the inevitable hurdles. She is my role model oftrue academic, real scholar
and firm teacher. I am honored and privileged to have been involved in many research
projects with her during my doctoral program. Sincere thanks also to my other committee
members, Drs. Christie Vogt and Sarah Nicholls for their continual support and
encouragement, and for their comments and suggestions which helped shape and refine
my manuscript.

Thank you, too. to other faculty, colleagues and specialists who helped in various
ways including my gathering and analysis of dissertation data, and myriad other aspects.
In these regards, thank you especially to Drs. Carl Borchgrevink , Jeffrey Beck, Sam
Larson, Jaeick Park, and to my international fellows in study, namely Jina, Suan, Hyeun,
Eunsoon and Ohoon. To Dr. Seoho Um, my graduate academic adviser in Korea, who
introduced me to destination marketing and inspired my professional passion for tourism,
my deepest thanks. Professor Michael Morris, my special mentor, has my highest praise
and recognition; as he has for so many international students similar to me, throughout
my doctoral process he held me firmly to the constant polishing ofmy dissertation

language and text format, pertinent oral and visual presentations. and associated

vi

communications. Without his skill, patience and unyielding strictness, myjourney would
have been even harder. Thank you to Dr. Ronald Cichy, Director and Professor of The
School of Hospitality Business, for his support in countless ways. He deserves and
certainly has my special recognition and gratitude.

How great a privilege it has been for me to have dear colleagues in tourism-
related academic society. Chang Huh, Youngrae Kim, Seoki Lee, Jonghee, and Jinyoung,
thank you for your friendship and collegiality on our East Lansing travails and triumphs.

None ofany ofthis would have been possible without the love and patience of my
family. My sincerest and. most deeply-felt gratitude is to my wife, Jaemin Cha, who
suffered more in and for my dissertation than did anyone. With aplomb and selfiessness
she underwent every emotion, endured all the fears and tears, the countless hours of my
solitude, detachment, anxiety. Absent her, I simply could not have completed this work.
Also thank you to my precious daughters, Seohee and soon-to-be-delivered newborn
baby, thank you for being alongside in this most important part of my life, thus also your
lives. Heartfelt thanks also to my parents, Sungyeol Kim and Moonsook Lee and my
parents—in-law, Joonsang Cha and Choonja Kwak, for their steadfast love and constant
prayers. I also am indebted to my American extended family, Michigan State University
Trustee Emerita Barbara Sawyer-Koch and Michigan State University Professor Emeritus
Donald Koch, for their extraordinary generosity and support.

Above all, my deepest gratitude is to God for being with me during my PhD. All

my privileges and success is His.

TABLE OF CONTENTS

LIST OF TABLES ............................................................................... XI
LIST OF FIGURES ............................................................................. xiii
CHAPTER I
INTRODUCTION ............................................................................... I
Overview ................................................................................... I
Background ................................................................................ I
Problem Statements ..................................................................... 2
Study Objectives ......................................................................... 4
Research Significance and Contributions ........................................ 5
A Different Approach to Predicting Group Travel Behavior ...... 6
Definitions of Terms for Main Factors ............................................ 6
Dissertation Organization ............................................................. 9
CHAPTER 2
LITERATURE REVIEW AND MODEL DEVELOPMENT ...................... I I
Overview ................................................................................... I I
Literature Review on Group Travel Behavior .................................... I 1
Vacation Decision-Making Research ...................................... I 1
Group Travel Behavior Studies in General ............................... 12
, Group Travel Behavior of College Students ............................. I4
Social Psychological Theories ....................................................... 15
Modified Model of Goal-directed Behavior (MGB) .................. 15
Social Identity Theory .......................................................... I8
An Alternative View of Intention: We-intention ....................... 22
The Conceptual Model ................................................................. 24
Hypothesized Relationships .................................................. 26
Moderating Effect of Group Size ............................................ 27
CHAPTER 3
METHOD .......................................................................................... 29
Overview ................................................................................... 2?
Data Collection ........................................................................... 29
Sample ............................................................................... 29
Data Collection Procedures ................................................... 29
Questionnaire Format ........................................................... 31
Sample Selection ................................................................. 32

viii

Measurements ............................................................................. 33
The Proposed Model ............................................................ 38
Data Analysis .............................................................................. 38
Descriptive Analysis ............................................................ 39
Structural Equation Modeling (SEM) ...................................... 40
Measurement Model ............................................................ 41
Structural Model and Hypothesis Testing ................................ 42
Multiple Group Analysis ...................................................... 43
CHAPTER 4
RESULTS .......................................................................................... 45
Overview ................................................................................... 45
Survey Response ......................................................................... 45
Data Collection ................................................................... 45
Characteristics of Survey Respondents .................................... 46
Study Subjects Selected from Survey Respondents ................... 47
Demographic Comparisons of Study Subjects to Other Subjects
and the Total College Population ............................................ 47
Profiling the Group Travel Market: Descriptive Analysis ................... 50
Past Vacation Experience with Friends .................................... 50
Group Composition on Past Vacation Trip with Friends ............. 51
Vacation Trip Plans for Coming Year ..................................... 53
Characteristics of Planned Trips with a Group of Friends ........... 53
Testing the Measurement Model: Confirmatory Factor Analysis in
SEM .......................................................................................... 60
Normality Test .................................................................... 60
Model Specification ............................................................. 62
Model Testing and Modification Procedures ............................ 62
Reliability .......................................................................... 65
Convergent Validity and Discriminant Validity ........................ 66
Testing the Hypothesized Structural Models .................................... 70
Goodness-of—fit of the Structural Model .................................. 70
Path Coefficients and Hypothesis Testing ................................ 71
Direct, Indirect, and Total Effects on We-intention ................... 75
Testing for Moderating Effects on the Proposed Structural Model ....... 76
Moderating Effect of Group Size ............................................ 76

CHAPTER 5
DISCUSSION AND CONCLUSIONS ....................................................

Overview ...................................................................................

Summary of Key Findings ............................................................
Characteristics of College Students’ Group Vacation Trips .........
Valid Measurement Model ....................................................
Mediating Path Model, with an Explanation ofa Non-significant
Effect ................................................................................

Theoretical Implications ...............................................................
Conceptual Development and Empirical Validation ..................
Incorporating Both Social and Personal Factors ........................

Managerial Implications ...............................................................
Understanding College Student Travel Market .........................
Targeting Friendship Group Rather Than Individual College
Students .............................................................................
Connecting Social Identity to Vacation-Related Products and
Services .............................................................................
Using Online Travel Community Enhancing Social Identity .......
Building Group-Friendly Environments Enhancing Social
Identity ...............................................................................

Limitations and Future Research ....................................................
The Subjective Norm Measurement Issues ...............................
Identifying More Social Factors .............................................
Generalizability of Results and External Validity ......................
Comparison Between Past and Potential New Partners ..............
Extending the Model of Intentions to Actual Participation
Behavior ............................................................................
Investigating Other Moderators ..............................................

APPENDICES ...................................................................................
Appendix A. Consent Form for Main Study .....................................
Appendix B. Survey Instruments ....................................................

REFERENCES ...................................................................................

79
79
79
79
81

82
83

84
85
86
86

87

88
89

90
91

91
92
92
93

94
94

96
97
98

l06

Table I

Table 2

Table 3

Table 4

Table 5

Table 6

Table 7

Table 8

Table 9

Table IO

Table I I

Table I2

Table 13

Table I4

Table 15

LIST OF TABLES

Description of factors and observed variables in the proposed
model ................................................................................... 35

Summary of data collection: total sample. returned sample. invalid

sample, valid sample ............................................................... 46
Institutions in which survey respondents were enrolled (n=697) ..... 46
Survey respondents’ academic majors (n=697) ............................ 47

Demographic characteristics of study subjects and other subjects
for all survey respondents, study subjects and other subjects ......... 48

Comparison of demographic characteristics of study subjects and

all US four-year college students .............................................. 49
Past vacation trip experience of survey respondents ..................... SI
Group composition on past vacation trips with a group of friends 52
Vacation trip planning intentions of survey respondents ............... 53

Characteristics of planned future vacation trips with a group friends 56

Normality test results of observed variables included in the
proposed model presented in Figure 3 ........................................ 61

Comparison of overall fit indices for proposed and modified full
measurement models (n=452) ................................................... 65

Variables removed from the proposed measurement model based
on the examination oflow factor loadings, multiple loadings and

large residuals ........................................................................ 66

Means, standard deviations and reliabilities for factors in Model 3
(n = 452) ............................................................................... 67

Confirmatory factory analysis results for the Model 3 (15 of
original 34 observed variables removed) (N = 452) ...................... 68

xi

Table 16

Table 17

Table 18

Table 19

Table 20

Table 21

Table 22

Table 23

Table 24

Table 25

Correlations among exogenous factors in the Model 3 for

examining discriminant validity (N = 452) ................................. 69
Chi-square difference tests for evaluating discriminant validity:

Model 3 (N = 452) .................................................................. 70
Overall fit indices for the proposed structural model (N = 452) ..... 70

Path coefficients in hypothesized structural model and hypotheses
testing results ........................................................................ 72

Comparison of goodness-of-fit tests for the proposed model and
nested models with paths dropped from the proposed model ......... 73

Comparison of explained variance in we-intention for 1) the theory

of planned behavior, 2) the theory of planned behavior plus desire,

and 3) the theory of planned behavior plus desire and social

identity ................................................................................. 75

Summary of indirect and direct effects of factors on we-intention to
take a vacation trip with a group of friends in hypothesized

structural model in SEM .......................................................... 76

Testing for moderating effects based on the proposed structural
model ................................................................................... 76

Results of testing for moderating effects based on the proposed
model (Figure 3): test of invariance for path coefficients ............... 78

Multiple group path analysis: comparison of path coefficients and t
values for the proposed model .................................................. 78

xii

Figure I
Figure 2

Figure 3

Figure 4

Figure 5

LIST OF FIGURES

Proposed Conceptual Model ....................................................
Depiction ofprocess used to select the primary study population

Proposed hypothetical model of intention to take a vacation trip
with friends in structural equation modeling (SEM) format ...........

Proposed measurement model ..................................................

Test results for the proposed structural model: standardized path
coefficients and squared multiple correlations (R2) ......................

xiii

b)
b)

39

63

72

CHAPTER 1

INTRODUCTION

Overview
General background and studyjustification are presented in chapter one. In the
first part, theoretical and practical background information is presented regarding college
students and their vacation travel decisions. In the second part, the problems identified to
serve as the focus of this study are discussed. Thereafter, study objectives and theoretical
justifications are provided, and the significance of the study and its contributions are
discussed. Finally. conceptual definitions of factors used in the proposed model are

presented.

Background

For the tourism industry, college students represent a large and growing market.
The US. Census Bureau (2005) estimates that there were more than 17 million students
enrolled in US. colleges in 2004. For the travel industry, one portion ofthe college
students’ travel phenomenon alone, namely spring break, generates close to US$I billion
(Mattila, Apostolopoulos, Sonmez, Yu, & Sasidharan, 2001).

Group travel is the most common phenomenon in tourism behavior, because
tourism often involves activities and experiences with friends, family, relatives and others.
Thus, exploring determinants of individual decisions to take a group vacation trip in
group is an important area for researchers and marketers in the fields of tourism

marketing and consumer behavior.

To date, consumer behavior studies have elaborated on both personal and social
factors in the context of group behavior (e.g., Bagozzi, 2000; Bagozzi & Dholakia, 2002;
Bagozzi & Dholakia, 2006; Doholakia, Bagozzi, & Pearo, 2004; Kropp, Lavack, &
Silvera, 2005; Mangleburg. Doney, & Bristol, 2004). In particular, Bagozzi and his
colleagues introduced the concept of intentional social action to the field of marketing
and consumer behavior. His examples of collective concepts or intentional social action
include soccer team members “celebrating a victory by patronizing a restaurant,” and a
couple “discussing our vacation in the Caribbean,” and a family “making a purchase of a
sports utility vehicle” (Bagozzi, 2000, pp 388-399). Although intentional social action or
collective concepts have not been investigated in the context of vacation decision-making,
it is believed that the notion ofintentional social action is also well suited to the context
of group travel with friends by college students, since, for example, college students often
refer to going to Florida or Cancun with a group of friends during spring break.

The following problems were identified, leading to the focus ofthis present study.

Problem Statements
First, the travel decision is salient in the group or social context (i.e., a vacation
trip with a group of friends among college students in this study). Although the role of
social influences on general consumer behaviors has been studied, there is limited
research incorporating social and individual influences to examine an individual’s
decision to travel with others. For example, many empirical studies have primarily
emphasized exploring the individual’s psychological, attitudinal, or motivational factors

(i.e., push and pull factors) in examining individual travel intentions, behaviors, and/or

destination choices, while ignoring social influences from others (cf. Crompton, I981;
Laing & Crouch, 2005; Mayo & Jarvis, I981; Middleton, I994; Moutinho, I987;
Schmoll, I977; Um & Crompton, I990; Woodside & Lysonski, I989). The influences of
others in the decision-making process is common theme in family vacation studies. For
example, Wang, Hsieh, Yeh, and Tsai (2004) examined family role relationships over
decision-making stages, comprising problem recognition, information search, and final
decision. They discovered that the role of parents and children in the travel decision-
making processes vary across different decision-making stages. Kang and Hsu (2005)
studied dyadic consensus between spouses on family vacation destination selection.
Although these studies of family vacation recognize the importance ofinfiuence of others
in the decision-making process, they still have not explored fully why individuals decide
to travel in groups. And, in the context of group travel with friends among college
students, it can be assumed that there are important social factors influencing the desire to
participate in such group travel.

Second, marketers have insufficient information to select marketing programs to
increase purchasing volume by college students who, in fact, are at the beginning of their
careers as consumers or at least in the early stages of their consumerist careers (Kropp, et
al, 2005). By 2000 the spending power of college students was estimated at more than
US$90 billion, with full-time, four-year students spending US$30 billion (Knutson, 2000)
and by 2006 had risen to US$182 billion (Harris Interactive, 2006). This spending power
is a function of students’ significant disposable incomes and few financial responsibilities
of an immediate nature. It is important to note that a vacation trip with a group of friends

among college students is often preferred to an individual trip or a family trip. For

example, 87% of college students who travel take a vacation trip with a group of friends
during spring break (Kim & Jogaratnam, 2002).

Tourism marketers have sought to motivate college students to take a vacation trip
or select a particular destination by emphasizing psychographic segmentation techniques.
They tend to assume that peoples’ stable personality traits can be identified and linked to
broad lifestyle orientations that then can fit general product classifications. Tourism
marketers also need to understand how the influence of others can shape college students’
decision making processes. Thus, in their marketing strategies, they need to focus more
on addressing various social influences that were explored in this study such as one’s
identity with friends (e.g., social identity), feelings of approval by friends (normative

influence), and perceived behavioral control over friends.

Study Objectives

The first objective of this study was to develop a profile of the college student
vacation trip market. The particular target market for profile development was college
students who take a vacation trip with a group of friends.

The second objective of this study was to examine why college students take a
vacation trip with a group of friends (i.e. decision making for individual participation in
group behavior). A special focus of this study was on the relative effects of social factors
(i.e., social identity and subjective norms) and individual factors (i.e., perceived
behavioral control and attitude) on a college student’s desire and intention to take a
vacation trip with a group of friends. Sirakaya and Woodside (2005), in their most

recently published article on decision making by travelers, argued that future research on

tourism decision making should integrate social factors and personal factors into a single,
theoretically sound model. This present study attempts to respond to this suggestion by
incorporating both social and personal factors in the context of travel decision-making.

The third objective of this study was to test and validate the mediating role of a
college student’s desire to take a vacation trip with a group of friends. Therefore, indirect
as well as direct effects of social factors and individual factors on a college student’s
intention to take a vacation trip with a group of friends were examined.

The final objective of this study was to examine the moderating effects of group
size (i.e., smaller versus larger group) on social processes for group vacation trips in the
proposed model framework. The magnitude of hypothesized relationships between social
identity and a college student’s desire to take a vacation trip with a group of friends and
between social identity and a college student’s intention to take a vacation trip with a

group of friends were examined across differing group travel sizes.

Research Significance and Contributions

The results from this study will contribute to the body of tourism research by
explaining social and individual influences on travel decision-making. In particular, the
role of social influences has not been included in most consumer behavior models of
group travel behavior, yet social theories can contribute to explaining this role of social
influences. More specifically, although decisions often are made in a social setting with a
group ofindividuals, each playing a specific role, most research has focused primarily on
psychological variables, not on social variables. The study of group travel as a social

activity requires a different approach.

To identify important factors influencing group travel behavior among college
students, a model depending upon the marketing and consumer behavior was adopted
which is built on three major frameworks. These include: the model of goal-directed
behavior, modified from the theory of planned behavior; the theory of social identity; and
concepts of we-intention (i.e., collective intention) from Bagozzi and his colleagues’
work (Bagozzi, 2000; Bagozzi & Dholakia, 2002; Bagozzi & Dholakia, 2006; Dholakia,
Bagozzi, & Pearo, 2004; Perugini & Bagozzi, 2001). These three frameworks were
integrated into the proposed framework for investigating factors that influence a college
student’s desire and intention to take a vacation with a group of friends. The findings of
this study were expected to extend the body of knowledge about travel decision making
with important implications for academic researchers and practitioners in the field of
tourism marketing and tourism consumer behavior. The additions of desire and social
identity to the theory of planned behavior were expected to lead to stronger exploratory
power than a model absent of these two constructs.

Group travel (traveling with a group of friends) is recognized increasingly for its
substantial potential in terms of size and economic importance. College students
primarily travel with friends and like being with friends while they travel (Josiam,
Clements, & Hobson, 1994). College students are at the beginning oftheir careers as
consumers. Their relative inexperience as consumers may render them uniquely
susceptible to group influence. In other words, they lack experience as consumers, which
may result in greater influence by peers and greater consumer susceptibility to
interpersonal influence (Kropp et al., 2005). Results from this study were expected to

provide valuable insights to travel marketers regarding college student travel behavior

since travel marketers can benefit from knowledge of social identity when developing
advertising content and marketing interventions.
A Different Approach to Predicting Group Travel Behavior

The direction and focus ofthis study differ from previous travel decision-making
studies in several ways. First, this study is limited to identifying factors leading to a
personal decision to participate in a vacation trip within a specific group, rather than
those leading an individual to select a particular destination. Second, while most ofthe
literature has focused on individual traveler’s behavioral intentions, here a college
student’s commitment to participating in ajoint act — traveling with a group of friends is
emphasized. Factors that influence a college student’s desire and intention to take a
vacation trip with a group of friends are of primary interest. Third, while previous
researchers mainly have explored internal motives, psychological or personal influences
predicting behavioral intention to take a vacation trip, here the importance of social

influences in the proposed framework is emphasized.

Definitions of Terms for Main Factors
Social Identity
Social identity is defined as “the individual’s knowledge that he or she belongs to
certain social groups together with some emotional and value significance to him or her
with group membership” (Tajfel, 1978, p. 63). In this study, social groups means a group
of friends that a college student identified representing those with whom he or she
previously has taken a vacation trip. This social group is based on friendship among

college students.

Cognitive social identity. Cognitive social identity refers to a cognitive awareness
of one’s membership in a group of friends (Ellemer, Kortekaas, & Ouwerkerk, I999)

Affective social identity. Affective social identity refers to a sense of emotional
involvement with the group of friends (Ellemer et al., 1999).

Evaluative social identity. Evaluative social identity connotes either a positive or
a negative value attached to group membership (Ellemer et al., 1999).
Attitude

Attitude towards taking a vacation trip with a group of friends is the degree to
which participation in a group vacation trip is positively or negatively valued (Ajzen,
I991).
Subjective Norms

Subjective norms are the perceived social pressures by significant others to
engage or not to engage in personal participation in a vacation trip with a group of friends
(Ajzen, I991).
Perceived Behavioral Control

Perceived behavioral control refers to individual perceptions of abilities to take a
vacation trip with a group of friends (Ajzen, 199]).
Desire to Take a Vacation Trip with a Group of Friends

Desire to take a vacation trip with a group of friends is defined as the motivational
state of mind wherein appraisals and reasons for group vacation behavior are transformed

into motivation to do so (Perugini & Bagozzi, 2001).

an

We-intention

We-intention to take a vacation trip with friends is defined as a commitment of an
individual or shared intention to group vacation trip behavior, and involves an implicit or
explicit agreement between participants to engage in thatjoint action (Tuomela, 1995).
Vacation Trip

Vacation trip is defined as an overnight trip (a night or longer) during vacation
time (spring or winter break. summer vacation, or holidays).
A Group of Friends

A group of friends refers to a group including one or more friends. In this study,
the notion of a group of friends in the proposed theoretical framework refers to those
identified by survey participants (college students). In particular, a group of friends does
not refer to the initially-formed travel friends group, but refers to those with whom
survey participants have taken a vacation trip previously, i.e., an namely ongoing
friendship vacation group. This definition is important, since the purpose of this study
was to investigate the issue of sustaining group intention (continuing to take vacation
trips with a group of friends which were already formed) rather than the initiation of new
travel groups (initiate taking a vacation trip with a group of friends who have never

traveled together).

Dissertation Organization
The organization ofthis dissertation is as follows. Chapter 1 provides the general
background and justification ofthe study. Chapter 2 discusses theoretical and empirical

issues and deficiencies in the area of vacation decision-making research. It includes a

literature review of group behavior and social influences, and it also provides an
alternative perspective to view decision-making research. Three underlying theories upon
which the research model is based are also introduced. On the basis of the literature
review, a group travel decision model (personal participation in group travel behavior) is
developed and research hypotheses are then explained. Chapter 3 focuses on the research
method employed, how the sample was defined, how data were collected, and how
constructs in the research model were operationalized. Characteristics of sample
statistics are also described. Chapter 4 describes how measures are validated in the
measurement model, how research hypotheses were tested in the structural model, and
provides data analyses results. Chapter 5 includes implications drawn from data analyses,

presents research limitations. and outlines future work to build on the results ofthis study.

CHAPTER 2

LITERATURE REVIEW AND MODEL DEVELOPMENT

Overview

Theoretical and empirical issues and deficiencies in the area oftravel decision-
making research are discussed in this chapter. It is emphasized that travel decision-
making research should focus on the role of social influences, and not solely on
individual or psychological variables since a vacation trip often occurs in the social
context ofa group, especially among college students. Three conceptual approaches — l)
a Model of Goal-directed Behavior, an extended Theory of Planned Behavior, 2) Social
Identify Theory, and 3) an alternative view of intention, namely, “We-intention” — are
discussed in explaining the proposed theoretical model. Finally, the hypothesized

relationships in the proposed conceptual model are presented and highlighted.

Literature Review on Group Travel Behavior

Vacation Decision-Making Research

To further understanding of why people take vacation trips and the factors which
influence their vacation trip behavior, previous researchers have focused mainly on the
decision — whether to go and where to go. The former refers to whether or not a traveler
decides to take a vacation trip, and the latter refers to a particular destination choice.
Certainly, investigation of destination choice models has advanced understanding of what
factors or cues are more important in deciding a destination choice for an individual
traveler (cf., Ankomah, Crompton, & Baker, 1996; Crompton, I992; Tussyadiah, Kono,

& Morisugi, 2006; Um & Crompton, I990; Woodside & Laysonski. I989). Traditionally,

push-pull theory has been used as an important framework for examining the different
forces that influence or push a traveler to consider taking a vacation and, given that
decision, the forces that attract or pull the traveler to select a particular vacation
destination (Crompton, I979; Dann, I981; Klenosky, Gengler, & Mulvey, 1993). In
particular, Crompton (1979) in studying motivations for pleasure vacations, identified
nine motivational factors, which translated into two main themes — l) “socio-
psychological” motives comprising “escape from a perceived mundane environment,”
“exploration and evaluation of self,” “relaxation,” “prestige,” “regression” and
“facilitation of social interaction,” and 2) “cultural” motives, comprising “novelty” and
“education.” The socio-psychological motives related to push factors were shown to be
more important than cultural motives related to pull factors when developing and
promoting tourism-related products. Tourism researchers have conducted numerous
studies to show how these kinds of push and pull factors influence attitudes, and how
these attitudes lead to taking a vacation trip or choosing a travel destination.
Group Travel Behavior Studies in General

The importance of group-based behavior is recognized in consumer behavior and
marketing research. Examples include: going for lunch with friends (Bogozzi & Lee,
2002), chatting with friends (Bagozzi, 2002), and shopping with friends (Mangleburg,
Doney, & Bristol, 2004), and participating in small group brand community (Bagozzi &
Dholakia, 2006). All these studies incorporated social influences as well as individual
influences to examine group-based behavior.

The notion of social influences and group-based behavior in travel and leisure

settings has been addressed by several researchers (Ajzen & Driver, 1992; Cho &

12

Kerstetter, 2002; Gitelson & Kerstetter, I994; Lam & Hsu, 2004; 2005; Um & Crompton,
I990; Woodside & MacDonald, 1994). In the late 1960’s, Burch (I969) emphasized that
personal motivational drives are filtered and redirected by the social circles such as
workmates, family and friends in the context ofleisure. Cheek, Field, and Burdge (1976),
in their study oftravel behavior at a zoological park, concluded that group membership
was more influential in channeling participation into particular activities than were
individual needs or desires. Crompton (1981) also recognized the need for a different
approach in studying a traveler’s behavior by emphasizing the importance of social
influences on a traveler’s behavior. He argued that individuals do not function as
independent entities in a society; rather they interact with other people. Crompton (1981)
suggested that “why do people go on vacation or go to a particular destination is
sociological as well as psychological, for the social group reinforces, modifies, and molds
the motivations of its members. It appears that research efforts in this area should be
designed to address both of these dimensions rather than either independently.”

Family travel has been recognized as one of the most common types of group-
based travel behavior. Previous researchers have investigated how the influences of
others play a role in the decision-making process for family vacations. For example,
Gitelson and Kerstetter(1994) studied the influence of information through social
interaction with family members on deciding destination choice and type of activities.
Wang et a1. (2004) examined family role relationships over decision-making stages,
comprising problem recognition, information search, and final decision. They discovered
that the role of parents and children in the travel decision-making processes vary among

different decision-making stages. Kang and Hsu (2005) studied dyadic consensus.

I3

between spouses. on family vacation destination selection. Although these previous
studies of family vacation decision-making recognize the importance ofinfluences from
others in the decision-making process, what kinds ofindividual and social factors
influence an individual’s decision to take a vacation trip with others have not been fully
investigated.
Group Travel Behavior of College Students

Just as spouses and children represent the relevant social group in family travel
decision-making, a group of friends represents the social group for college students
relevant to predicting their intentions to take a vacation trip with a group of friends.
Going on vacation with a group of friends is a common tourism phenomenon among
college students. Thus, taking a vacation trip with a group of friends should be viewed as
a group behavior, rather than an individual behavior. Olmsted and Hare (1978) defined a
group as “a plurality of individuals who are in contact with one another, who take one
another into account, and who are aware of some significant commonality” Their
definition is directly applicable to college students’ travel. The study of group travel
behavior focusing on a group of friends, however, rarely has been investigated in the
context of college travel. Sirakaya and Woodside (2005) recently published a conceptual
paper, building and testing theories of decision making by travelers in general. Among
many important issues in theories of decision making, they stressed that tourism is a
social activity that involves others (especially in group travel), and its decisions may be
made in a social setting with a group ofindividuals (Sirakaya & Woodside, 2005). Their
suggestions seem applicable to the context of vacation decision-making by college

students.

14

In summary, most travel decision-making studies have failed to account for both
individual behavior and group behavior (i.e., traveling alone or with companions). That is,
most tourism studies show little concern with how others may influence personal travel
decision-making for group behavior. For college students, taking a vacation trip occurs in
the context ofa social group, namely a group of friends. In predicting their desire and
intention to take a vacation trip with a particular group of friends, both individual and
social factors should be examined together. To investigate this process, this study
integrated social psychological theories into the proposed decision making model

framework.

Social Psychological Theories

A proposed theoretical model was built on three prominent social psychological
frameworks and theories, namely 1) the extended Theory of Planned Behavior,
particularly the Model of Goal-directed Behavior (Perugini & Bagozzi, 2001), 2) the
Social Identify Theory (Tajfel, 1978), and 3) the framework of “We-intention” (Bagozzi,
2000). These three frameworks were integrated into the proposed model to determine
what factors influence a college student’s desire and intention to take a vacation trip with
a group of friends.
Modified Model of Goal-directed Behavior (M GB)

The Theory of Planned Behavior (TPB) (Ajzen, 1991) has been used broadly in a
variety of fields including tourism. The TPB posits that one’s personal intention to enact
behavior is a function of the individual’s attitude toward that behavior, his/her subjective

norms and perceived behavioral control. In the TPB, attitude and perceived behavioral

15

control are considered individual influences, while subjective norms are considered a
social influence. Perugini and Bagozzi (2001) extended the TPB by adding three
variables — anticipated emotions, past behavior, and desires — along with attitude,
subjective norms, and perceived behavioral control, in the hope of better explaining and
predicting purposive behaviors. Recently, the Model of Goal-directed Behavior (MGB) in
group behavior has been shown to improve the predictive and explanatory power of the
TPB in the context of brand and virtual communities (Perugini & Bagozzi, 2001; Bagozzi
& Dholakia, 2002; Bagozzi & Dholakia, 2006; Dholakia, Bagozzi, & Pearo; 2004). MGB
is especially applicable to group vacation trip in that it is a behavior to accomplish a
group goal. The main highlight of MGB is the mediating role of“desire” in predicting a
purposive behavior to goal achievement, meaning that desires were proposed as the
proximal determinants of intentions. The present study also includes the desire to take a
vacation trip with a group of friends, as a mediator to explain the effects of individual and
social factors on the We-intention to take a vacation trip with a group of friends. MGB
components are presented below.

Attitude. Attitude toward a behavior is the degree to which performance of the
behavior is positively or negatively valued (Ajzen, 1991). Attitude toward taking a
vacation trip with a group of friends is conceived of as an evaluative response (e.g., Eagly
& Chaiken 1993). Such responses, in turn, are assumed to motivate decisions to act. Thus,
when a college student positively evaluates taking a vacation trip with a particular group
of friends, his or her desire and intention to take a vacation trip with a group of friends is

expected to increase.

16

Perceived behavioral control. Perceived behavioral control is defined as the
perception of how easy or difficult it is to perform the behavior (Ajzen, 1991) Travel
behavior may pose difficulties in execution to college students that may limit volitional
control. Perceived behavioral control was expected to directly influence both a college

student’s desire and We-intention to take a vacation trip with a group.

Subjective norm. Subjective norm is defined as perceived social pressure to
perform or not perform a behavior. Subjective norm has been identified as one ofthe
social factors that influence behavioral intention. An individual’s perceived subjective
norm reflects social pressure from significant others to perform a focal act. Subjective
norms reflect the influence of expectations of others (not themselves) and constitute what
Kelman (1974) terms “compliance.” Individuals holding subjective norms are believed to
be motivated by the need for approval from significant others.

Desire. Desire is defined as “the motivational state ofmind wherein appraisals
and reasons to act are transformed into a motivation to do so” (Perugini & Bagozzi, 2001,
p.84). For desires to function as motivators, Davis (1984) maintained that decision
makers (college students, namely travelers who consider taking a vacation trip with a
group of friends in this example) must be aware of their desires and accept them as
motivating reasons for acting. Frankfurt (1988) also saw a special role for desires. In his
view, desires function as determinants of decisions when decision makers give self-
reflective consideration to their desires and come to endorse them as motivators to act.
Perugini and Bagozzi (2001) explained that “desires” provide direction for intentions and
transform the motivational content to act embedded in attitudes toward the act, subjective

norms (SN) and perceived behavioral control (PBC). According to them, “attitudes, SN,

17

and PBC provide reasons for acting, it is claimed, but do not incorporate explicit
motivational content needed to induce an intention to act” (Perugini & Bagozzi, 2001).
This motivation of desire is hypothesized as the most proximal determinant of intentions
in the MGB. Empirical support for the motivational role of desires in predicting a
collective intention can be found in the literature. Namely, strong relationships between
desire and We-intentions were empirically supported in settings such as a virtual
community (Dholakia et al., 2004; Lee & Bagozzi, 2002) and brand community of auto
bikers (Bagozzi & Dholakia, 2006). According to the tourism literature, push motivations
such as the main variables in MGB (i.e., attitude, subjective norms) have been useful in
explaining the desire for travel (Christensen 1983; Crompton, 1979).
Social Identity Theory

Concept of social identity. The rationale for including the effects of social identity
on a college student’s desire and We-intentions to take a vacation trip with a group of
friends draws on Social Identify Theory (e.g., Hogg & Abrams 1988; Taj fel 1981; Terry,
Hogg, & Duck 1999; Turner 1985). Social Identify Theory explains group activity as a
kind of shared concept rather than a personal mental state. Social Identify Theory
provides the basis for specifying what shared consciousness is and how it creates a
readiness to engage in group action (Bergami & Bagozzi, 2000). Social identity affects
individuals’ perceptions, cognitions, evaluations, and attributions about issues and events,
and stimulates behaviors for the benefit of group members (Underwood, Bond, & Baer,
2001). Taj fel and Turner (1986) posited a distinction between personal and social identity.
According to them, personal identity is mainly under the control of personal variables,

while social identity is derived primarily from group memberships. That is. the concept

18

of social identity deals with the structure and function of identity as related to peoples’
membership in groups. The essence of Social Identify Theory is that people take part in
the creation of their own “categories” (Schouten & McAlexander, 1995). Categories
means here that one tries to determine a social category into which one falls, and to
which one feels one belongs, and provides a definition of who one is in terms ofthe
defining characteristics of the category (Hogg, Terry, & White, 1995). For example,
Deaux, Kim, and Kathleen (1995) identified categories that meaningfully distinguish
various forms of social identity. In this present study, a college student identifies himself
or herself as related to a friendship group. Their research (Deaux et al., 1995) suggests
several trait clusters including personal relationships (e.g., wife, friend), political
affiliations (e.g., democrat, independent), stigmatized groups (e.g., alcoholic, homeless)
and ethnic and religious groups (e.g., Asian Catholic). Others have suggested that people
gain a sense of who they are and derive much of their esteem from their memberships in
such social groups and categories (Tajfel & Turner, 1979). Research has shown that
people are motivated to maintain a positive social identity. In the present study, a group
of vacation trip companions can be identified as an important social category for college
students. Thus, this group is one form of social identity. It is called “Vacation Friends”
social identity. This social group is a social construction arising from friendship groups
who vacation together. This group receives the most significant attention from college
students in their travel decision making. College students are likely to attend only to the
“friendship group” that is especially self-relevant, and it is this self—relevant social

category that constitutes an individual’s social identity.

Multidimensional social identity. An important feature of Social Identify Theory
is that one achieves social identity through self-awareness of one’s membership in a
group, emotional attachment to a group, and evaluative significance of this membership
(Tajfel, 1978). Ellemers, Kortekaas, and Ouwerkerk (1999, p. 372) recently proposed
three components of social identity: “a cognitive component (a cognitive awareness of
one’s membership in a social group — self-categorization), an emotional component (a
sense of emotional involvement with the group — affective commitment), and an
evaluative component (a positive and negative value connotation attached to this group
membership — group-based self-esteem).”Cognitive Social Identify Theory postulates a
cognitive categorization process where similarities between self and group members are
emphasized as comparisons of dissimilarities with nonmembers (Hogg, I992). Emotional
social identity, which refers to a feeling of affective commitment, posits that those who
belong to the same social group may show differential affective responses, depending on
the extent to which they feel affectively committed to that group (Ellemers et al., 1999).
Finally, the evaluative component of social identity arises from evaluations of self-worth
derived from membership (Ellemers et al., 1999). Group-based self-esteem has been
found to promote actions that produce in-group welfare (e.g., Long & Spears, 1997).
With respect to construct validity, there is an argument between two different factor
structures: 1) first-order three-factor, and 2) second-order three-factor of social identity.
For example, Ellemers et al. (1999) proposed social identity as a first-order three-factor
structure. Besides viewing social identity as a first-order three factor model, comprising
cognitive, emotional, and evaluative social identity, it seems reasonable to consider

overall social identity as a common high-order factor. Recent previous research has

supported a second-order three-factor structure for social identity (Bagozzi & Dholakia,
2002; Bagozzi & Dholakia, 2006; Bagozzi & Lee, 2002; Dholakia et al., 2004). Thus, in
this study, social identity was conceptualized as having a second-order three-factor
structure.

Applications of social identity in literature. Recently, marketing researchers have
used this theory to explain consumer behavior under social situations. The link between
social identity and decision-making / judgments has been investigated in numerous
consumer domains, including small group brand communities (Bagozzi & Dholakia,
2006), the persuasiveness of spokespersons (Deshpande & Stayman, 1994), media use
(Saegert, Hoover, & Hilger, 1985), brand loyalty (Oliver, 1999), brand equity for service
marketers and development planners (Underwood, Bond, & Baer, 2001), virtual
community participation (Baggozzi & Dholakia, 2002; Dholakia, Bagozzi & Pearo,
2004), consumer self-importance (Reed 11, 2004) and information-processing tendencies
(Meyers-Levy & Sternthal, 1991). These research studies have clearly demonstrated that
classifying consumers based on some social identity leads to differences in the
descriptions of how consumers behave or respond to marketing stimuli.

In summary, Social Identify Theory provides an important theoretical background
for this present study predicting a college student’s desire and We-intention to take a
vacation trip with a group of friends (e.g., group behavior). For the friendship groups
investigated in the present research, it was expected that social identity influences
intentions to participate with group members in the shared activity of traveling together

on vacations.

21

An Alternative View of Intention: We-intention

The intention concept has been commonly studied by social psychologists.
Consumer behavior researchers in the field oftourism also have used the concept of
behavioral intention ofthe targeted or purposive behavior. According to the Theory of
Planned Behavior (Ajzen, 1991), an individual’s openly stated intention to act is the most
proximal predictor of behavior. This type of behavioral intention refers to the individual
or personal intention, which can be used to predict an individual’s immediate concept of
real behavior. Eagly and Chaiken (1993, p. 168) defined such a personal intention as a
“person’s motivation in the sense of his or her conscious plan to exert effort to carry out a
behavior.”

Taking a vacation trip with a group of friends is a group-based behavior. When
participating in such a group behavior, individuals are guided by collective group
intentions, rather than by individual or personal intentions. The concept of personal
intention should be differentiated from group or collective intentions, which can be used
to predict group behavior. In the literature, group intentions are known as “We-
intentions” (Tuomela, 1995) and “Share-intentions” (Bratman, I997). Tuomela (1995, p.
129) defines “We-intention” as a “commitment of an individual to participate in joint
action and involves an implicit or explicit agreement between the participants to engage
in thatjoint action. Bratman (I997) expressed a “shared intention” in the form “I intend
that we act.” Many individual actions occur as a form of social action. When two people
interact, neither person’s activities can be accounted as wholly their own (Nowak &
Vallacher, 1998). Many individual actions are best characterized in collective rather than

purely individual terms. Collective behavior is not simply the sum of individual actions,

“A

__i

and the difference resides in the intentions ofthe actors. When engaged in collective
activity, individuals are guided by collective intentions, sometimes called “We-
intentions” (Scarlc, 1990).

In this study, a college student’s intention to take a vacation trip with a group of
friends is conceptualized as a group intention which have been defined as “We-
intentions” (Tuomela, 1995) and “Shared-intentions” (Bratman, 1997). When a college
student intends to take a vacation trip with a group of friends, this is “We-intention” to
perform a group act, namely take a vacation trip together, rather than “I-intention,”
deciding to take a vacation trip by myself. Thus, when a college student intends to take a
vacation trip with a group of friends, he or she conceives of this activity not so much as
individuals performing personal acts that contribute to a group performance, but rather as
a group action in which he or she is a member ofthe group. Bagozzi and Lee (2002) have
reported that members within a group regard themselves as part of the social fabric of the
travel group in certain specific ways.

Tuomela (1995, p.129) asserts that We-intentions require individual commitment,
and commitment to provide mutual support in the sense that a member is “not only
committed to performing the pre—assigned part.... He or she is also committed to
furthering” thejoint action. While Tuomela (1995) and Bratman (1997) have provided
the foundations of social actions for predicting group behavior, Bagozzi (2000)
contributed to development of the measurement of the “We-intentions” concept. Bagozzi
(2000) conceptualized this type of group or collective intention as “intentional social
action” which, according to him, emphasizes that group members act intentionally (i.e.,

engage in purposive and goal-directed action that remains under the individual’s

volitional control), and that these actions have a collective basis in that both what is done
(a group-based behavior) and why it is done (i.e., members ofa groupjointly hold a
common goal and agree in a particular way to do their parts) to achieve the goal (Bagozzi,
2000). Further, empirical studies have provided support for the measurement model
developed by Bagozzi (2000) of We-intention in the consumer behavior field, such as
virtual community in communication environment (Bagozzi & Dholakia, 2002) and
friendship group behavior (Bagozzi & Lee, 2002).

In the present study the concept of We-intention (i.e. group intention), advocated
by previous researchers (e.g., Bagozzi, 2000; Bagozzi & Dholakia, 2002; Bagozzi & Lee,
2002; Bratman, I999; Tuomela; 1995), is tested in the context oftaking a vacation trip

with a group of friends among college students.

The Conceptual Model

The proposed model incorporates both individual factors (attitude and perceived
behavioral control) and social factors (subjective norm and social identity) as antecedents
that influence a college student’s desire and/or We-intention to take a vacation trip with a
group of friends. The main outcome variable is a college student’s We—intention,
indicating a college student’s intention to take a vacation trip with a group of friends. The
proposed model is presented in Figure I. It is not designed for predicting an individual
intention, but rather for predicting collective intention or an individual’s group intention.
Social identity is conceptualized as a second-order three-factor comprising cognitive,
emotional and evaluative components of social identity.

As shown in Figure 1, personal factors comprising attitude and perceived

behavioral control and social factors comprising subjective norm and social identity

influence a college student’s desire to take a vacation trip with a group of friends
positively and directly. In turn, a college student’s desire to take a vacation trip with a
group of friends influences a college student’s We-intention positively and directly. In
accordance with the Model of Goal-directed Behavior (MGD), it should be noted that the
desire to take a vacation trip with a group of friends is a central factor to mediate
relationships between these identified antecedents (attitude, perceived behavioral control,
subjective norms, and social identity) and the outcome variable (We-intention). More
specifically, desire to take a vacation trip with a group of friends plays both full mediator
and partial mediator roles in this proposed model. Perceived behavioral control and social
identity have both indirect and direct effects through desire on We-intention to take a
vacation trip with a group of friends. Group size is expected to moderate the relationships

between social identity and desire, and social identity and We-intention.

 
   

  

Perceived
behavioral control

  
 
 

H7: +
—>

    

Subjective norms

Social identity

  

Figure 1. Proposed Conceptual Model.

25

H ypothesized Relationships

The conceptual model in Figure l posits the following nine main hypotheses:
__l_: Perceived behavioral control is expected to be positively associated with a
college student’s desire to take a vacation trip with a group of friends.
H2: Attitude is expected to be positively associated with a college student’s desire to
take a vacation trip with a group of friends.
_3: Subjective norms are expected to be positively associated with a college
student’s desire to take a vacation trip with a group of friends.
M1 Social identity is expected to be positively associated with a college student’s
desire to take a vacation trip with a group of friends.
__5_: Perceived behavioral control is expected to be positively associated with a
college student’s We-intention to take a vacation trip with a group of friends.
HQ: Social identity is expected to be positively associated with a college student’s
We-intention to take a vacation trip with a group of friends.
fl: Desire is expected to be positively associated with a college student’s We-
intention to take a vacation trip with a group of friends.
_H_8: The smaller the size of group, the more social identity affects a college student’s
desire to take a vacation trip with a group of friends.

_9: The smaller the size of group, the more social identity affects a college student’s

We-intention to take a vacation trip with a group of friends.

26

Moderating Effect of Group Size

As posited in H8 and H 9, the magnitudes of the relatidnships 1) between social
identity and desire, and 2) between social identity and We-intention are expected to be
different, depending on the size ofthe travel party. That is, group size is introduced to
identify the moderating effect on these two relationships in the proposed model. Social
impact theory (Latane, 1981) has explained the effect of group size. Latane (1981)
proposed that the amount of impact an individual experiences as a result of a group
setting will be a function ofthe number, strength, and immediacy ofthe sources of
impact. The number of group members is the attribute that has received the most
attention in this regard.

There has been substantial empirical research on the effects of relative group size
on ingroup identification and ingroup favoritism (e.g. Mummendey & Simon, 1989;
Sachdev & Bourhis, 1984; Simon & Brown, 1987). However, previous researchers
(McGuire, McGuire, Child, & Gujioka, I978; McGuire & Padawer-Singer, I976;
Ellemers et al. (1999) have found that small group size results in stronger in-group
identification than large group size because it is more salient or distinctive than majority
group memberships, and not because membership in a minority group is more attractive.
Simon and Hamilton (1994) indicated that membership in a small or distinctive group
implies a relatively large overlap between collective self and individual self. Brewer
(1991) also supported this notion by explaining that, compared to inclusion in a large
group, small group membership offers better opportunity to balance the need to retain
some sense of individuality with the need to belong to a group, which should result in a

greater readiness to perceive or define oneself as a group member. Based on this

27

literature relating to group size and social influences, it is reasonable to predict a
moderating effect ofgroup size in explaining the relative contribution of social identity

on the college student’s desire and We-intention to take a vacation trip with a group of

friends.

28

CHAPTER 3

METHOD

Overview

A cross-sectional survey was used in this study for empirical, explanatory and for
descriptive purposes. The focus ofthis chapter is on : 1) how the sample was defined, 2)
how data were collected, 3) how factors in the proposed model were operationalized
based on validated scales from other studies, and 4) the statistical analyses that were
employed.

Data Collection

Sample

The purpose of this study was to identity college students who take vacation trips
with friends, and to investigate the factors that underlie college students’ intentions to
take a vacation trip with friends Thus, the target sample was those who have taken a
vacation trip with one friend or more at least once since becoming college students.
Data Collection Procedures

Pre-test. Before data collection, a two-staged pilot test was conducted. The first
stage was designed to confirm the face validity of measures and scales. The measures and
scales used in this study were based on prior validated instruments drawn from the
existing relevant literature. Personal interviews were conducted with doctoral students
and professors majoring in tourism, communication and education to check for the face

validity of the instruments that were developed for use in this study.

"n

The second-stage involved evaluation of survey questions via a pre-test process.
Thirty-one college students were selected and asked to provide feedback regarding: I) the
on-line survey’s graphic design, 2) the clarity ofthe questions asked, and 3) the amount
of time required to complete the questionnaire. They completed the questionnaire in the
researcher’s presence. Any questions and/or problems concerning format and wording
were recorded, and necessary changes were subsequently made. The feedback collected
from this pretest was used to refine the survey instrument for the main survey and
inappropriate items were eliminated. Each student who participated in the pretest
received compensation in the amount of US$12-$15 for time spent, i.e., approximately
45-60 minutes. The above steps insured that the questionnaire was satisfactory in terms
of content and face validity.

Survey administrations. At the beginning of spring semester 2006, prior to data
collection, instructors teaching courses at 1 I universities across the Eastem- Midwestern-
and Southern USA were contacted. From nine of those universities, a total of 21
instructors agreed to introduce the survey to their students emphasizing that participation
was voluntary. Some distributed on-line invitation letters to their students to participate in
the survey. These letters included a direct link to the questionnaire. Others directly
distributed invitation letters to their students together with paper-based questionnaires. As
an incentive, all participants were given the opportunity to enter a drawing from which
six participants each would be awarded a $50 gift certificate. Each participant was asked
to voluntarily provide his or her email address in order to participate in the drawing.
Fifteen of the 21 instructors also awarded course credits to students who successfully

completed the survey.

30

Most students completed the online survey because, in this population, the
Internet is a popular mode of communications for both educational and social purposes.
The paper-based surveys were distributed and completed in class under instructors’
supervision. Data collection was conducted for two weeks in 2006 (February 21 to March
10) which included this population’s spring break, for them a major vacation.

Since the focus ofthis study was on all US college students, the ideal sample to
use would have been a random sample ofthe total population of all college students.
Drawing such a sample was not possible, therefore a convenience sample was used which
limits generalizability ofthe study. Since the focus ofthis study is primarily on
relationships between variables, this type of research question is typically less vulnerable
to generalizability problems (Burnett & Dunne, 1986; Sears, 1986).

Questionnaire Format

The survey instrument consists of five main sections: past vacation experience
with friends, group composition on past vacation trips with friends, main variables for
measuring factors in the proposed model, trip-related variables for the next planned
vacation, and socio-demographics. A copy ofthe survey instrument is provided in
Appendix A.

Section I included questions regarding past experience in or on vacation trips
with or without friends. In section 2 ofthe questionnaire, participants were directed to
“IMAGINE that you and your FRIEND/S (one friend or more) are deciding (or have
decided) where to travel together on your coming vacation.” Participants were then asked
to “Now, in your mind, picture that friend or those friends, while you answer the

following questions:” This procedure is similar to determining one’s intention to have

31

lunch with a group of friends as reported, in Bagozzi and Lee’s study of intentional social
actions. Bagozzi and Lee (2002, p.235) explained that “these instructions were designed
to focus attention as vividly as possible on a specific group and its members, and to
provide a target format to which certain questions referred later in the questionnaire” (for
the main variables in their proposed model). After that, questions were included
regarding past experiences on a vacation trip with friends. Number of friends on this trip
(group size), their ages, type of relationships, and length oftheir relationships also were
asked of respondents for identifying past vacation group composition.

In section 3, questions to measure the main factors in the proposed model were
included. The format used in section 2 was also applied to section 3 questions by
explaining “The following questions relate to your vacation trip with friend/s within this
year. In this section, the friend/s you listed refers to the same friend/s you identified at the
beginning ofthe survey.”

Section 4 included the following questions about the next planned vacation trip:
composition of the travel party, main destinations, length of stay, projected expenditures,
sources of funding, planned activities, information sources, lodging type, transportation,
and trip organization. In the fifth and final section of the questionnaire, a series of socio-
demographic questions was included.

Sample Selection

The survey population comprised college students enrolled in selected American

universities. It included college students who had or had not taken a vacation trip. A sub

sample of them was selected for the main analysis.

’Ifll
-I‘U

As shown in shadowed part of Figure 2, the subsample consisted only of
respondents who have taken a vacation trip with one or more friends at least once since
becoming college students. Respondents whose trips were with only a boyfriend /
girlfriend were also excluded from the subsample used for detailed analyses. This

minimized the potential bias resulting from dyadic relations that may not be group acts.

 

Travelers with one or more friends,
but not boy or girl friend

 

 

 

 

Travelers

 

 

 

 

All respondents Other travelers

 

 

 

 

 

 

 

Non-travelers

 

 

Figure 2. Depiction ofthe process used to select the primary study population.

Measurements
Components in the proposed conceptual model depicted in Figure l were derived
from respondents’ self-reported perceptions in response to problem stimuli. Perceptions
were gathered via on-line and paper questionnaires. Factors for use in the proposed model
were assessed using multiple item measures. All scales used were based on previous
empirical studies using the model of goal-directed behavior, social identity theory, and a
view of we-intention (Ajzen & Driver, 1992; Ajzen, I991; Bagozzi; 2000; Bagozzi &

Dholakia, 2002; Bagozzi, Dholakia, & Mookerjee, 2006; Perugini & Bagozzi, 2001). All

items were modified to reflect the context of taking a vacation trip with friends identified
by respondents.

The proposed model included nine factors which were unobserved. These
included social identity and its three dimensions (i.e., cognitive, affective, and evaluative
social identity), together with attitude, subjective norms, perceived behavioral control,
desire, and we—intention. It also included 34 observed variables associated with those
factors. These are described in detail in Table I.

Attitudes. Attitudes toward taking a vacation trip with a group of friends were
measured with five seven-point semantic differential scales based upon items from Ajzen
(1991). Participants were asked “How do you personally feel about taking a vacation trip
with the friends you listed within this year.” The following five adjective pairs were
provided in this series: negative-positive. unpleasant-pleasant, boring-fun, unenjoyable-

enjoyable, and unfavorable-favorable.

Perceived Behavioral Control. Four items were measured, with four seven-point
rating scales from Ajzen (1991). These questions consisted of: How much control do you
have over taking a vacation trip with the group of friends you identified above?” (I: no
control to 7: total control); “For me to engage in taking a vacation trip with a group of
friends I identified is (1: easy to 7: difficult)”; “I believe I have the resources required to
travel with a group of friends I identified” (1: false to 7: true); “The number of events
outside my control that can prevent me from taking a vacation trip with a group of friends

identified above are (1: numerous to 7: very few).”

Table 1. Description of factors and observed variables in the proposed model.

 

Observed
__ Factors vafl ables

Survey questions

Scale

 

A" A]
A2
A3
A4
A5

SN“ SN]

8N2

5N4
PBCc PBCI
PBC2
PBC3
PBC4

csr‘ csn

C812

C813
C814
ASIc ASII
ASIZ
A513

A514

Taking a vacation trip with my friends is
negative or positive.

Taking a vacation trip with my friends is
unpleasant or pleasant.

Taking a vacation trip with friends is boring or
fun.

Taking a vacation trip with friends is
unenjoyable or enjoyable.

Taking a vacation trip with friends is
unfavorable or favorable.

Most people important in your life think you
should take a vacation trip with your friend/s
you listed within this year.

People in your life whose opinions you value
would approve ofyour taking a vacation trip
with the friend/s you listed within this year.

It is expected of me that I take this vacation trip
with the friend/s I listed within this year.

My most important friends take a vacation trip
with friend/s within this year

How much control do you believe you have
over taking a vacation trip with your friend/s?
For you to take a vacation trip, with the friend/s
you listed within this year, is

I believe I have the resources required to take
this vacation trip with my friend/s

It is mostly up to me whether or not I take a
vacation trip with the friend/s I listed

Please indicate the degree to which your self-
image overlaps with the identity ofthe group of
friends you listed above, as you perceive it

I am like other members ofthe group 1
indicated.

1 identify with other members ofthe group I
indicated.

The group I indicated is an important reflection
of who I am.

My behavior often depends on how I feel others
wish me to behave.

How attached are you to the group you
indicated?

1 like being a member ofthe group 1 indicated.

1 would rather belong to the other groups.

I:extremely negative to 7:
extremely positive
I:extremely unpleasant to 7:
extremely pleasant
Izextremely boring to 7:
extreme fun

lzextremely unenjoyable to
7: extremely enjoyable
I:extremely unfavorable to 7:
extremely favorable
Izshould not to 7: should

Izstrongly disapprove to 7:
strongly approve

lzextremely unlikely to 7:
extremely likely
I:definitely false to 7:
definitely true

Izno control to 7: complete
control

lzextremely difficult to 7:
extremely easy

I:definitely false to 7:
definitely true

Izstrongly disagree to 7:
strongly agree

lznot at all overlapped to 7:
overlapped very much

I:does not describe me at all
to 7: describes me very well

I:does not describe me at all
to 7: describes me very well
I:does not describe me at all
to 7: describes me very well
I:definitely false to 7:
definitely true

Iznot at all attached to 7:
attached very much

I:does not describe me at all
to 7: describes me very well
I:does not describe me at all
to 7: describes me very well

Note: aAttitude: overall evaluative response of group vacation trip,13 Subjective norm: the perceived social
pressure to perform group vacation trip, c Perceived behavioral control: perception of how easy or difficult
it is to perform group vacation trip, d Cognitive social identity refers to a cognitive awareness of one’s
membership in the vacation trip group, ° Affective social identity: a sense of emotional involvement with

the vacation trip group.

35

Table 1. Description of factors and observed variables in hypothesized model (continued).

 

 

 

 

 

 

Observed
_ _Factor variable Survey questions Scale
ESIf ESII I am a valuable member ofthe group 1 indicated. I:does not describe me at all
to 7: describes me very well
E512 I am an important member ofthe group I I:does not describe me at all
indicated. to 7: describes me very well
ESI3 I would rather not tell that I belong to the group I I:does not describe me at all
indicated. to 7: describes me very well
ESI4 I think the group I indicated has little to be proud I:does not describe me at all
of. to 7: describes me very well
Dg DI Your desire for taking a vacation trip with your l:no desire at all to 7: strong
friend/s you listed can be described as... desire
DZ I desire to take a vacation trip with my friend/s I lzvery strongly disagree to 7:
listed. very strongly agree
D3 I want to take a vacation trip with my friend/'5 I I:does not describe me at all
listed. to 7: describes me very well
D4 I feel an urge or need to take a vacation trip with I:does not describe me at all
my friend/s listed. to 7: describes me very well
WIh W11 We ( l and my friend/s I listed) intend to take a lzextremely unlikely to 7:
vacation trip together within this year. extremely likely
WIZ We (1 and my friend/s I listed) will try to take a I:definitely false to 7:
vacation trip together within this year. definitely true
W13 I intended that I and my friend/s I listed take a lzvery strongly disagree to 7:
vacation trip together within this year. very strongly agree
W14 We (1 and my friend/s I listed) plan to take a Izvery strongly disagree to 7:

vacation trip together within this year.

very strongly agree

 

Notes: rEvaluative social identity: a value attached to vacation trip group membership. 3 Desire: motivation
to take vacation trip with friends, h We-Intention: a commitment to participate in joint vacation trip with

friends

Subjective norms. Subjective norms were measured by four items, using seven-
point rating scales from Ajzen (1991). These questions consisted of: “Most people who
are important to me approve / disapprove of my taking a vacation trip with the group of
friends I identified above.” (I: disapprove to 7: approve); “Most people who are
important in my life think I should / should not travel with the group of friends identified
above.” (I: should not travel to 7: should travel); “Most friends I know would support/
not support my travel with the group of friends identified above.” ( I: not support to 7:
support); “People who are important to me would like me to travel with a group of

friends.” (I: unlikely to 7: likely).

36

Social identity. All three social identity factors were measured with seven-point
rating scales. The cognitive component of social identity (i.e., self-categorization) was
measured by four items, modified from Bergami and Bagozzi (2000). An example is:
“Please indicate the degree to which your self-image overlaps with the identity ofthe
group of friends you identified above as you perceive it.” (I: not at all, to very much).
Affective social identity (i.e., emotional commitment) was measured by four items based
on research on commitment (e.g., Allen & Meyer 1996). An example is: “How attached
are you to the group you identified above?” (I: not at all attached, to 7: strongly attached).
Evaluative social identity (i.e., group self-esteem) was measured by four seven-point
items, modified by Luhtanen and Crocker (1992). An example is: “I am a valuable
member of the group 1 indicated above.” (1: describes me inaccurately to 7: describes me

well). All items and scales are presented in Table l.

Desire. Desire was measured by four-items, modified from Perugini and Bagozzi
(2001) and Bagozzi and Dholakia (2002), using a seven-point Likert scale, ranging from
strongly disagree (I) to strongly agree (7). An example is “I desire to travel in the next

six months with the group of friends I mentioned above.”

We-Intentions. Four items measured wc-intentions, using four seven-point Likert
scales, modified from other studies (e.g., Bagozzi & Dholakia, 2002; Bagozzi & Lee,
2002). An example is: “I intend that our group (i.e., I and the group of friends I identified
above) take a vacation trip together within this year.” using a seven-point scale ranging
from very strongly disagree (I) to very strongly agree (7).

Moderating variable. Group size was asked as a potential moderating variable in

the model. “In your mind, and including yourself, how many persons might be in your

37

group for this vacation trip?” A median-split approach was used to separate the data set
into two groups. As a result, one group contained the study subjects scoring high (over
median) on group size, while the other consisted ofthose subjects scoring low(below
median).
The Proposed Model

This study was conducted to examine the causal relationships among the factors
of attitude, subjective norms, perceived behavioral control, social identity, desire and we-
intention. A path diagram ofthe full structural model including observed variables is
presented in Figure 3. A path diagram provides a pictorial portrayal of relationships of
observed variables (in rectangles), factors - i.e., unobserved or latent variables - (in ovals),
and associated error terms (in circles) (Hair, Anderson, Tatham, & Black, 1998). Straight
arrows depict impacts of exogenous variables — i.e., independent variables - on
endogenous variables - i.e., dependent variables - and curved arrows depict correlation
among variables. A list of factors in the proposed model and their corresponding

observed variables are presented in Table I.

Data Analysis
A two-stage data analysis procedure was employed. In stage one, the descriptive
analysis aimed to profile college students who take a vacation trip with friends in terms
of past vacation trip experience, future vacation planning and socio-demographics. Stage
two used structural equation modeling (SEM) to assess construct measures in the
proposed model, in order to examine the causal relationships and to examine moderating

effects.

38

Descriptive Analysis

Descriptive analyses were performed to: I) estimate the potential market size of
college students who take a vacation trip with friends, 2) identify group composition on
past vacation trips, 3) determine vacation trip characteristics of college students, and 4)

describe the socio-demographic characteristics of respondents. SPSS 13.0 was adopted as

the tool for this analysis.

Perceived
behavioral
control

 

 

    
     
 

 

 
  
   
 

   
    

        

 

l_m_ll_b2_JL_I2.LLm_J
'\ WII
We-Intention
Int!
Social identity
5N4
Cognitive Affective Evaluative
social identit .. social identity social identit
l csn II csn II (25871 csm IFAsn 11 A812 ll A813 11 A314 11 csn 11 £512 11 5313 II 5514 I

 

Figure 3. Proposed hypothetical model of intention to take a vacation trip with friends in
structural equation modeling (SEM) format.

39

Structural Equation Modeling (SEM)

Structural equation modeling (SEM) is a multivariate statistical technique used to
test for causal relations among latent factors. SEM is an analytic approach that combines
features of confirmatory factor analysis and path analysis. It is possible in SEM to specify
a model that has a measurement component (like a factor model) and a structural
component (like a path model). Since SEM encourages confirmatory rather than
exploratory modeling, it is suited to theory testing rather than theory development. Thus,
SEM is the most appropriate tool for this study since its focus is on confirming the fit of
existing theory to the travel behavior of college students.

This study used a two-step procedure proposed by Anderson and Gerbing (1988).
The first step involved evaluating and refining the measurement model, while the second
step tested the estimation of the structural model and hypotheses. The AMOS 5.0
statistical package program was used for SEM. The maximum likelihood estimation
procedure (ML) was used as a parameter estimation method in SEM. The ML estimation
procedure allows all model parameters to be simultaneously estimated.

Sample size for SEM. Sample size is important to estimate and interpret SEM
results. Small sample sizes have the greatest potential to increase the likelihood of
specification errors and produce biased goodness of fit indices, a critical element in
maximum likelihood estimation. At a minimum to ensure appropriate use of maximum
likelihood estimation, the sample size should be 100. Sample size should be sufficient to
include five observations for each estimated parameter (Bentler, 1995). Optimal sample
size is about 200 according to Kline (I998). The number of subjects obtained for this

study met both suggested criteria (N = 452).

J r

Model fit indices. SEM provides some measures of the overall fit of the
measurement model and the structural model. The chi-square statistic is commonly used
to evaluate model fit, but this is known to be sensitive to sample size (Kline, 1998). Thus,
this study used multiple indices including the chi-square statistic adjusted for the degrees
of freedom (xz/df), the comparative fit index (CFI), the nonnormed fit index (NNFI), and
the root mean square error of approximation (RMSEA) in assessing the model fit
following suggestions by Kline (1998). General rules of thumb for model fit are that
xz/dfshould be less than 3 fit indices, such as CPI and NNFI, should be greater than .90
and, RMSEA should not be larger than .05 (Kline, 1998).

Furthermore, 12 difference was used when comparing fit of an adequate model
and another competing model. Two competing models were evaluated using a
significance test based on difference in x2 and degrees of freedom between the two.
Measurement Model

The measurement model specifies how factors are measured in terms of the
observed variables, and factors describe the measurement properties of the observed
variables. The measurement model was evaluated with confirmatory factor analysis
(CFA). During measurement model testing, CFA estimates only associations among
factors, not direct causal effects. In other words, causal relationships among factors in the
proposed structural model are not a product of the measurement model. The measurement
model including 9 factors and 34 observed variables was tested and evaluated through the

model specification procedure suggested by Joreskog and Sbrbom (1993)

Model specification procedure. First, the measurement model was specified based

on previous theoretical and empirical studies. The proposed model was tested and

41

evaluated using overall fit indices and parameters estimated by CFA. Second, when the
proposed model was found to be a poor fit to the data modification procedures were
conducted to identify candidate observed variables for deletion from the measurement
model. These procedures are customarily used to maximize model fit (Bolen 1989; Byme
I994).

The model modification procedures employed observed variables that had low
factor loading, significant cross-loadings, and large residuals using three measures:
standardized factor loadings, modification index (MI) in regression weight, and
modification index (MI) in covariance between two error term variables respectively.
AMOS results provided these three measures. The M1 test objective was to determine
whether, in subsequent runs, better models resulted, with certain parameters specified as
free rather than fixed. As the minimum cut-off, it was suggested that a standardized factor
loading should be greater than .50, and each MI should not exceed 100 (Kline, 1998).
After removing observed variables in accordance with these three criteria (factor loading
and two kinds of M15), model modifications were continued until all parameters
estimated and overall fit measures (12 difference, CFI, NNFI, and RMSEA) werejudged
statistically and substantively satisfactory.

Furthermore, all observed variables were evaluated based on the measurement
model with observed variables retained through model specification procedures,
convergent validity, discriminant validity, and reliability.

Structural Model and Hypothesis Testing
Once the appropriate measurement model was obtained after revising it and

testing it against the competing measurement model, a full structural model was tested. In

42

assessing the path model, hypothesized relationships (relative magnitudes and direction
by path coefficients) were examined. Evaluation of a structural model is similar to
conducting a path analysis but with latent variables instead of observed variables. Path
analysis allows researchers to specify and test structural models that reflect a priori
assumptions about direct or indirect causal effects among observed variables.

Multiple Group Analysis

Multiple group analysis is used to evaluate a whole model across independent
samples, each of which represents one level of the categorical variable. This study used
the analytical strategy of Singh (1995) to examine the existence ofthe moderating effects
on the structural model in SEM.

First, for all moderating variables, a median-split approach was used to separate
the data set into two groups. As a result, one group contained the study subjects scoring
high on a respective variable, while the other consisted of those subjects scoring low.
Second, an unconstrained structural model was estimated, in which path coefficients were
allowed to vary across the cross-group datasets. Next, a fully constrained structural model
was estimated by requiring that all path coefficients are constrained to be equal for cross-
group datasets. The fully constrained structural model thus was based on the notion of
cross-group variance in model relationships. The )6 difference test was performed to
compare the unconstrained and fully constrained structural models. Finally, ifa
significant difference was found, the x2 difference test was used again to test for the
moderating effects of individual paths. However, )8 statistics for the unconstrained and
the partially constrained models are compared here. Partially constrained means that only

target path coefficients are set to be equal for cross-group datasets. If the chi-square

43

difference is significant, the relative magnitude oftarget paths can be compared between

groups.

44

CHAPTER 4

RESULTS

Overview

Study results are presented in this chapter. First, survey response information is
presented. Second, descriptive statistics are reported to include: sample characteristics,
percent of past vacation experience with friends, group composition of college students
who have taken a vacation trip with friends, and characteristics of planned future
vacations with friends. Third, normality test results are presented prior to structural
equation modeling (SEM). Fourth, confirmatory factor analysis results are presented to
evaluate validation of study measures used in subsequent SEM. Fifih, hypotheses-testing
results are discussed to assess the quality of the proposed structural model derived
employing SEM analysis procedures. Finally, multiple group analysis results are

presented to identify possible moderating effects in the proposed structural model.

Survey Response
Data Collection
As can be seen in Table 2, the survey was distributed to college students who
enrolled in one of 21 courses at nine universities. Completed surveys were returned from
758 students. Of those, 61 incomplete or duplicate responses were identified and deleted.

Thus. 697 responses were kept in the final sample (on-line 558, in-class 139) for analysis.

45

Table 2. Summary ofdata collection: total sample. returned sample, invalid sample, valid
sample.

 

 

 

__ ____9_n-.lins$.twL_-_ln_-c_l§§_s_svrvey Tmal __
Returned sample 603 155 758
Invalid sample a 45 I6 61
Valid sample 558 I39 697

 

Note: a Invalid sample refers to sample with incomplete or double responses.

Characteristics of Survey Respondents

Institutions in which survey respondents were enrolled are presented in Table 3.
Michigan State University (31.6%) and the University of Missouri at St. Louis (25.6%)
were the main institutions in which they were enrolled. The distribution oftheir academic
majors was as follows: Tourism/Hospitality management 36.1%, Liberal
Arts/Communication 25.3% and Business 15.3%, as can be seen in Table 4. While
convenience sampling was employed in this study, considerable diversity was achieved

across the institutions and academic majors of respondents.

Table 3. Institutions in which survey respondents were enrolled (N = 697).

 

 

Category Descriptions Frequency Percent
Institutions
Michigan State University (MI) 221 31.6%
University of Missouri at St. Louis (MO) 179 25.6%
Niagara University (NY) 85 12.2%
University at Buffalo (NY) 63 9.0%
Pennsylvania State University (PA) 46 6.6%
Arkansas Tech University (AR) 38 5.4%
Ferris State University (MI) 32 4.6%
University of Minnesota (MN) 25 3.6%
Ohio University (OH) 10 1.4%

 

46

Table 4. Survey respondents’ academic majors (N = 697).

 

 

Category Descriptions Frequency Percent

Majors
Liberal Arts / Communication 177 25.3%
Tourism / Hospitality Management 253 36.1%
Business (Management, Accounting, Marketing) 107 15.3%
Social Science / Education 60 8.6%
Natural Science / Engineering / Medical 43 6.1%
Engineering 41 5.9%
Agriculture / Environment / Natural Resources 13 1.8%
Other 6 .9%

 

Study Subjects Selected from Survey Respondents

Survey respondents included not only college students who took a vacation trip
with a group of friends, the targeted study population, but also those who did not. The
sample screening procedure resulted in a final sample of 452 college students,
representing 67% ofthe 697 survey respondents, who took a vacation trip with a group of
friends at least once while at college prior to when the survey was conducted.
Demographic Comparisons of Study Subjects to Other Subjects and the Total College
Population

Demographic characteristics of study subjects (N = 452) were compared with
subjects excluded in the full study sample (they are called “other subjects” below) and all
US four-year college students based on US census data to assess non-response bias and
generalizability ofthe sample respectively. Results ofthese comparisons are presented in
Tables 5 and 6.

Study subject’s (N = 452) genders are found to be almost equal, being 45.6%

male and 54.4% female. These percentages are similar to other subjects (43.1% vs.

47

56.9%) and US four year college students (45.5% vs. 54.5%). Study subjects’ ages were

20 to 24 years old (57%) followed by 19 or younger (29%) and 25 or older (15.3%).

Table 5. Demographic characteristics of study subjects and other subjects for all survey
respondents, study subjects and other subjects.

 

 

 

Democraphic . . Study 8 Other b Test statisticsC
variZbles Descriptions subjects subjects ,
(N = 452) (N = 245) x“ p
Gender
Male 45.6 % 43.1% x =.38 .538
Female 54.4 % 56.9%
Age
19 or younger 29.0% 29.3% X =.07 .966
20 to 24 55.7% 56.1%
25 or older 15.3% 14.6%
Academic year
Freshman 19.1% 19.8% x2=-17 .982
Sophomore 24.4% 23.2%
Junior 30.1% 32.1%
Senior 26.4% 25.9%
Ethnicity
White / Caucasian 77.1% 75.4% 12:1.5 .820
Asian / Pacific Islander 17.2% 16.8%
Black/ African 32% 3.1%
American
Spanish / Latino 2.5% 2.4%
Citizenship
U.S.A. 88.3% 89.1% x =.05 .997
Asia 9.5% 9.1%
Other American 1.8% 1.5%
nat1ons
Europe .5% .3%

 

Note: a Study subjects refer to respondents who have had vacation trip experience/s with a group
of friends since becoming college students. This sample was used for main analyses, testing the

i’

roposed model and developing a profile of vacation trip market of college students in this study.
Other subjects refers to respondents who had no vacation experiences with a group of friends

since becoming college students. Other subjects were excluded for main analyses in this study.
° All demographic variables in the above table exhibit no significant differences between the two

groups, at p<.05.

Table 6. Comparison of demographic characteristics of study subjects and all US four-
year college students.

 

 

Demographic Descriptions Study All US four year
variables subjects a college students b
(N = 452) (N = 9,664,000)
Gender
Male 45.6 % 45.5%
Female 54.4 % 54.5%
Age
19 or younger 29.0% 27.1%
20 to 24 55.7% 49.3%
25 or older 15.3% 23.6%
Academic year
First to second year 43.5% 45.6%
Third to fourth year 56.5% 54.4%
Ethnicity
White / Caucasian 77.1% 67.5%
Asian / Pacific Islander 17.2% 5.8%
Black / African American 3.2% 13.5%
Spanish / Latino 1.6% 11.2%
Other 2.5% 2.0%

 

Note: a Study subjects refer to respondents who have had vacation trip experience/s with a group
of friends since becoming college students. This sample was used for main analyses, testing the
proposed model and developing a profile of vacation trip market of college students in this study.
b Source: Four year college students in Table 10 “Type of College and Year Enrolled for College
Students 15 Years Old and Over, by Age, Sex, Attendance Status, Race, and Hispanic Origin:
October 2004”, US. Census Bureau, Current Population Survey, October, 2004.

There is no age difference between study subjects and other subjects. Compared to the
age of all US four year college students, study subjects are distributed as follows:
younger: 20 to 24 years 49.3%, 19 or younger 27.1%, and 25 or older 23.6%. Fifty-seven
percent of study subjects are academic juniors and seniors, and there are 43.5% freshmen
and sophomores. This result is almost the same as for other subjects (58.9% juniors and

seniors) as well as for all US four year college students (54.4%juniors and seniors).

Regarding ethnic background, the majority of study subjects are White/Caucasian

(77.1%), followed by Asian/Pacific Islander (17.2%), Black (3.2%), and Spanish/Latino
(2.5%). Compared to other subjects, there is no statistical difference in ethnic background.
However. the Asian/ Pacific Islander component of study subjects comprised a much
higher percentage than those of all US four year college students (17.2% vs. 5.8%) while
the percent of Black and Spanish study subjects is slightly lower than that of all US four
year college students (13.5% vs. 1 1.3%). The citizenship ofthe majority of study subjects
are American (88.3%), followed by Asian (9.5%), other South- Central- and North-
American nations’ citizens (1.8%) and from Europe (0.5%). Other subjects exhibited no
citizenship differences from study subjects. Unavailability of data precluded comparison
ofthe citizenship distribution of US four year college students and study subjects.

The subjects selected for this study (N = 452) have demographic characteristics
similar to those respondents who never took a vacation trip with a group of friends (N =
252) and to the total population ofUS college students. This supports concluding that
non-response bias is not due to any demographic differences and that generalizability of
results is unlikely to be problematic given the similar demographic profiles of subjects

and the overall US college population.

Profiling the Group Travel Market: Descriptive Analysis

Past Vacation Experience with Friends

As can be seen in Table 7, the majority of survey respondents (76.9%) have had
vacation experiences while at college (average: four trips). Their vacation trip/s was/were
taken mainly with friends (87.8%) or with a group of family/relatives (1 1.6%). Less than

one percent of these college students took a vacation trip alone.

50

Table 7. Past vacation trip experience of survey respondents.

 

 

Variables Descriptions Survey
respondents

Have you taken vacation trip/s since

becoming a college student? (N = 697)
Yes 76.9%
No 23.1%

Main companion/s in travel party (N = 530)
Friends 87.8%
Family/Relatives 11.6%
Alone .6%

Average number of vacation trips since becoming college student (N = 527)

Average number of vacation trips with friends since becoming

college student

4.1 trips (3.9 SD)

(N = 462)
3.3 trips (3.2 SD)

 

Group Composition on Past Vacation Trip with Friends

Information about with whom study subjects took a vacation trip in the past while

at college is presented in Table 8. Their friends’ genders were either mixed (57.1%) or

the same (42.9%). Fifty-five percent of study subjects responded that 21 to 23 years old

was the average age of their group of friends. Their average group size was 5.2 members.

Regarding types of relationships among friends, the majority of respondents were long-

term close friends (65.4%), followed by previous middle/high school friends (48.3%),

class mates (38.9%), and boyfriends or girlfriends (35.6%), roommates (33.3%), friends

from clubs or student organizations in college (33.1%). Relationship duration among

friends was 5.3 years on average. Friends included in travel parties mainly were longer

time friends rather than shorter time friends (74.7% > 2 years).

51

Table 8. Group composition on past vacation trips with a group of friends. a

 

 

Demographic variables Descriptions Study subjects
Gender composition (N = 446)
Same gender (men only, women only) 42.9 0/0
Mixed gender 57.1 %
Average age ofgroup members (N = 450)
18-20 33.6%
21-23 55.2%
24 or older I 1.2%
Average 21.3 (1.8 so b)
Group size (N = 449)
2 12.6%
3 I 1.5%
4 27.7%
5 16.6%
6 10.6%
7-10 16.6%
I 1 or more 4.4%
Average 5.2 (2.9 SD)
Relationship types among members ° (N = 447)
Long-term close friend 65.4%
Previous middle/high school friends 48.3%
Classmates 38.9%
Boyfriends/ girlfriends 35.6%
Roommates 33.3%
Friends from clubs / student
organizations in college 33.1%
Other 20.4%
Length of relationship among members (N = 446)
1-2 years 25.3%
3-4 years 24.1%
5-9 years 38.0%
10 or more years 12.5%
Average 5.3 (3.6 SD)

 

Note: a Vacation trip group refers to the group of friends with whom study subjects took vacation
trip/s in the past while at college. b SD in parentheses indicates standard deviation. c Percentages

add to more than 100% due to multiple responses.

52

Vacation Trip Plans for Coming Year
As can be seen in Table 9, the majority of survey respondents (78%) planned to
take a vacation trip in the coming year (2006). About 80% ofthose planning such a trip

indicated that friends would be their main companions on their planned trip.

Table 9. Vacation trip planning intentions of survey respondents.

 

 

Variables Descriptions A”
respondents
Do you plan to take a vacation trip/s within this year? (N = 697)
Yes 77.7%
No 22.3%
Main companion/s in travel party for next vacation trip (N = 540)
Friends 80.2%
Alone 2.8%
Family/Relatives 16.6%
Other 0.4%

 

Characteristics of Planned Trips with a Group of Friends

In this section, a trip-related profile for future vacations is presented. It is
summarized in Table 10. In terms of timing of the planned trip, more than half the target
sample (55%) planned to take a vacation trip with a group of friends during their summer
vacation, followed by spring break (36%), winter break (2.0%), and other holidays
(0.9%). The reported average planned length of stay away from home was 6.7 days, with
5.8 days at their main destinations. As the main destination city or place, Las Vegas
ranked as the top city (5.9%), followed by Cancun (4.7%), Chicago (3.6%), New York
City (3.4%), Miami (3.2%), Orlando (2.9%), and San Francisco (2.5%). Regarding the
main destination states or countries, Florida ranked first (18.8%), Mexico second (7.1%).

followed by California (6.3%), Nevada (5.4%) and New York State (4.8%). Three-

53

fourths ofthe students planned to visit domestic (75.1%) rather than foreign destinations
(24.9%). Among foreign destinations, Mexico ranked as the top country (7.1%), followed
by Europe (6.1%), Canada (5.2%), Caribbean countries (4.0%), and Asia (2.5%). The
majority of study subjects (61.5%) organized their own trips, i.e., by themselves or with
several friends, rather than purchasing commercially organized trips such as package
tours. Over 33% responded that pleasure activity was the main purpose oftheir vacation
trips. Entertainment and recreation activity ranked second (21.7%) and third (14.8%)
respectively. The primary source of information used was previous experience with the
destination (34.2%). Recommendation of friends or colleagues (29.0%) was also an
important factor in planning vacation trips, indicating the power of past experience and
word of mouth in travel decisions. Internet search was the most important secondary
information source used (27.6%).

Study subjects’ main lodging type used was hotel/motel (52.5%). Over half
(54.0%) of the study subjects used commercial air as transportation to their main
destination. Another 39% used private auto/truck. Rental car was an important
transportation mode at the main destination (24.0%). Individual expenditures on vacation
trips averaged $719 with a standard deviation of $626. Approximately one-fourth
(28.8%) of study subjects responded that they would spend over $1000 for their next
vacation trip. A majority of study subjects planned to utilize personal employment
earnings (69.4%) and savings (58.6%) to pay for their trips. Parents or relatives were also
important funding sources for their next vacation trip (49.1%). The average number of
activities participated in on their trips was 6.8, and shopping was their most popular

activity (71.4%). More than half the study subjects planned to participate in city/urban

54

sightseeing (56.8%), visit a unique or themed restaurant/bar (56.4%), or participate in
beach activities (51.2%). Attending social events such as parties (49.8%), outdoor
recreation (48.1%) and water sports (44.0%) were also reported to be preferred activities

during vacations.

Table 10. Characteristics of planned future vacation trips with a group friends.

 

Trip-planning variables

 

Time of planned trip
Summer vacation
Spring break
Winter break
Other holidays
Other

Number ofnights away from home
1-3 night/s
4-6 nights
7-9 nights
10 or more nights

Average number ofnights away from home
Average number of nights spent at main destination

Main destination by city or place b
Las Vegas
Cancun
Chicago
New York City
Miami
Orlando
San Francisco

Main destination by state or country
C

Florida

Mexico

California

Nevada

New York State
Main destination by country

USA

Mexico

Europe

Canada

Caribbean countries

Asian countries

(N = 438)
55.0%
36.0%

2.0%
0.9%
6.1%

(N = 443)
17.5%
43.2%
28.0%
11.3%

(N = 448)
6.7 (5.5 81):)

(N = 448)
5.8 (4.7 30)

(N = 446)
5.9%
4.7%
3.6%
3.4%
3.2%
2.9%
2.5%

(N = 446)

18.8%
7.1%
6.3%
5.4%
4.8%

(N = 446)
75.1%
7.1%
6.1%
5.2%
4.0%
2.5%

Note: 3 SD in parenthesis indicates standard deviation. bTop seven cities or places which
respondents gave as main destinations. ° Top five US states or other countries which respondents

gave as main destinations.

56

Table 10. Characteristics of planned future vacation trips with a group of friends
(continued).

 

 

Trip-planning
variables
Trip organizer(s) (N = 441)
With several persons in a self-organized trip 61.5%
By yourselfor With one person 1n a self- 26.4%
orgamzed trip
With several persons in a commercially
. . 7.5%
orgamzed tr1p (package tour)
By yourselfor with one person in a
. . . 3.3%
commerc1ally orgamzed tr1p (package tour)
Other 1.2%
Main purpose of vacation trip (N = 451)
Pleasure activity 33.3%
Entertainment 21.7%
Recreation activity 14.8%
Visiting friends 14.8%
Sightseeing 3.4%
Visiting home or family 3.0%
Education activity 2.4%
Cultural activity 1.7%
Volunteering activity 0.6%
Other 4.1%
Primary information source used (N = 445)
Previous experience with the same destination 34.2%
Recommendations of friends or colleagues 29.0%
Internet 19.6%
Recommendations of parents or relatives 10.8%
Recommendations of travel agents 2.8%
Newspaper or magazine articles 1.3%
TV programs 0.9%
Other 1.5%
Secondary information source used (N = 445)
Internet 27.6%
Recommendations of friends or colleagues 24.4%
Recommendations of parents or relatives 18.0%
Previous experience with the same destination 15.5%
Recommendations of a travel agent 7.3%
Newspaper or magazine articles 2.7%
TV programs 2.1%
Other 2.3%

 

Table 10. Characteristics of planned future vacation trips with a group of friends
(continued).

 

Trip-planning variables

 

Lodging type used (N = 445)
Hotel/motel 52.5%
Private home 20.8%
Condo/time share 11.5%
Campground 8.1%
Cruise ship 2.3%
Youth hostel 2.3%
Dormitory 1 .9%
Bed & breakfast 0.4%
Organizational camp 0.0%
Other 0.2%
Transportation to main destination (N = 442)
Airplane 54.0%
Private auto/truck 39.0%
Rental car 3.7%
Bus/motor coach 2.1%
Ship/boat 0.8%
Train 0.4%
Transportation at main destination (N = 442)
Private auto/truck 44.9%
Rental car 24.0%
Bus/motor coach 9.9%
Train 5.7%
Taxi 4.4%
Ship/boat 4.2%
Airplane 2.9%
Subway 0.7%
Other 3.3%
Individual expenditures on trip (US$) (N = 449)
$199 or less I 1.6%
$200-399 23.3%
$400-599 19.5%
$600-799 l 1.8%
$800-999 5.0%
$1000 or more 28.8%
Average $719 (625.6 SD 3)

 

Note: “ SD in parenthesis indicates standard deviation. * Percentages add to more than 100% due
to multiple responses.

58

Table 10. Characteristics of planned future vacation trips with a group of friends
(continued).

 

Trip-planning variables

 

Main funding sources oftrip expenses 3 (N = 450)
Personal employment earnings 69.4%
Savings 58.6%
Parents or relatives 49.1%
Loan 3.9%
Grants or scholarship 2.9%
Other 2.7%

Average number of activities participated in (N = 442)

6.8 (3.6 SD b)

Activities participated in a (N = 442)
Shopping 71.4%
Sightseeing — City/Urban 56.8%
Visit a unique or themed restaurant / bar 56 4%
(e.g., wild west, rock ’n roll, rainforest) °
Beach activities 51.2%
Attend social events (party) 49.8%
Outdoor recreation (e.g., camping, hiking,

. . 48.1%
hunting, golf, tennis)
Participate in water sports (e.g., boating,
. . . . . 44.0%
canoeing, rafting,jet skinning etc.)
Attend performing arts events (e.g., concert,
27.5%
plays, shows)
Visit theme/amusement park 26.6%
Indoor recreation (e.g., tennis, swimming,
. . . 26.3%
bowling, rock climbing)
Visit historic sites 25.9%
Sightseeing — rural 24.3%
Attend sports events 22.6%
Visit 200 or aquarium 19.9%
Visit national- or state park 19.7%
Attend festival 19.5%
Guided tour 17.2%
Visit art museum or gallery 15.6%
Visit history- or science museum 14.7%
Visit urban park or botanical garden 14.5%
Winter sports (e.g., skiing) 7.9%
Attend camp 5.0%
Attend lecture 4.8%
Volunteering program (e.g. conservation or 3 5%

helping project)

 

Note: “ Percentages add to more than 100% due to multiple responses. 1’ SD in parenthesis

indicates standard deviation.

Testing the Measurement Model: Confirmatory Factor Analysis in SEM

This section concentrates on assessing and refining the measurement model that
represents relations between observed variables and factors. First, SEM assumption tests
measuring normality ofthe data are discussed. Next, the measurement model was
assessed and refined through confirmatory factor analysis (CFA). The procedures for
improving model fit are presented, along with the discussions of reliability, convergent
validity, and discriminant validity for the measurement model.
Normality Test

SEM is highly sensitive to distributional characteristics of data, and maximum
likelihood estimation (MLE) used in SEM is based on the assumption of normality (Hair
et al., 1998). Normality for each variable in the proposed model was examined to
determine if the data meet the normality assumption for the MLE method. The normality
test is an important preliminary analysis step since test results must fall within acceptable
standards for subsequent SEM analyses to be meaningful. Skewness and kurtosis tests
were performed to evaluate normality. Data in Table I 1 show that the value for
univariate skewness and kurtosis ranged from —2.36 (A1) to —0.28 (PBC2) and from —
0.33 (SN3) to 8.0 (A1) respectively. Values of all variables in the model for univariate
skewness and kurtosis were found to fall within conventional criteria (Kline, 1998) of

normality (-3 to 3 for skewness and —10 to 10 for kurtosis).

to

Table l 1. Normality" test results of observed variables included in the proposed model
presented in Figure 3.

 

 

 

 

Constructs Variable names Skewness” Kurtosisc

(> I 3 I = extremely (> I 10I : extremely
_i' __ . _______ - . skewed) pc_ake_d)_~w _
Attitude Al -2.36 8.00
A2 -1.70 4.29
3 -I .95 4.66
A4 -2.09 5.86
A5 -2.14 6.95
Subjective norm 8N1 -0.76 -0.06
8N2 -1.20 2.40
8N3 -0.73 -0.33
8N4 -0.62 0.00
Perceived behavioral control PBCI -0.46 0.14
PBC2 -0.28 -0.44
PBC3 -0.87 -0.04
PBC4 -0.79 -0.25
Cognitive social identity C811 -0.59 0.37
C812 -1.06 1.01
C813 -1 .21 1.81
C814 -1 .06 0.67
Affective social identity A811 -I.I8 1.70
A812 -1 .37 2.46
A813 -2.06 3.58
A814 -1 .68 1.96
Evaluative social identity E811 -l.21 1.42
E812 -I .21 1.50
E813 -1.84 2.50
E814 -1.43 0.76
Desire D1 -1.60 3.34
D2 -l.89 4.43
D3 -1.57 2.85
D4 -0.94 0.21
We-intention W11 -1.41 1.36
W12 -1 .33 1.52
W13 -1.21 1.16
W14 -1.12 0.65

 

Note: aNormality was examined in terms of skewness and kurtosis. B Skewness refers to the symmetry of
the distribution. Skewness with a value above three is conventionally considered as extremely skewed. °
Kurtosis indicates a relative excess of cases in the tails ofa distribution relative to a normal distribution. A
Kurtosis value of 10 is a conventional criterion indicating normality distribution in terms ofits peakedness.
Values above 10 are considered extremely peaked.

()1

Model Specification

The prOposed measurement model was specified for the relations between the
observed variables and the factors through loadings of the observed variables and their
error term. All factors and observed variables were specified based on previous empirical
studies discussed in Chapter 3.

As shown in Figure 4, the proposed measurement model consisted of nine factors
and 34 observed variables. Social identity was included as a second-order factor in the
proposed model, which was specified by three factors comprising cognitive social
identity, affective social identity, and evaluative social identity, each of which was
specified by four observed variables. Attitude, subjective norms, perceived behavioral
control, desire, and we-intention were each specified by four observed variables. All
observed variables in the proposed measurement model were presented earlier in Table 1.
Model Testing and Modification Procedures

Overall fit indices for proposed and modified measurement models through
modification procedures are presented in Table 12. In step 1, the proposed measurement
model was tested. According to overall fit indices, the proposed model did not produce a
good fit with the data, 76(477) = 2268.2, p < .001 (xz/df= 4.76, CFI = .820,NNF1 = .800,
RMSEA = .091) in this initial step (xz/dfsmall than 3.0, RMSEA small than .05, CPI and
NNFI all greater than .90 as suggested by Kline(1998). Thus, the proposed model needed

further modification.

62

l

 

 

Subjective

   
 
   

 

  
   
     

Perceived
behavioral
control

I

 

 

 
  
 
 

 
 
 

  
 

 
 
  

  
    

Cognitive

social identity

Affective

social identity

Evaluative
social identity

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A 1 <—
A 5 <—
SN 1 <—
SN 4 «—
PBCI <——
PBC4 <——
DI <——
D4 <—
WI 1 <—
w14 <—

 

 

Figure 4. Proposed measurement model.

63

 
  

e13

e17
e18

621

e22

e25

e33

 

 

CSI 1

 

 

 

 

 

C814

 

 

 

 

A81 1

 

 

 

 

 

A81 4

 

 

 

E81 1

 

 

 

 

 

E814

 

 

e1
e4
e5

e8

e9

e12

Model modification procedures was used to identify observed variables that had
low factor loadings, significant cross-loadings, and large residuals using standardized
factor loadings, modification index (MI) in factor loading, and modification index (M1) in
covariance between two error term variables. As the minimum cut-off, it has been
suggested that a standardized factor loading should be greater than .50, and each M1
should not exceed 100 (Kline, 1998). After removing observed variables with above
standardized factor loadings and M1 , goodness of fit indices (CFI, NNFI and RMSEA)
were iteratively used to determine how well the CFA model fit the data.

In the second step, six observed variables with low factor loadings (below
suggested level of .50 for the expected constructs.) were removed from Model I. The
CFA test results showed that Model I was a significantly better fit than the proposed
model in terms of chi-square difference (ME (171) = 739.1, p < .001). But this model still
exhibited a poor fit to other indices ( xz/df= 5.00, CFI = .862, NNFI = .842, RMSEA
= .094). Five observed variables (W13, W14, 8N3, D1, A I) shared factor loadings (above
the MI suggested level of 100) with other observed variables. These five variables were
deleted to create Model 2.

In step 3, Model 2 was tested. Even though Model 2 was a significantly better fit
than Model I in terms of chi—square difference, it was determined that Model 2 needed
further improvement, Ax2(115)= 562.3, p < .001 (for: 5.06, CFI = .875, NNFI
= .849, RMSEA = .095). Four observed variables (8N4, A813, A814, A4) were
identified and deleted due to shared large residuals with other observed variable loadings

(above the MI suggested level of 100) to create Model 3.

64

Table 12. Comparison of overall fit indices for proposed and modified full measurement
models (n=452).

 

 

Model _ x2 df xz/df NNFI CFI RMSEA A 212
Proposed model 2268.2 477 4.76 .800 .820 .091 -
Model 13 1529.1 306 5.00 .842 .862 .094 739, 1 2
Model 2b 966.8 191 5.06 .849 .875 .095 5623*
Model 3c 213.9 1 17 1.83 .974 .980 .043 7529*

 

Note: NNFI = non-normed fit index; CFI = comparative fit index; RMSEA = root mean square
error of approximation; A )6 = )8 difference between two models (i.e., xzof Model 1 minus xzof
Model 2 = 562.3). Criteria to accept model: [i/df should be small than 3; NNFI and CFI should
be greater than .90; RMSEA should be small than .05. a Model 1: Proposed model minus (E813,
E814, C811, C814, PBCI, PBC4). I’Model 2 : Model 1 minus (W13, W14, 8N3, D1, Al). " Model
3 = Model 2 minus (8N4, A813, A814, A4). * significant at p < .05

Finally, Model 3 was re-estimated. Results from assessing this model in the last
step showed that all fit indices indicated a good fit to the data, A76 (74) = 752.9, p<.01,
xz/alf = 2.6, CFI = .99, NNFI = .974, RMSEA = .043. The modified model (Model 3)
exceeded the initial proposed model on all fit criteria, which confirmed that the
modifications were meaningful and appropriate.

During this process to assess model fit, a total of 15 observed variables were
removed from the proposed model as shown in Table 13. Every factor in the final
measurement model (Model 3) was measured using at least two observed variables as
shown in Table 14.

Reliability

The reliability test was conducted to evaluate the internal consistency and
precision of the observed variables measuring a given factor. Results are presented in
Table 14 to include means, standard deviations and reliabilities for factors in the final

measurement model (Model 3). Reliability scores range from (1 = .66 to (it = .91.

Observed variables should have a Cronbach’s alpha of .7 or higher to be judged reliable

65

measures (Nunnally, 1978). All scales demonstrate generally good reliability except for

perceived behavioral control ((1 = .66).

Table 13. Variables removed from the proposed measurement model based on the
examination oflow factor loadings, multiple loadings and large residuals.

 

Reasons for removing from proposed model

 

. Low factor Multiple loadings Large residuals
Variables . . .
Model removed loadings (MI of regressmn (M1 for covariance
(Standardized weight b >100) of error term °
loading ‘1 <.50) >100)

 

E813
E814
C811
C814
PBC]
PBC4
W13
W14
M2 8N3
D1
A1
8N4
A813
A814
A4
Note: a Standardized loadings indicate relationships between observed variables and their
associated factors. b Modification index (M1) for regression weights is the change in chi-
square if the path between two variables is correlated. ° Modification index for covariance
of error term variables is the decrease in chi-square if two error term variables were
allowed to correlate.

Ml

XXXXXX

XXXXXX

M3

XXXX

 

Convergent Validity and Discriminant Validity

Convergent validity. Convergent validity is used to determine if different
observed variables used to measure the same factor are highly correlated. In SEM,
convergent validity can be assessed by reviewing the t-test for the factor loadings
(Hatcher, 1994). As can be seen Table 15, all factor loadings for the observed variables

measuring the same factor are statistically significant. showing that all observed variables

66

Table 14. Means, standard deviations and reliabilities for factors in Model 3 (N = 452).

 

 

 

 

 

Number
of Reliability
_ _ _ F _actors' variables Mean (SD (Cronbach’s 0t)

Attitude 3 19.1 2.3 .889
Subjective norm 2 I 1.4 2.2 .809
Perceived behavioral control 2 9.7 2.7 .657
Social identity

Cognitive social identity 2 11.4 2.3 .773

Affective social identity 2 l 1.9 2.0 .763

Evaluative social identity 2 1 1.6 2.4 .905
Desire 3 17.6 3.5 .843
We-intention 2 1 1.2 3.0 .909

 

Note: C ronbach’s (1 provides an estimate of the inter-item reliability or consistency.

effectively measure their corresponding factor thereby supporting the convergent validity
of the results. Observed variables that were specified to measure each of the factors in the
revised final model (modified model 3) all have relatively high loadings (statistically
significant at p < .05), ranging from .62 to .93, which is also evidence of convergent
validity.

Discriminant validity. Discriminant validity was assessed in two ways. First,
inspection of the correlations among factors was conducted. As can be seen Table 16,
estimated correlations between factors were not excessively high, and none of the pairs
for the 95% confidence interval approach 1.00, thus providing support for discriminant

validity (Anderson & Gerbing, I988).

67

Table 15. Confirmatory factor analysis results for the Model 3 (15 of original 34
observed variables removed) (N = 452).

 

 

”Attitude

Social identity

Desire

We-intention

Social identity

Factor Observed variables Standardized t-value
retained in the loading a
Modified model 3
A2 .902* -
A3 .802* 21.0
A5 .865* 23.2
Subjective norm 8N1 .824* -
SN2 .828* 12.7
Perceived behavioral control PBC2 .615* -
PBC3 .636* 9.7
Cognitive social identity C812 .735* -
C813 .871* 15.6
Affective social identity A811 .732* -
A812 .848* 14.8
Evaluative social identity E811 .854* —
E812 969* 23.6
D2 .873* -
D3 873* 24.0
D4 .715* 17.6
W11 .900* -
W12 .929* 26.0
C81 .872* -
A81 .853* 1 1.0
E81 .808* 11.9

 

Note: a Standardized loadings indicate relationships between observed variables and their
associated factors in order to examine convergent validity. Convergent validity refers to the
degree of association between observed variables of a factor. * Factor loadings were all

significant at p < .01.

Table I6. Correlations among exogenous factors in the Model 3 for examining
discriminant validity (N = 452).

 

 

1 2 3 4
1. Attitude 1 - - -
2. Subjective norm .264 1 - -
3. Perceived behavioral control .039 .536 1 -
4. Social identity .431 .476 .288 1

 

The second method used is a stronger test of discriminant validity. A series of chi-
square difference tests were conducted (Anderson & Gerbing, 1988; Bagozzi & Phillips,
1982). The chi-square difference test can be used to assess the discriminant validity of
two constructs by calculating the difference ofthe chi-square statistics for the constrained
and unconstrained measurement model (Hatcher, 1994). The constrained model is
identical to the unconstrained model, in which all factors are allowed to co-vary, except
that the correlation between the two constructs of interest is fixed at one.

A significant chi-square difference indicates discriminant validity between the
pair of factors (Anderson & Gerbing, 1988). Discriminant validity is demonstrated if the
chi-square difference (with Idf) is significant indicating that the chi square of the
constrained model is significantly lower than that of the unconstrained model, implying
that the model in which the two factors are viewed as distinct (but correlated) factors is
superior. As shown in Table 17, all chi-square differences ranged from 26.0 to 153.7
which all exceed 1 1.58 (p < .01), demonstrating adequate discriminant validity for all

factors.

69

Table 17. Chi-square difference tests for evaluating discriminant validity: Model 3
(N = 452).

 

Unconstrained model a
22(117)= 213.9

 

Constrained model b

 

Factor pair )6 (1 18) A 3:6
(A, 8N) 339.8 1259*
(A, PBC) 377.6 163.7*
(A, 81) 313.0 991*
(SN, PBC) 239.9 260*
(SN, 81) 256.4 425*
(PBC, 81) 286.3 72.4*

 

Note: a Model 3. b Constrained model having correlation between two factors that set to 1.0. c )8
difference between unconstrained model (Model 3) and constrained model ’1' Significant at the .05
overall significant level. A significant 12 difference ([112) indicates discriminant validity between
the pair of factors.

Testing the Hypothesized Structural Models
Goodness-of-fit of the Structural Model

In this study, the proposed structural model presented earlier in Figure 3 was
analyzed using measures resulting from the measurement model analysis. Results of the
analysis of path coefficients for this structural model are shown in Table 19.

As can be seen in Table 18, the proposed structural model adequately fits the data
with RMSEA equal to .04 (which is smaller than .08) and xz/dfequal to 1.81 (which is
smaller than 3). Other fit indices included: x2: 215.0 (df= 119, p < .001), CFI = .98, and
NNFI = .97.

Table 18. Overall fit indices for the proposed structural model (N = 452)
Model 22 df xz/df CFI NNFI RMSEA

Proposed model 215.0 1 19 1.81 .980 .974 .04

Note: CFI = comparative fit index; NNFI = non-normed fit index; RMSEA = root mean square
error of approximation. Criteria to accept model: x2/df should be small than 3; NNFI and CF I
should be greater than .90; RMSEA should be small than .05.

 

 

70

Path Coefficients and Hypothesis Testing

Path coefficients estimated by SEM and hypothesis testing results are presented in
Table 19. The path coefficient from subjective norms to desire (8N —> D) was not
significant at the .05 level, indicating a weak and positive relationship (B = .010, t = 0.01,
p < .05). But, the path coefficients from social identity to desire (81 —> D: B = .733, t =
11.2, p < .01), from social identity to we-intention (81 ——> WI: B = .319, t = 4.2, p < .01),
and from desire to we-intention (D —> WI: B = .677, t = 5.3, p < .01) were significant
at .01 with strong and positive relationships. Further, path coefficients from attitude to
desire (A —» D: B = .132, t = 2.6, p < .05) and from perceived behavioral control to desire
(PBC —> D: B = .163, t = 4.5, p < .05) were significant at .05 level with strong and
positive relationships.

The significant test results for path coefficients support all hypotheses (H l , H3,
H4, H5, and H6) excepting one (H2) (SN —+ D).

Based on the relative values of path coefficients that are shown in Figure 5, social
identity (B = .733) has the highest level of explanatory power for desire when compared
to perceived behavioral control (B = .163), attitude (B = .132) and subjective norm (B
= .010). These factors explain 74% (R2 = .74) ofthe variance in desire, as can be seen in
Figure 5. Regarding factors influencing we-intention, desire among three factors had the
strongest effects on we-intention (B = .733). The effect of perceived behavioral control on
we-intention is shown by its high level of explanatory power (B = .622). Compared to the
other two factors, effect of social identity on we-intention is somewhat weak (B = .319).
Perceived behavioral control, social identity, and desire explained 79% (R2 = .75) ofthe

variance in we-intention.

71

Table 19. Path coefficients in the hypothesized structural model and hypotheses testing
results.

 

 

 

 

Standardized Standard Hypotheses
P_a_t_l_i__ _ _ __coefficient (B) t-value______ error testing results
A ——2 D .132* 2.56 .058 Supported
8N ——> D .010 .05 .056 Not supported
PBC —+ D .163* 4.51 .068 Supported
81 —> D .733** 11.2 .078 Supported
PBC —> WI .622** 9.12 .121 Supported
81 —> WI .319* 4.18 .179 Supported
D —> WI .677** 5.34 .162 Supported

 

Note: A=attitude; SN=subjective norm; PBC=perceived behavioral control; 81=social identity;
D=desire; W1=we intention. *Coefficients were significant at p < .05, "Coefficients were
significant at p < .01,

   
    
  
  

Perceived
behavioral control

  

R2=.79l

 

————> We-intention

Social identity

** Significant at p < .01

Figure 5. Test results for the proposed structural model: standardized path coefficients
and squared multiple correlations (R2).

Comparison with the theory of planned behavior. The first row data in Table 20
indicate the goodness-of-fit results for the model depicted in Figure 5. It also indicates for
two nested models, so that the theory of planned behavior can be compared with models
in which desire and social identity respectively were included. The significance ofthe
added paths corresponding respectively to the introduction of desire and social identity
were examined using the chi-square difference test.

The second row of data in Table 20 shows the goodness-of—fit indices for the
model with removal of the paths from social identity to desire and we-intention (81 ——> D
and 81 —> WI) from the proposed model. Because this modification produces a model
nested in the full model (the proposed model) displayed in Figure 5, the two can be
compared via the chi-square values for the two models (full model vs. nested model). The
purpose of this comparison is to test for the significance of both paths from social identity
to desire and we-intention. The chi-square difference tests revealed that the effects for
social identity were highly significant (ME (2): 189.3, p < .001). In other words, social
identity increases the explanatory power of the theory of planned behavior.

Table 20. Comparison of goodness-of-fit tests for the proposed model and nested models
with paths dropped from the proposed model.

 

.7

Model ._.x‘ a'f xz/df NNFI CFI RMSEA A 22
Proposed model 215.0 1 19 1.81 .974 .980 .042 -
Pmposed "10““ “1‘ 404.3 121 3.34 .925 .941 .072 189.3**
paths from S1 removed
Proposed model with
paths from D and SI 688.0 123 5.59 .854 .882 .101 283.7”
removed
Note: Sl=social identity; D=desire; NNFI = non-normed fit index; CFI = comparative fit index;
RMSEA = root mean square errdr of approximation. A X2 = xz difference between two models
(i.e., xzofproposed model minus x2 ofnested model). Criteria to accept model: y'i/df should be
small than 3; NNFI and CFI should be greater than .90; RMSEA should be small than .05. "Chi
square differences (2112) were significant at p < .01.

 

 

 

73

The third row of data in Table 20 summarizes results for the model with paths
”from both social identity and desire removed. This model, too, was nested in the full
model; the chi-square value for this model can be compared with the chi—square values
for the model summarized in the second row of data in Table 20, to test for the
significance of paths from attitude, perceived behavioral control, subjective norms, and
social identity to desire (A -+ D, SN —> D, PBC —2 D and 81 —> D and from desire to we-
intention (D ——> WI). The chi-square difference tests showed that the effects of desire are
highly significant (A36 (2) = 283.7, p < .001). Thus, desire enhanced the explanatory
power of the theory of planned behavior.

Further indications of the contributions made by social identity and desire were
evident in comparisons of explained variances for we-intentions across the three models
of interest. Table 21 shows a summary of explained variance in we-intention. The
addition of desire increases the explained variance in we-intention. R2 increases
from .696 to .740, when desire is added to the model of the theory of planned behavior.
Further, Table 21 also shows that the inclusion of social identity also has more effect on
the explained variance of we-intention: R2 increase from .740 to .791, when social
identity is added to the model of the theory of planned behavior plus desire. Based on the
results, it was concluded that the proposed model is an improvement on the model of the

theory of planned behavior to explain taking a vacation trip with a group of friends.

74

Table 21. Comparison of explained variance in we-intention for 1) the theory of planned
behavior, 2) the theory of planned behavior plus desire, and 3) the theory of planned
behavior plus desire and social identity.

 

 

Model We-intention
Theory of planned behavior .696
Theory of planned behavior plus desire .740
Theory of planned behavior plus desire and .791

social identity

 

Direct, Indirect and Total Effects on We-intention

Table 22 summarizes indirect, direct and total effects of factors on we-intention
based on test results of the proposed structural model presented in Table 19. In the model,
indirect effects worked through only desire since desire played a mediating role as well as
being a main. influence on we-intention. Indirect effects through desire were estimated by
production of standardized path coefficients obtained by SEM. For example, the product
for a path (A ——+ WI) through desire was .089 by .132 ofA -—> D multiply .677 ofD —->
WI. The indirect effect of social identity on we-intention through desire (.496) was
substantially stronger than that of perceived behavioral control (.1 10) and attitude (.089).

Direct effects were standardized path coefficients which already were obtained by
SEM. Attitude and subjective norm had no direct effects on we-intention while perceived
behavioral control and social identity had both direct and indirect effects in the proposed
structural model. Perceived behavioral control and desire had substantial direct effects on
we-intention (.677 and .622). Social identity also had a moderate direct effect on we-
intention (.319).

Total effects were estimated by summing the indirect effects and direct effects.
Table 22 indicates that total effect of social identity was high (.815). Perceived
behavioral control also was a significant factor for we-intention (.732). Further, desire

had a strong total effect on we-intention (.677), indicating that it played a significant role

75

of mediation in the model, especially, between social identity and we-intention (desire is
the key consideration with respect to social identity).

Table 22. Summary of indirect and direct effects of factors on we-intention to take a
vacation trip with a group of friends in the hypothesized structural model in SEM.

 

 

 

Path Indirect effect Direct effect Total effect
through desire
A —2 WI 0.089 - 0.089
SN ‘—) WI 0.007 - 0.007
PBC —> WI 0.110 0.622 0.732
81 —> WI 0.496 0.319 0.815
D—> WI - 0.677 0.677

 

Note: A=attitude; SN=subjective norm; PBC=perceived behavioral control; Sl=social identity;
D=desire; Wl=we intention.

Testing for Moderating Effects on the Proposed Structural Model

The moderating effects of group size on several paths were investigated in this
stage. The purpose of this analysis was to see whether there are differences in paths in the
tested model across groups by group size.
Moderating Effect of Group Size

The role of group size as a moderator of the relationships in the proposed
structural model was assessed with the multiple group analysis in SEM. To determine a
moderating influence, the four steps discussed in Chapter 3 were employed for this
analysis in SEM.

Table 23. Testing for moderating effects based on thgproposed structural model

b

X

 

 

Unconstrained a Fully constrained ” AX: C
(df= 238) (df= 257)
Model 465.19 526.78 6158*

 

Note: a Unconstrained model is estimated, in which path coefficients were allowed to vary across
the cross-group datasets (small vs. large group). b Fully constrained model is estimated by
requiring that all path coefficients were constrained to be equal for cross-group datasets. c x2
difference between unconstrained model and fully constrained model *The significant difference
(at p < .05) indicates that moderating effects do exist in group size.

76

Based on median group size, data were divided into two groups — one, a large
group of friends with more than five members in travel party and the other, a small group
of friends with fewer than five members in travel party. The indicators retained in both
groups were identical, to enable a comparison.

Group size difference in overall model. The x2 difference test was performed to
compare the unconstrained and fully constrained structural model. Results are presented
in Table 23. The )6 statistics for the unconstrained model and constrained structural
models were 465.19 (df= 238) and 526.78 (df= 257), respectively. The difference is
61.58, with 19 degrees of freedom. The significant difference (at p < .05) indicates that
moderating effects exist'with respect to group size.

Group size diffi’rences in individual paths. The x2 difference test was used again
to test for the moderating effects on individual paths. Table 24 presents chi-square
difference results for path coefficients between the small and the large group. The chi—
square difference tests showed that the paths of 81 —-—> D, PBC ~——> WI, D ——> W1, and 81
—2 WI were highly significant as indicated in the following: Ax2(1)= 3.99, p<.01, Ax2(l)
= 2.67, p < .05, (3)6 (I) = 2.95, p < .05, sz (I)= 2.90, p<.05 . These results mean that
there were differences in four paths across group size. But no statistically significant
difference could be found for the paths from attitude, subjective norms, and perceived
behavioral control to desire (A —-> D, SN —~> D, and PBC —> D ).

Comparison of path coefficients between two groups. Test results provided in
Table 25 indicate that the effect of social identity on desire was stronger when group size
was small than when large. We-intention was also more strongly influenced by social

identity when group size was small. But, the relationship between perceived behavioral

77

control and we-intention was significantly weakened by small group. The effect of desire

on we-intention was stronger for the small than the large group due to stronger effects of

social identity on desire for the small groups.

Table 24. Results oftesting for moderating effects based on the proposed model (Figure

3): test of invariance for path coefficients.

 

2

 

X
Path Unconstrained a Partially constrained b Ax2 c
(df= 238) (df= 239)
A —+ D 465.19 465.36 .17
SN —> D 465.19 465.62 .43
PBC —» D 465.19 465.97 .78
81—) D 465.19 469.19 3.99*
PBC —* WI 465.19 465.86 267*
D —> WI 465.19 466.14 295*
81 —2 WI 465.19 467.09 290*

 

Note: 3The unconstrained model was estimated, with path coefficients allowed to vary across the
cross-group datasets (small vs. large group). b Partially constrained means that only the target path
coefficients were set to be equal for cross-group datasets. c x2 difference between the
unconstrained models and partially constrained models * The significant difference (at p < .05)
indicates a difference in path coefficient across group size.

Table 25. Multiple group path analysis: comparison of path coefficients and t values for
the proposed model.

 

Small group (N = 234)
Standardized

Large group (N = 218)
Standardized

 

Paths coefficient tvalue coefficient tvalue S vs. L"
A——>D .100 1.48 .113 1.89 s=L
8N—>D .012 .15 .072 1.02 s=L
PBC —-4 1) .130 1.83 .092 1.06 s = L
81 —» D .757 7.49" .742 6.85‘ s > L
PBC —> w1 .536 6.28“ .698 6.07‘ s < L
D —2 WI .798 5.30‘ .518 3.38' s > L
81—» w1 .404 2.72‘ .178 1.19 s > L

 

Note: a 8: small group with fewer than 5 companions in travel party, L: large group with more
than five companions in travel party. * Significant at p < .05

\J

CHAPTER 5

DISCUSSION AND CONCLUSIONS

Overview
A summary of key findings is presented in this chapter including: 1)
characteristics of college students’ group vacation trips, 2) the valid and reliable
measurement model developed, and 3) the mediating path model, with an explanation of
the non-significant path between subjective norm and desire. Theoretical and practical
implications and contributions are discussed, and limitations of the study and directions

for future study are addressed.

Summary of Key Findings

Characteristics of College Students ’ Group Vacation Trips

Group vacation trip. It was found that college students mostly take their vacation
trips with others (99.4% of past trips and 96.6% of planned trips) rather than alone. Their
main companions on vacation trips are friends (about 80%) rather than family/relatives
(about 12%). These results supported previous studies which have found that group
behavior is the most common phenomenon in the tourism decision-making setting (Burch,
1969; Check et al., 1976; Field, 1971; Kim & Jogaratnam, 2002; Pizam et al, 2004;
Shoham, Schrage, & Eeden, 2004). It was found that college students’ travel in groups of
mainly friends with longer relationships (i.e., middle or high school classmates or
hometown friends) rather than with groups involving shorter relationships (i.e.,

classmates or roommates in college). The average size of these groups was reported to be

79

about five members, and the majority of group trips taken include both male and female
members.

Characteristics of planned group vacation trips with friends. College students’
group vacation trips occur during their summer vacations (55% of group vacation
travelers). This result was consistent with other tourism studies regarding college
students’ general travel behavior (Hobson & Josiam, 1992; Shoham et al., 2004). Their
planned average length of stay at the main destination was found to be about six days.
Their most popular destination is Florida (18.8%) followed by Mexico (7.1%) and
California (6.3%). The purposes of their trips are pleasure and entertainment activity
(55.0%). The main lodging type which will be used is hotel/motel (52.5%). Their planned
main transportation to their destination is commercial air (54.0%), and rental cars are
often (24%) used at their destinations. They planned spending is $719 on average for the
vacation trip while some 30% of them planned to spend over $1000 on their trips. Trips
are funded primarily from personal employment earnings and savings. Parents or
relatives were also found to be an important source of trip funding. They typically
participate in about seven activities, the most popular being: shopping, city or urban
sightseeing, visiting unique or themed restaurants/bars, and participating in beach
activities. Attending social events also was a preferred activity.

The role of friends in vacation trip planning among college students. Most college
students plan their trips with several friends (61.5%) rather than relying on a packaged
tour (1.2%) or planning alone (26.4%). The Internet was identified as an important
information source (19.6%), but previous vacation experience at the same destination

(34.2%) and recommendations of friends or colleagues(29.0%) were the two most

80

important sources. Overall, these findings show that “friends” play important roles as
main travel companions as well as information sources for college students’ vacation
trips. These findings are consistent with the tourism literature, which suggests that word
of mouth or personal communication from an immediate and trusted source (i.e., advice
of friends and relatives) is typically more influential than media communications (i.e.,
Internet) (Berestein & Steiner 1964; Nolan, 1976; Pizam et al., 2004).
Valid Measurement Model

In the first stage of structural equation modeling, confirmatory factor analysis was
employed to test measurement validity. Modification procedures were performed since
the proposed measurement model did not adequately fit the data set. A total of 15
observed variables were deleted from the full measurement model. The size of
standardized factor loadings and modification indices were used in selecting the variables
that were deleted. The modified measurement model was found to fit the data well. C FA
results established evidence of reliability, convergent validity, and discriminant validity
in the modified measurement model. The factor structure of social identity is inconsistent
with the literature, showing either three separate first-order factors or a second-order
three-factor structure. After conducting a series of CFAs, it was found that the second-
order three-factor model of social identify had the most parsimonious and better fit,
among all alternatives. The second-order three-factor model of social identify consisted
of three constructs — cognitive, affective, and evaluative social identifies, which are
separate but related. This second-order three-factor structure of social identity provides a
parsimonious picture of how college students’ perceptions of overall social identity

influence group vacation trip behavior. These results are consistent with other studies

81

(Bagozzi & Dholakia, 2002; Bagozzi & Dholakia, 2006; Bagozzi & Lee, 2002; Dholakia,
Bagozzi & Pearo, 2004))
Mediating Path Model, with an Explanation of a Non-significant Effect

In the second stage of structural equation modeling, the structural model was
evaluated to assess the hypothesized relationships among variables. The main objective
of this study was to examine individual and social effects on a college student’s desire
and we-intention to take a vacation trip with group of friends. Desire to take a vacation
trip with a group of friends was identified as a central factor that mediates relationships
between identified antecedents (attitude, perceived behavioral control, subjective norms,
and social identity) and the outcome variable (We-intention). More specifically, desire to
take a vacation trip with a group of friends was preposed as both a full mediator and a
partial mediator in the proposed model. Findings showed that overall fit for this
mediating construct was adequate. Findings were generally consistent with the proposed
hypotheses. All hypothesized relationships were strong and positive, as predicted, except
for the hypothesized relationship between subjective norm and desire to take a vacation
trip with a group of friends. The link between these two was identified as positive, but not
statistically significant. In fact, the path coefficient linking subjective norms and desire
almost approached zero (B = .01), indicating there is almost no relationship between
these two variables. Some possible explanations for this finding are discussed below.

One problem is how this construct of subjective norms was conceptualized and
measured. Subjective norms reflect social pressure or expectations from significant
others to perform a focal act (Ajen, 1991). Generally speaking, significant others may

include many different reference groups. For example, for the college student, such

82

“significant others” could include all important people, including those in romantic
relationships, parents, relatives and/or group vacation trip members. Thus, a particular
group influence is not singled out, so the interpretation of significant others may have
differed across college students. Since this study focused on interpersonal influences, and
since this reference group was allowed to vary across respondents, inconsistency in the
make up of reference groups selected could have masked an existing relationship between
subjective norms and desire.

Furthermore, the finding of non-significant path between subjective norms and
desire is somewhat consistent with other studies. Several social psychology researchers
have found that subjective norms are inadequate and rarely predict intentions (Ajzen &
Driver, 1992; Amitage & Conner, 2001; Goldin & Kok, 1996; Shepperd, Hartwick, &
Warshaw, 1988;). Furthermore, according to the study by Bagozzi and Lee (2002)
examining the group activity of going for lunch together with a group of friends,
subjective norms were not pervasive determinants of an outcome, in both Korean and
American samples, compared to other predictors. In several other examples in the tourism
and leisure literature (Ajzen & Driver, 1992; Lam & Hsu, 2004), subjective norms have

not been found to contribute significantly to the behavioral outcome variable.

Theoretical Implications
Conceptual Development and Empirical Validation
One of the fundamental theoretical contributions of this study is the conceptual .
development and empirical validation of each factor in the proposed model. The concepts

of social identity and a shared intention or collective intention (called We-intention) were

83

introduced to the context of travel decision-making in this study. Applicability of social
identity and We-intention to the tourism setting was evaluated through a series of
measurement model tests. Results provide evidence of construct validity for both factors.
This evidence suggests that they should be more widely included in tourism research
studies.

The concept of social identity, grounded in social psychology theory, has been
used in consumer and marketing research to determine social intentions or group
behaviors relating to selecting and purchasing a particular brand or product. In this study,
social identity was applied successfully regarding taking a vacation trip with a group of
friends among college students. People achieve social identity through self-awareness of
their membership in a group and the emotional and evaluative significance of this
membership (Tajfel, 1978). In this study, a group of friends, with whom a college student
took a previous vacation trip, was identified as an important social group for college
students. In line with social identity theory, it was confirmed that the factor structure of
social identity is a second-order three-factor, comprising cognitive (a cognitive awareness
of one’s membership in a social group — self categorization), evaluative (a positive and
negative value connotation attached to this group membership — group self-esteem), and
emotional (a sense of emotional involvement with the group — affective commitment)
components of social identity (Ellemers et al., 1999).

Traditional travel-decision making frameworks have emphasized individual
traveler’s intention to take a vacation trip or why individual travelers choose a particular
destination. In this study, as a main outcome variable, intention was conceptualized as a

shared or collective intention, instead of personal intention. Bagozzi and his colleagues

84

(Bagozzi & Dholakia, 2000; Bagozzi & Dholakia, 2006; Bagozzi & Lee, 2002; Dholakia
et al., 2004) define this shared intention as intentional social action, and call it We-
intention. In this study, their conceptualization of We-intention proved to be applicable
to explaining taking a vacation trip with a group of friends identified by an individual
traveler. Measures of We-intention used in this study proved to be reliable and valid.
Application ofa shared-intention or collective intention in the context oftravel-decision
making was successful in this case.
Incorporating Both Social and Personal Factors

Findings from this study help explain why college students continue to take
vacation trips with the same group of friends. Traditionally, studies of college student
travel have focused mainly on identifying their planning tools, travel patterns, activities,
and travel involvement levels. Furthermore, the traditional framework of push and pull
factors has been applied to explain college student motivations. In this study, findings
showed that college students take and plan the majority of their vacation trips with a
group of friends, rather than alone. In this study, social influences as well as
psychological influences were investigated to examine college students’ unique
motivation to taking vacation trips with a group of friends. In particular, social influences
- specifically, social identity — rarely have been addressed in previous tourism research
investigating travelers’ decision-making processes. Understanding of both effects on
college students’ decision-making has important theoretical implications for examining
college students’ group travel motivations.

When compared to perceived behavioral control, attitude and subjective norm,

social identity was found to have the highest level of explanatory power for predicting a

85

college student’s desire to take a vacation trip with a group of friends. When examining
the total effect of each variable on predicting the main outcome variable, a college
student’s We-intention to take a vacation trip with a group of friends, social identity was
shown to be the most important decision motivator. That is, a college student’s social
identity is the main motivating factor predicting a shared intention or collective intention
to take a vacation trip with a group of friends.

Sirakaya and Woodside (2005, p. 829) posited that “tourism is a social activity
that involves family, relatives, friends and others (if in a group travel setting), thus a
different approach is needed. Existing models lack the integration of these issues into a
single, unique model that is theoretically sound, complex, and still useful for practical
purpose.” This study represents an attempt to respond to their suggestions by integrating
both personal and social factors in an integrated model and empirically testing it to
identify the relative importance of these factors on travel decision-making among college
students. The findings of this study are consistent with other empirical marketing studies .
In summary, social identity is a salient variable in explaining a college student’s desire
and a shared intention (We-intention) to take a vacation trip with a group of friends. Thus,
among college students, both sociological and psychological variables enter their travel

decision-making.

Managerial Implications
Understanding the College Student Travel Market
The college student population is steadily increasing (for example, college

students enrolled in US colleges and universities will exceed 17 million by 2012 (US

86

Census Bureau, 2004) and the economic impact ofthe college student market should not
be underestimated. The college student market segment is especially important to tourism
marketers, since college students tend to travel more than other market segment (Shoham,
Schrage, & van Eeden, 2004; Schrage et al., 2001). Given rapidly increasing competition
in the tourism market, marketers should adapt their marketing strategies to address the
unique needs of college students. Thus, understanding the nature and characteristics of
this travel market segment is the first important step in developing an effective marketing
strategy targeting college students.
Targeting Friendship Group Rather Than Individual College Students

For college students, it was found that friends play important roles as main travel
companions as well as information sources when planning and taking vacation trips.
College students often make joint commitments to participate in group travel, rather than
deciding to take a vacation trip, or where to go, alone. A travel decision with a group of
friends has important tourism marketing implications since such groups, once formed, are
extremely influential in shaping and changing individual college student preferences,
attitudes and behavior. Accordingly, when developing tourism marketing programs,
tourism marketers need to place emphasis on the wants and needs of group travelers
rather than on those of individual travelers’. Accordingly, when tourism marketers target
college students’ vacation trips during spring- or summer breaks, marketing programs
should target groups of students rather than individual students. And, marketers need to
encourage college students to take a vacation trip with friends; perhaps promotional

events such as “bring your friends with you” or “discount program when five people

book at the same time” could be used to encourage more such group vacation trips among
college students.
Connecting Social Identity to Vacation-Related Products and Services

People gain a sense of whom they are and derive much oftheir esteem from their
memberships in their social groups and categories (Tajfel & Turner, 1979; 1986). This is
a key concept of social identity theory. Social identity is when individuals conform to
standard roles and rules defined and borrowed from society (Treholm & Jensen, 1996). In
consumer behavior and marketing research, social identity has been found to influence a
wide variety of consumer attitudes and behaviors. Since people are motivated to maintain
a positive social identity, consumers often are attracted to products and brands linked to
their social identity (Forehand, Desphpande, & Reed, 2002; Stayman & Desphpande,
1989). That is, a consumer conforms to social expectations and presents himself or
herself as “one of the group members,” including when he or she considers product-
and/or brand selection.

College students may identify with group memberships or categories that
meaningfully distinguish various forms of social identity(SI), depending on their interests
and preferences. In this study, college students identified themselves as related to a
friendship group which took a vacation trip together previously. This type of vacation
friend 81 was shown to be a most important factor in motivating group vacation trip
behavior among college students. An important question is: what lessons can tourism
marketers learn from this important finding? Tourism and destination marketers need to
understand how such college students’ social identities potentially provide valuable

insights into developing tourism-related products and services targeting college students.

88

Tourism marketers need to explore ways of enhancing social identity among college
students. Bond and Baer (2001) suggested social identity can be enhanced by offering
environmental characteristics linking to social identity in service industries. For example,
the Hard Rock Café targets young adults such as college students, and provides their
social identity with its themes of rock music and decor. College students who take a
vacation trip with a group of friends may relate this Hard Rock Café to their
identification through group membership and strengthen their “vacation friends” social
identities.
Using Online Travel Community Enhancing Social Identity

Marketers can create meaningful connections between their products and
consumer social identities (Oliver, 1999). In this study, shopping was found to be the
most preferred activity at main destinations. When college students take a vacation trip,
they are likely to purchase tourism—related products that can enhance social identity with
a group of friends, such as logos, badges, bumper stickers, special uniforms, or group
pictures with special destination/venue frames. Gift shops in tourism destinations
promote sales of these symbolic items enhancing social identity among the travel party.
When tourism marketers advertise a particular destination targeting college students, they
should emphasize aspects enhancing the social identity (e.g., recreational activities a
group of friends can enjoy together) among a group of friends with whom a college
student plans a vacation trip, rather than simply focusing on the attributes of the
destination. A finding ofthis study was that word-of-mouth communication from friends
or significant others and their own previous experience at the same destination were the

most important primary information sources, while the Internet was the most important

89

secondary information source. Based on these findings, tourism marketers should
consider using word-of mouth communication via the Internet medium to enhance social
identity among college student travelers. Social interactions occur not only personally -
such as interacting with a group of friends, as in this study, but also online or in the
virtual travel community. Sharing information about their personal experiences at a
particular destination or vacation-related products or services can facilitate social
interaction among the members. Currently, many existing virtual travel communities
exist on the Internet, and tourism marketers should consider how they can utilize these
online community mediums to encourage group travel by enhancing social identity.
Building Group-Friendly Environments Enhancing Social Identity

To increase group travel motivation among college students, tourism marketers
should emphasize in their destination advertising, group-friendly environmental
characteristics that enhance social identity in the group of friends with whom a college
student will travel. Furthermore, tourism marketers need to develop events, programs,
and facilities emphasizing social activities among friends. Tourism marketers can
develop recreational programs delivering group experiences to enhance social identity.
For example, vacation marketers can offer group-based activities (e.g., rafting, beach
volleyball, hiking, etc.), which allow participants to develop feelings of communion with
their friends with whom they travel. It is expected that the greater the degree to which the
tourism marketers provide recreational activities enhancing social interaction and social

cohesion, the greater will be the enhancement ofthe social identity.

\0
<3

Limitations and Future Research

Despite its theoretical and practical contributions to the field oftourism marketing
and tourism consumer behavior, several limitations ofthe present study need to be
addressed. These include: 1) the measurement issue associated with subjective norm, 2)
generalizability of results and external validity, 3) the limited number of social factors
considered, 4) only past and not new travel partners were allowed, and 5) investigating
other moderators. Future studies need to address these identified issues and limitations to
extend the body of knowledge on the effects of social and individual factors on group
travel behavior.
The Subjective Norm Measurement Issues

In this study, the hypothesized relationship between subjective norm and desire to
take a vacation trip with a group of friends was not found to be significant. It is believed
that the conceptual definition and measurement of this construct could be changed to
better capture its influence. Researchers have argued that the way in which subjective
norms are conceptualized within the TPB framework fails to tap important facets of
social influences (e.g. Conner & Armitage, 1998; Terry, Hogg, & White, 1999).
According to the Theory of Reasoned Action (Ajzen & Fishbein, 1980) and the Theory of
Planned Behavior (Ajzen, 1992), subjective norm was originally conceptualized as
perceived social pressure from significant others to perform a focal act. “Significant
others” may include family, relatives, co-workers, friends, etc. An important objective of
this study was to examine the social influence ofa group of friends on group travel
behavior. Unfortunately, the respondents in this study were not required to reference only

“a group of friends” in questions relating to “significant others.” This flaw in

91

questionnaire design confounded interpretation ofthe results related to subjective norms.
Once recognized, it can be mitigated easily through appropriate future questionnaire
design. Thus. in future studies, a reconceptualization of subjective norms is required, to
distinguish the scope of subjective norm between a particular group of friends, and
significant others (i.e., family and/or coworker/s), in examining college students’ desire
and intention to take a vacation trip.
Identifying More Social Factors

Only two social factors were included in this study: I) subjective norm and 2)
social identity. The findings ofthis study have demonstrated that social identity is an
important factor influencing a college student’s desire and intention to take a vacation
trip with a group of friends. Since group travel is the most common travel phenomenon
among college students. it is important to assess other social influences to deepen
understanding ofthe social factors that influence college students’ travel decisions.
Adding more social/group variables such as group norms, social normative perception,
attention to social comparison information and susceptibility to interpersonal influences
could prove useful in extending understanding ofthe college students’ group travel
decision-making process.
Generalizability of Results and External Validity

Convenience sampling targeting only nine universities in the US may limit the
generalizability ofthe findings ofthis study to the entire travel market of college students.
Instructors at nine universities were contacted by the researcher and study participants
were self-selected rather than randomly selected. Western US universities were not

included. However, this study focuses primarily on relationships between variables, thus

92

this type of research question is typically less vulnerable to generalizability problems
(Burnett and Dunne, 1986; Sears, 1986).

Future studies are needed that apply the same theoretical framework to other age
populations to achieve external validity. For example, it would be interesting to examine
the effects of individual and social factors on the vacation decision-making process
among youth groups (age 12-17) or among older population segments.

Comparison Between Past and Potential New Partners

The main outcome variable in this study was a college student’s We-intention to
take a vacation trip with a group of friends. This reference group of friends was
identified by each college student and was specified to be a group with whom the
respondent had taken a vacation trip previously. Thus, the focus was on past travel
partners and not on potential new travel partners. The choice of past travel partners was
consistent with Bagozzi and his colleague’s We-intention literature (Bagozzi & Dholakia,
2000; Bagozzi & Dholakia, 2006; Bagozzi & Lee, 2002; Dholakia et al., 2004). They
argued that restriction to an on-going (i.e., past travel partners) was deemed necessary for
manageability and to control boundary conditions.

This study only investigated the effects of social and individual influences on a
college student’s desire and intention to take a vacation trip with an on-going group of
friends. Thus, it would be informative to compare the relative effects of social and
individual influences on a college student’s desire and intention to take a vacation trip
with friends including both past and potential new travel partner groups. It is quite
possible that social factors may play different roles depending on the length of

relationships within groups.

93

Extending the Mode of Intentions to Actual Participation Behavior

The travel decision process is complex involving many different stages from need
recognition to post evaluation. In this study, data collection was conducted during the
planning stage among college students. Thus, the main outcome variable for this study
was not actual tourism behavior, but rather behavioral intention to take a vacation trip
with a group of friends. Although behavioral intention is the most proximate
measurement of actual behavior, it would be meaningful to extend this proposed
framework to link behavioral intention to actual travel participation behavior. To do so,
data would need to be collected twice, i.e., prior to taking a vacation trip and after taking
a vacation trip.

Travel decision models generally deal with one of two aspects of travel behavior:
whether to go and where to go. This study focused on whether or not a college student
intended take a vacation trip with a group of friends identified by him or her, namely an
on-going group. Social factors may also be important determinants of college students’
destination choices, although empirical studies to test this proposition do not exist. The
existing literature suggests that destination choice is largely determined by pull-factors,
namely attributes of a destination. Since group travel is a social act, college students are
less likely to be influenced by pull-factors than by friends when making destination
choices.

Investigating Other Moderators

The traveler’s decision process does not end once the vacation is completed.

Previous literature has suggested that experiences at and with the destination tend to

influence revisit decisions. By the same token, experiences (satisfaction or

94

dissatisfaction) may contribute to future travel behavior. In the proposed framework,
social identity was proposed as an independent variable influencing a college student’s
desire and intention to take a vacation trip with a group of friends. Interestingly, positive
travel experiences with a group of friends (i.e., group experience) likely strengthen one’s
level of identity with that group, which means that social identity now becomes the main
outcome variable.

Study findings indicate that group size moderates the relationships between social
identity and desire and between social identity and intention to take a vacation trip with a
group of friends. The magnitudes of these relationships were found to be stronger for
small group sizes than for large group sizes. Future studies need to evaluate other
potential moderating variables besides group size. Perhaps cultural background could be
investigated. Due to the small number of international students in the current study,
international students could not be compared to American students to investigate the
relative effects of cultural / ethnic background on college students’ group travel decision
—making. Previous literature is consistent in indicating that cultural background is a
major determinant of group behavior (Bagozzi & Lee, 2002; Markus & Kitayama, 1991;

Shoham, Schrange, & Van Eden, 2004).

95

APPENDICES

96

Appendix A.

Consent Form For Main Study

You are invited to participate in the study, “Intention to take a vacation trip with a group
of friends.” The purpose of this study is to examine personal and social influences on
college students’ decision-making in taking a vacation trip with a group of friends. This
survey asks you for information about your attitudes, desires, and intentions toward
taking a vacation trip with a group of friends. This survey also seeks general
demographic information and questions relating to travel information search, and travel
planning.

The survey will take 20—30 minutes. This study is for research purposes only. Your
responses will not be associated with you in any way and will remain strictly
confidential. Your privacy will be protected to the maximum extent allowable by law.
Your identity will not be linked to the data you provide. There are no anticipated risks
associated with participation.

You will receive extra credit in your class in exchange for your participation if your
instructor has previously agreed to provide extra credit. If you choose not to participate
in this study, you can earn equivalent extra credit by doing an alternative project.
Providing your email address on the last part of the survey is completely voluntary, but is
used for drawing prizes. At the close of the survey, each of six respondents selected by
random pick will receive an Amazon.com electronic gift certificate or money order in the
amount of US$50. This gift certificate or money order will be emailed to the email
address provided by you. After selecting and sending to six respondents, all email
addresses in the data will be deleted to protect your identity.

Participation in this study is voluntary, and you may choose not to participate at all, you
may decline to participate in certain procedures or answer certain questions, or you may
discontinue your participation at any point without penalty or loss of benefits. You also
have the right to withdraw your consent to participate from this study at any time without
penahy.

Please direct any questions about this study to Dr. D. Holecek at dholcccktfiimsucdu,
phone: (517) 353-0793, fax: (517) 432-2296 or to Seunghyun Kim at
kimseun8@msu.cdu, phone (517) 803-6536. For questions regarding your rights as a
human subject of research, please contact Peter Vasilenko, Ph.D., Director of Human
Research Protection, (517) 355-2180, fax (517) 432-4503, e-mail irb@msu.edu, mail 202
Olds Hall, Michigan State University, East Lansing, MI 48824-1047.

 

 

 

You indicate your voluntary agreement to participate in this study by signing below:

 

Your Signature

97

Appendix B.
Survey Instruments

VACATION TRIP SURVEY

 

Vacation trip is defined as an overnight trig (A NIGHT OR MUCH LONGER) during vacation time
(spring or winter break, summer vacation, holidays)

 

 

 

 

Section 1. Vacation Trip Experience

I. Have you taken any vacation trip/s since becoming a college student?
:1 No (If no, continue to question #2)

1:1 Yes—p

 

1-1. If yes, how many trips? trips (continue to question #2)

 

 

 

2. Have you ever taken any vacation trip/s with your friend/s since becoming a college student?

:1 No _,.
1:1 Yes

 

2-1. If no, your main companion/s in your travel paity was/were? (check only ONE)
1:1 I traveled alone 13 Family / Relatives 1:1 Other (please specify)
(continue to question #3)

 

 

 

2-2. If yes, how many trips with your friend/s? trips (continue to question

 

 

 

 

Section II. Group [Members/or l’ucution Trip

3. Please read carefully the following situation:

IMAGINE that you and your FRIEND/S (one friend or more) are deciding (or have decided) where to
travel together in the coming vacation. Now, in your mind, picture that friend or those friends, while
you answer the following questions.

3-1. In your mind, and including yourself, how many persons might be in your group for this vacation trip?
Including yourself, persons

 

3-2. Please provide the requested information in the table below about eachfriend.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Friend Gender Age How many years have you
known each other?
Yourself :1 M [:1 F N.A.
Friend I a M 13 F year/s
Friend 2 1:1 M :1 F year/s
Friend 3 13 M E] F year/s
Friend 4 1:1 M Cl F year/s
Friend 5 13 M 13 F year/s
Friend 6 D M 1:1 F year/s
Friend 7 1:1 M :1 F year/s
Friend 8 1:1 M 1:1 F year/s
Friend 9 Cl M E] F year/s
Friend 10 C] M D F year/s
Over 10 Ifyou have more than ten. please skip to NEXT question.

 

 

98

 

Section II. Group Members for lt’uctttion Trip (C ont'tl)

3-3. How would you classify your friend/s you listed above? (check ALL that apply)

:1 Classmates [3 Long-term close friend

1:1 Boyfriend / girlfriend 1:1 Previous school friends (middle or high school)

C] Roommates 1:1 Friends from recreational / sports / hobby clubs in college
C1 Religion-related friends 1:1 Friends from student organizations in college

(e.g., sororities, fraternities, student groups, and clubs)
C1 Other (please specify)

 

3-4. Since you became a college student, have you taken any vacation trip with your friends you listed
on page 1?
1:1 No (if "No", continue to question #4)
1:1 Yes

 

3-4-1. If yes, how many trips with this group of friend/s since becoming a college
student? trips (continue to question #4)

 

 

 

4. What is the main destination you might visit with your friend/s you listed within this year?

 

City or Place

 

State or Country (if other than US)

 

 

 

 

Section 111. Opinions of It’acution Trip with Group of F rientI/s

 

The following questions relate to your vacation trip with friend/s WITHIN THIS YEAR. Please answer
each ofthe following questions by circling the number that best describes your opinion. In this section,
“YOUR FRIEND/S YOU LISTED” refers to the SAME FRIEND/S you identified at the beginning ofthe
survey.

5. Forgetting what others think you should do and their reactions, how do you PERSONALLY feel about
taking a vacation trip with your friend/s you listed within this year. (please circle a number you choose
for each)

(For example, if you choose “Extremely negative”, circle “I”, or if “Slightly positive”, circle “5”)

 

 

Extremely Quite Slightly Neither Slightly Quite Extremely
negative 1 2 3 4 5 6 7 positive
unpleasant l 2 3 4 5 6 7 pleasant
bored 1 2 3 4 5 6 7 fun
not enjoyable I 2 3 4 5 6 7 enjoyable
unfavorable 1 2 3 4 5 6 7 favorable

 

 

 

6. How much control do you believe you have over taking a vacation trip, with your friend/s you listed?

no I I 2 I 3 I 4 I 5 I 6 I 7 Complete
control Control

 

7. For you to take a vacation trip, with your friend/s you listed within this year, is

extremely 1 I 2 I 3 I 4 I 5 I 6 I 7 Extreme] y
difficult Easy

 

99

Section III. Opinions of Vacation Trip with Group o/‘Frieml/s (C out?!)

 

8. Most people who are important in your life think you

shouldnot 1 I 2 | 3 | 4 | 5 | 6 l 7 Should

 

 

 

take a vacation trip with your friend/s you listed within this year.

9. Your desire for taking a vacation trip with your friend/s you listed can be described as

no desire I I 2 I 3 I 4 I 5 I 6 I 7 Strong
at all --------- desire

 

 

 

10. People in your life whose opinions you value would

strongly 1 I 2 I 3 I 4 I 5 I 6 I 7 strongly
disapprove --------- approve
of your taking a vacation trip with your friend/s.

 

 

 

l 1. Please indicate the degree to which your self-image overlaps with the identity of your friend/s you
listed, as you perceive it.

not at all I I 2 I 3 I 4 I 5 I 6 I 7 overlapped
overlapped very much

 

 

 

12. How strong are your feelings of belonging to the group you indicated?

Not at all 1 I 2 I 3 I 4 I 5 I 6 I 7 very strong
strong ---------

 

 

 

13. How attached are you to the group you indicated?

not at all 1 I 2 I 3 I 4 I 5 I 6 I 7 attached
attached very much
Not at all attached: you have no positive feelings toward group,
Attached very much: you have substantial positive feelings toward the group

 

 

 

14. Please indicate whether the following statements are likely or unlikely for you. (circle a number you
choose for each)

 

Extreme ly . Ex tremely
unliker Neither likely
We (1 and my friend/s I listed) intend to take a

. . . . . 1 2 3 4 5 6 7
vacation trip together Within this year.

It is expected of me that 1 take this vacation trip with 1 2 3
my friend/s I listed within this year.

I intend to take a vacation trip with my friend/s I 1
listed within this year.

10
w
A
LII
C5
\l

 

 

100

Section 111. Opinions of Vacation Trip with Group of F rieml/s (Cont’tl)

 

15. Please indicate whether the following statements are true or false. (circle a number you choose for each)

 

 

 

Definitely Definitely
False Neither true

My behavior often depends on how I feel others 1 7 3 4 5 6 7
wish me to behave. 7
My most important friends take a vacation trip with 1 2 3 4 5 6 7
friend/s within this year.
We (1 and my friend/s 1 listed) will try to take a 1 2 3 4 5 6 7
vacation trip together within this year.
When I am uncertain how to act in a social 1 2 3 4 5 6 7
situation, I look to the behavior of others for cues.
I believe I have the resources required to take this I 2 3 4 5 6 7
vacation trip, with my friend/s.
When in a social situation, I tend not to follow the
crowd, but instead behave in a manner that suits my I 2 3 4 5 6 7
particular mood at the time.
It is my feeling that if everyone else in a group is
behaving in a certain manner, this must be the I 2 3 4 5 6 7
proper way to behave.

 

16. How would you rate each of following statements regarding a vacation trip with your friend/s you
listed? (circle a number you choose for each)

 

 

Very Very
strongly Neutral strongly
disagree agree

I desire to take a vacation trip with my friend/s l 7 3 4 5 6 7
I listed.

I intend that I and my friend/s I listed take a 1 2 3 4 5 6 7
vacation trip together within this year.

It is mostly up to me whether or not I take a 1 7 3 4 5 6 7
vacation trip with my friend/s I listed. 7

We (1 and my friend/s I listed) plan to take a 1 2 3 4 5 6 7
vacation trip together within this year.

 

 

101

 

Section III. Opinions of 1"ucuti0n Trip with Group o/‘Frientl/s ( C out?!)

17. Please circle the number that best describes you in each ofthe following statements. (In these questions,
“THE GROUP I INDICATED” refers to the SAME FRIEND/S you listed at the beginning ofthe

survey.)

 

Describes

 

Does not Describes
describe dmf: t 1 me W"?
me at all m0 :5, e y well
I want to take a vacation trip with my I 2 3 4 5 6 7
friend/s I listed.
I feel an urge or need to take a vacation trip 1 2 3 4 5 6 7
with my friend/s I listed.
1 am like other members ofthe group 1 1 2 3 4 5 6 7
indicated.
I identify with other members ofthe group I 1 2 3 4 5 6 7
indicated.
I dislike being a member ofthe group 1 1 2 3 4 5 6 7
indicated.
I would rather belong to the other groups. I 2 3 4 5 6 7
I am a valuable member of the group I 1 2 3 4 5 6 7
indicated.
1 am an important member ofthe group I 1 2 3 4 5 6 7
indicated.
1 would rather not tell that I belong to the 1 2 3 4 5 6 7
group 1 indicated.
1 think the group 1 indicated has little to be 1 2 3 4 5 6 7
proud of.
The group 1 indicated is an important 1 2 3 4 5 6 7

reflection of who I am.

 

 

18. Between you and the friend/s you listed. who might make the following decision for the vacation trip?

(circle a number you choose for each)

 

 

2“ f
Selecting a destination (where to go) I 2 3 4 5 6 7
Vacation budget 1 2 3 4 5 6 7
Information collection I 2 3 4 5 6 7
Overall vacation decisions I 2 3 4 5 6 7

 

 

102

   

Section I I". Next Vacation Trip

Please answer the following questions about the next vacation trip you plan to take.

I9. Do you plan to take any vacation trip/s within this year?
[3 Yes (if“Yes"’, continue to question #20) D No (gif“No”, go to question #33)

20. Your next vacation trip is likely to be made with you and your friend/s you listed at the beginning
ofthe survey?
[3 Yes (if“Yes”, continue to question #21)
D No —>

 

20-1. lfNo, who might be in your travel party on this vacation trip?
D Other/another friend/s D Family / Relatives D Other (please specify)

20-2. If No, including yourself, how many persons might be in your travel party on this
vacation trip? persons (continue to question # 21)

 

 

 

21. When might you take the next vacation trip (check only ONE)
C] Spring break [3 Holidays (please specify)
[3 Summer vacation D Other (please specify)

 

 

22. How many nights are you likely to be awayfrom home on this vacation trip? nights

23. How many nights are you likely to be at the main destination for this vacation trip? nights

24. What is the main destination you might visit on your next vacation trip

 

City or Place

 

State or Country (if other than US)

 

 

 

 

Example of "City or Place Daytona Beach, New Orleans, Disney World. Aspen resort, Cancun

25. How likely is your next vacation trip to be mainly organized? (please check only ONE)
C] By yourselfor with one person in a self-organized trip

With several persons in a self-organized trip

With several persons in a commercially organized trip (package tour)

By yourselfor with one person in a commercially organized trip (package tour)

Other (please specify)

DUDE]

 

26. What is likely to be the main purpose ofthe next vacation trip?
(Please check ONE at primary purpose, and ONE at secondary purpose)
Primarypurpose Secondary pmose
D D Recreation activity (Indoor or Outdoor)
Entertainment
Pleasure activity
Cultural activity
Sightseeing
Educational activity
Visiting friends or relatives
Volunteering
Other (please specify)

 

 

UUDUDDDD
DDDDDDDU

 

I03

Section IV. Next Vocation Trip (Cont ’(l)

 

27. When organizing this trip, what is likely to be the main sources ofinformation you might use?
(please check ONE at primary source, and ONE at secondary source)

 

 

Primary source Secondary source
C] [:1 Recommendations of friends or colleagues
[3 [3 Recommendations of parents or relatives
[:1 E] Previous experience with the same destination
[3 C] Articles in newspapers or magazines
E] D Search on the Internet
C] [3 TV program
D [3 Recommendations ofa travel agent
[3 D Other (please specify)

 

28. Primarily where would you likely stay? (check only ONE)

[3 Hotel/motel Cl Bed & Breakfast U Campground

a Dormitory [:1 Condo / time share [3 Cruise ship

[:1 Youth hostel E1 Private home (3 Organizational camp
[3 Other (please specify)

 

29. What might be the primary mode oftransportation? (please check ONE to destination, and ONE i_n
destination) (For example: if you take an airplane to Orlando, and you use rental car in Orlando, you
check “airplane” to destination and “rental car” in destination)

To Destination In Destination

D [3 Private Auto /' Truck
Ship / Boat

Bus / Motorcoach
Rental car

Airplane

Train

Other (please specify)

DDDDDD
DUDDDD

 

30. Please estimate the total dollars you might spend individually on the next vacation trip (including
transportation, meals, lodging, etc.) 5 .00

 

31. Please complete the following with regard to MAIN funding sources oftrip expenses.
(check ALL that apply)

U Parents or relatives (3 Grants or scholarships
[3 Savings [1 Jobs
E] Loan [:1 Other (please specify)

 

104

 

Section IV. Next V acotion Trip (Cont '(l)

32.

37.

39.

. What university / college are you enrolled in?

. What is your major?

. Please provide the name of your instructor who introduced this survey.

Please check the following activities you are likely to participate in during your next vacation trip.
(check ALL that apply)

:1 Attend sports events C] Attend performing arts event (e.g.. concert. plays. shows)

E] Attend a festival [:1 Participate in outdoor recreation (e.g.. camping. hiking. hunting.
golf. tennis)

[:1 Attend a camp [3 Participate in indoor recreation (e.g.. tennis. swimming. bowling.
rock climbing)

C] Attend a lecture D Participate in water sports (e.g.. boating. canoeing. raftingjct
skinning etc.)

I: Participate in beach E] Participate in winter sports (e.g.. skiing)

activities

[:1 Participate in a guided tour [3 Participate in volunteering program (e.g. conservation or helping
project)

[:1 Sightseeing —— City/Urban :1 Visit a unique or themed restaurant/bar (e.g.. wild west.
rock‘n’roll. rainforest)

Cl Sightseeing — Rural [1 Visit an urban park or botanical garden

Cl Shopping a Visit a theme/amusement park

[3 Visit a 200 or aquarium [:1 Visit a history or science museum

[I] Visit a national or state [:1 Visit an art museum or gallery

park
[:1 Visit historic place or sites Cl Attend social events (e.g., party)
Cl Other (please specify)

 

Section V. Demographics

 

 

 

. Please indicate your year, now?

D Freshman D Sophomore D Junior [:1 Senior Cl Graduate

. Your ethnic background? (check ALL that apply)

[I] White/Caucasian El Black / African American
C] Mexican American CJ Hispanic / Spanish American
Cl American Indian / Native American D Asian / Pacific Islander

Of which country are you a citizen?

 

 

This question will be used only for awarding you extra credits if your instructor has agreed previously.
Ifnot, please go to NEXT question.

What is your email address? (providing your email address is completely
optional and voluntary. Your email address will be used only for awarding you extra credits (if your
instructor has agreed previously) and/or drawing the prizes in the amount of either U53 5 0 money order
or US$50 Amazon. com e-gift certificate).

 

Thank you for your participation!!

105

REFERENCES

106

References

Ajzen. I. (1991). The theory of planned behavior. Organizational Behavior and Human
Decision Processes, 50, 179-21 1.

Ajzen, I., & Driver, B. L. (I992). Application ofthe theory of planned behavior to leisure
choice. Journal of Leisure Research, 24, 207-224.

Ajzen, I., & Fishbein, M. (I980). Understanding attitudes and predicting social
behavior. Englewood Cliffs, NJ: Prentice-Hall.

Allen, N. J., & Meyer, J. P. (I 996). Affective, continuance, and normative commitment
to the organization. Journal of Vocational Behavior, 49, 252-276.

Anderson, J., & Gerbing, D. (1988). Structural equation modeling in practice: A review
and recommended two—step approach. Psychological Bulletin, 103, 41 1-423.

Ankomah, P. K., Crompton. J. L., & Baker, D. (1996). Influence of cognitive distance in
vacation choice. Annals of Tourism Research, 23(1), 138-150.

Armitage, C. J., & Conner, M. (2001). Efficacy of the theory of planned behavior: A
meta-analytic review. British Journal of Social Psychology, 40, 471-499.

Bagozzi, R. P. (2000). On the concept ofintentional social action in consumer behavior.
Journal of Consumer Research, 27(3), 388-396.

Bagozzi, R. P., & Dholakia, U. M. (2002). Intentional social action in virtual
communities. Journal of Interactive Marketing. 16(2), 2-20.

Bagozzi, R. P., Dholakia, U. M., & Mookerjee, A. (2006). Individual and group bases of
social influence in online environments. Media Psychology, 8, 95-126.

Bagozzi, R. P., & Lee, K. H. (2002). Multiple routes for social influence: The role of
compliance, internalization, and social identity. Social Psychology Quarterly,
65(3), 226-247.

Bagozzi, R. P., & Dholakia, U. M. (2006). Antecedents and purchase consequences of
customer participation in small group brand communities. International Journal
of Research in Marketing, 23, 45-6].

Bearden, W. O., & Rose, R. L. (1990). Attention to social comparison information: An

individual difference factor affecting consumer conformity. Journal of Consumer
Research, 16, 461-471.

I07

Bergami, M., & Bagozzi, R. P. (2000). Self-categorization, affective commitment, and
group self-esteem as distinct aspects of social identity in the organization. British
Journal of Social Psychology, 39, 555-577.

Bentler, P. M. (1995). EQS structural equations program manual. Encino, CA:
Multivariate Software, Inc.

Bollen, K. A. (1989). Structural equations with latent variables. New York: John Wiley.

Bratman, M. E. (1997). I intend that we J. In G. Holstrom-Hintikka & R. Tuomela (Eds),
Contemporary action theory (pp. 49-63). Dordrecht: Kluwer.

Brewer, M. B. (1991). The social self: On being the same and different at the same time.
Personality and Social Psychology Bulletin, 1 7, 475-482.

Burch, W. R. (1969). The social circles ofleisure: Competing explanations. Journal of
Leisure Research, 1(2), 125-147.

Bureau, U. S. C. (2001). Statistical abstract of the United States. Washington DC: US.
Census Bureau.

Burnett, J. J ., & Dunne, P. M. (1986). An appraisal of the use of student subjects in
marketing research. Journal of Business Research, 14(4), 329-343.

Byme, B. M., (1994). Structural equation modeling with EQS and EQS/ Windows.
Newbury Park: Sage publications.

Cheek, N. H., Field, D. R., & Burdge, R. J. (1976). Leisure and recreation places. Ann
Arbor, Michigan: Ann Arbor Science.

Cho, M., & Kerstetter, D. L. (2004). The Influence of sign value on travel-related
information search. Leisure Science, 26, 19-34.

Christensen, J. E. (1983). An exposition ofcanonical correlation in leisure research.
Journal of Leisure Research, 14(4), 31 1-322.

Conner, M., & Armitage, C. J. (1998). Extending the theory of planned behavior: A
review and avenues for future research. Journal of Applied Social Psychology, 28,

1429-1464.

Crompton, J. L. (1981). Dimensions ofthe social group role in pleasure vacations. Annals
of Tourism Research, 8(4), 550-568.

Crompton, J. L. (1979). Motivations for pleasure vacation. Annals of Tourism Research,
6, 409-424.

108

Crompton, J. L. (1992). Structure of vacation destination choice sets. Annals of Tourism
Research, 19(3), 420-434.

Dann, G. M. S. (1981). Tourist motivations: An appraisal. Annals ofTourism Research,
8(2), 187-219.

Davis, W. A. (1984). The two senses of desire. Philosophical Studies, 45, 181-195.

Deaux, K., Kim, M., & Kathleen, E. (1995). Parameters of social identity. Journal of
Personality and Social Psychology, 68, 280-291.

Deshpande, R., & Stayman, D. M. (1994). A tale of two cities: Distinctiveness theory and
advertising effectiveness. Journal of Marketing Research, 31(1), 57-64.

Dholakia, U. M., Bagozzi, R. P., & Pearo, L. K. (2004). A social influence model of
consumer participation in network— and small-group-based virtual communities.
International Journal of Research in Marketing, 21, 241-263.

Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes. Fort Worth: Harcourt
Brace Jovanovich.

Ellemers, N., Kortekaas, P., & Ouwerkerk, J. (1999). Self-categorization, commitment to
the group and group self-esteem as related but distinct aspects of social identity.
European Journal of Social Psychology, 29, 371-389.

Frankfurt, H. (1988). The importance of what we care about. Cambridge: Cambridge
University Press.

Gerald, H., & Hoyt, M. F. (1974). Distinctiveness of social categorization and attitude
toward ingroup members. Personality and Social Psychology Bulletin, 29, 836-
842.

Gilbert, M. (1989). On social facts. London, UK: Routledge.

Gitelson, R., & Kerstetter, D. (1994). The influence of friends and relatives on travel
decision-making. Journal of Travel and Tourism Marketing, 3(3), 59-68.

Godin, G., & Kok, G. (1996). The theory of planned behavior: A review of its application
to health-related behaviors. American Journal of Health Promotion, 11, 87-98.

Grier, S. A., & Deshpande, R. (2001). Social dimensions of consumer distinctiveness: the
influence of group social status and identity on advertising persuasion. Journal of
Marketing Research, 38(2), 216-224.

Hair. .I. 19.. Anderson. R. 13.. ’l‘utham. R. I-.. & Black. W. C. ( 1998). 1:1/ltllll'tll‘ltllt' data
. . . -/ . . _ , . . .
inn/lists ir/th readings (3" edition). Itpper Saddle River. NJ: Prentice-l lull.

Harris Interactive (2006). College students surfback to campus on a wave of digital
connection 2006.
htt p :// wwx-vharri sinteract i ve.com/ne\\-'s/ newsletters/cl ientnewsl2006 al lov2 .pd f

Hartup, W. W. (1993). Adolescents and their friends. In B. Laursen Ed.), New
directions for child development (pp. 3-22). San Francisco, CA: Jossey-
Bass.

Hobson, J. S. P., & Josiam, B. (1992). Spring break student travel - an exploratory study.
Journal of Travel and Tourism Marketing, 1(3), 89-97.

Hogg, M., Terry, D., & White, K. (1995). A tale oftwo theories: A critical comparison of
identity theory with social identity theory. Social Psychology Quarterly, 58, 255—
269.

Hogg, M. A. (1992). The social psychology of group cohesiveness: from attraction to
social identity. Hemel Hempstead, UK: Harvester Wheatsheaf.

Hogg, M. A., & Abrams, D. (1988). Social identification: A socialpsychologv of
intergroup relations and group processes. London: Routledge.

Holt, D. B. (1997). Poststructuralist lifestyle analysis: conceptualizing the social
patterning of consumption in postmodernity. Journal of Consumer Research,

23(March), 326-350.

Jiireskog, K. G., & Sbrbom, D. (1993), LISREL 8: Structural equation modeling with the
SIMPLIS command language [Computer software]. Chicago: Scientific Software
International.

Josiam, B. M., Hobson, J. S. P., Dietrich, U. C., & Smeaton, G. (1998). An analysis of
the sexual, alcohol, and drug related behavioral patterns of students on spring
break. Tourism Management, 19(6), 501-513.

Josiam, B. M., Smeaton, G., & Clements, C. J. (1998). Involvement: travel motivation
and destination selection. Journal of Vacation Marketing, 5(2), 167-175.

.Iosiam, B. M., Clements, C. J., & Hobson, J. S. P. (1994). Youth travel in the USA:
Understanding the spring break market. In A. Seaton (Ed), T ourism: The state of
the art (pp. 322-331). Chichester, UK: Wiley.

Kang, S. K., & Hsu, C. H. C. (2005). Dyadic consensus on family vacation destination
selection. Tourism Management, 26, 571-582.

110

Kelman, H. C. (1974). Further thoughts on the processes of compliance, identification,
and internalization. In J. T. Tedeschi (Ed). Perspectives on social power (pp.
125-171). Chicago: Aldine.

Kim, K., & Jogaratnam, G. (2002). Travel motivations: A comparative study of Asian
international and domestic American college students. Journal of Travel and
Tourism Marketing, 13(4), 61-82.

Kleine, R. E., Kleine, S. S., & Kernan, J. B. (1993). Mundane consumption and the self:
A social-identity perspective. Journal of Consumer Psychology, 2(3), 209-235.

Klenosky, D. B., Gengler, C. D., & Mulvey, M. S. (1993). Understanding the factors
influencing ski destination choice: a means-end analytic approach. Journal of
Leisure Research, 25(4), 362-379.

Kline, R. B. (1998). Principles and practice of structural equation modeling. New
York: The Guilford Press.

Knutson, B. J. (2000). College students and fast food - how students perceive restaurant
brand. Cornell Hotel and Restaurant Administration Quarterly, 41(3), 68-74.

Kropp, F., Lavack, A. M., & Silvera, D. H. (2005). Values and collective self-esteem as
predictors of consumer susceptibility to interpersonal influence among university
students. International Marketing Review, 22(1), 7—33.

Kygtabeb, R. L., & Crocker, J. (1992). A collection self-esteem scale: Self-evaluation of
one's social identity. Personality and Social Psychology Bulletin, 18, 302-318.

Laing, J. H. & Crouch, G. 1. (2005). Extraordinaryjourneys: an exploratory cross-cultural
study of tourists on the frontier. Journal of Vacation Marketing, 11(3): 209-223

Lam, T., & Hsu, C. H. C. (2004). Theory of planned behavior: Potential travelers from
China. Journal of Hospitality and Tourism Research, 28(4), 463-482.

Lam, T., & Hsu, C. H. C. (2005). Predicting behavioral intention ofchoosing a travel
destination. Tourism Management, 27(4), 584-599.

Lascu, D. N., & Zinkhan, G. (1999). Consumer conformity: Review and applications for
marketing theory and practice. Journal of Marketing Theory and Practice, 7(3), 1-
12.

Latane, B. (1981). The psychology of social impact. American Psychologist. 36, 343-356.

Lennox, R. D., & Wolfe, R. N. (1984). Revision of the self-monitoring sale. Journal of
Personality and Social Psychology, 46(6), 1349-1369.

111

Long. K., & Spears, R. (1997). The self-esteem hypothesis revisited: Differentiation and
the disaffected. In R. Spears, P. J. Oakes, N. Ellemers, & S. A. Haslam (Eds),
The Social psychology of stereotyping and group life (pp. 296-317). Oxford: Basil
Blackwell.

Luhtanen, R. L., & Crocker, J. (1992). A collection self-esteem scale: Self-evaluation of
one's social identity. Personality and Social Psychology Bulletin, 18, 302-318.

Mangleburg, T. F., Doney, P. M., & Bristol, T. (2004). Shopping with friends and teens’
susceptibility to peer influence. Journal ofRetailing, 80, 101-1 16.

Markus, H. R., & Kitayama, S. (1991). Culture and self: implications for cognition,
emotion, and motivation. Psychology Review, 98, 224-253.

Mathieson, A., & Wall, G. (1982). Tourism economic, physical and social impacts.
London: Longman.

Mattila, A. S., Apostolopoulos, Y., Sonmez, 8., Yu, L., & Sasidharan, V. (2001).
The impact of gender and religion on college students' spring break behavior.
Journal of Travel Research, 40, 193-200.

Mayo, E. J., & Jarvis, L. P. (1981). The psychology ofleisure travel. Boston, MA: CBI
Publishing Company, Inc.

McGuire, W. J., McGuire, C. V., Child, P., & Fujioka, T. (1978). Salience of ethnicity in
the spontaneous self-concept as a function of one's ethnic distinctiveness in the
social environment. Journal of Personality and Social Psychology, 36, 51 1-520.

McGuire, W. J., & Padawer-Singer, A. (1976). Trait salience in the spontaneous self-
concept. Journal of Personality and Social Psychology, 33, 743-754.

Meyers-Levy, J., & Sternthal, B. (1991). Gender differences in the use of message cues
and judgments. Journal of Marketing Research, 28, 84-96.

Middleton, V. T. C. (1994). Marketing in travel and tourism. Boston: Butterworth-
Heinemann Ltd.

Miniard, P. W., & Cohen, J. B. (1983). Modeling personal and normative influences on
behavior. Journal of Consumer Research, 10, 169-180.

Moutinho, L. (I 987). Consumer behavior in tourism. European Journal ofMarketing,
21(10), 3-44.

Mummendey, A., & Simon, B. (1989). Better or different? III: the impact ofimportance

of comparison dimension and relative ingroup size upon intergroup discrimination.
British Journal of Social Psychology, 28, 1-16.

112

Netemeyer, R. G. (1986). A propositional inventory of interpersonal influence research
and tests of the role of normative and informational influence in two models of
behavioral intention. Unpublished doctoral dissertation, University of South
Carolina.

Nowak, A., & Vallacher, R. R. (1998). Toward computational social psychology:
Cellular automata and neural network models of interpersonal dynamics. In S. J.
Read & L. C. Miller (Eds), Connectionist models of social reasoning and social
behavior. Mahwah, NJ: Erlbaum.

Oliver, R. L. (1999). Whence consumer loyalty. Journal ofMarketing, 63(Special Issue),
33-44.

Olmsted, M. s, & Hare, A. P. (1978). The small group (2nd edition), New York: Random
House.

Perugini, M., & Bagozzi, R. P. (2001). The role of desires and anticipated emotions in
goal-directed theory of planned behavior. British Journal of Social Psychology.
40, 79-98.

Pizam, A., Jeong, G.-H., Reichel, A., Boemmel, H. V., Lusson, J. M., Steynberg, L.,
State-Costache, O., Volo, 8., Claudia, K., Kucerova, J ., & Montmany, N. (2004).
The relationship between risk-taking, sensation-seeking, and the tourist

behavior of young adults: A cross-cultural study. Journal of Travel Research,
42(February), 251-260.

Raaij, W. F. V., & Francken, D. A. (1984). Vacation destinations, activities and
satisfactions. Annals ofTourism Research, 11(1), 101-1 12.

Reed 11, A. (2004). Activating the self-importance of consumer selves: exploring identity
salience effects on judgments. Journal of Consumer Research, 31(2), 286-295.

Sachdev, L., & Bourhis, R. Y. (1984). Minimal majorities and minorities. European
Journal of Social Psychology, 14, 35-52.

Saegert, J., Hoover, R. J., & Hilger, M. T. (1985). Characteristics of Mexican American
consumers. Journal of Consumer Research, 12(June), 104-109.

Schmoll, G. A. (1977). Tourism promotion. London: Tourism International Press.

Schouten, J. W., & McAlexander, J. H. (1995). Subculture of consumption: An
ethnography of the new bikers. Journal of Consumer Research, 22(1), 43-61.

Searle, J. R. (1990). Collective intentions and action. In P. R. Cohen, J. Morgan, & M. E.

Pollack (Eds), Intentions in communication (pp. 401-415). Cambridge, MA: MIT
Press.

113

Sears, D. (1986). College sophomores in the laboratory: Influences ofa narrow database
on social psychology's view of human nature. Journal of Personality and Social
Psychology, 51(3), 515—530.

Sheperd, B. H., Hartwick, J., & Warshaw, P. R. (I 988). The theory of reasoned action: A
meta-analysis of past research with recommendations for modifications and future
research. Journal of Consumer Research, 15, 325-343.

Shoham, A., Schrage, C., & Eeden, S. V. (2004). Student travel behavior: a cross-
national study. Journal ofTravel and Tourism Marketing, 1 7(4), 1-10.

Simon, B., & Brown, R. (1987). Perceived intragroup homogeneity in minority-majority
contexts. Journal of Personality and Social Psychology, 53, 703-71 1.

Simon, B., & Hamilton, D. L. (1994). Self-stereotyping and social context: The effects of
relative in-group size and in-group status. Journal of Personality and Social
Psychology, 66, 699-711.

Singh, J. (1995). Measurement issues in cross-national research. Journal of International
Business Studies, 26(3), 597-619.

Sirakaya, E., & Woodside, A. G. (2005). Building and testing theories of decision making
by travelers. Tourism Management, 26(6), 815-832.

Sirsi, A. K., James, C. W., & Reingen, P. H. (1996). Microcultural analysis of variation
in sharing of causal reasoning about behavior. Journal of Consumer Research.

22(March), 345-372.

Sparks, P., Shepherd, R., Wierings, N., & Zimmermans, N. (1995). Perceived behavioral
control, unrealistic optimism and dietary change: an exploratory study. Appetite.
24, 243-255.

Tajfel, H. (1978). Social categorization, social identity, and social comparison. In H.
Tajfel (Ed), Differentiation between social groups: Studies in the social
psychology of intergroup relations (pp. 61-79). London: Academic Press.

Tajfel, H. (1981). Human groups and social categories: Studies in social psychology.
Cambridge, UK: Cambridge University Press.

Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G.
Austin & S. Worchel (Eds), The Social psychology of integroup relations (pp. 33-
47). Monterey: Brooks-Cole.

Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behavior. In S.

Worchel & W. G. Austin (Eds), Psychology of intergroup relations (pp. 7-24).
Chicago: Nelson-Hall.

114

Terry. D. J., & Ho g, M. A. (1996). Group norms and the attitude-behavior relationships:
A role for group identification. Personality and Social Psychology Bulletin, 22,
776-793.

Terry, D. J., Hogg, M. A., & Duck, J. M. (1999). Group membership, social identity,
attitudes. In D. Abrams & M. A. Hogg (Eds), Social identity and social cognition
(pp. 280-314). Oxford: Blackwell.

Terry. D. J., Hogg, M. A., & White, K. M. (I 999). The theory ofplanned behavior: Self-
identity, social identity, and group norms. British Journal of Social Psychology,
38, 25-244.

Tuomela, R. (I 995). The importance of us: A philosophical study of basic social nations.
Stanford, CA: Stanford University Press.

Turner, .1. C. (1985). Social categorization and the self-concept: A social cognitive theory
of group behavior. In E. J. Lawler (Ed), Advances in group processes: Theory
and research (pp. 77-121). Greenwich, CT: JAI Press.

Tussyadiah, L. P., Kono, T., & Morisugi, H. (2006). A model of multidestination travel:
Implications for marketing strategies, Journal of Travel Research, 44, 407-417.

Um, S., & Crompton, J. L. (1990). Attitude determinants in tourism destination choice.
Annals of Tourism Research, 1 7(3), 432-448.

Underwood, R., Bond, E., & Baer, R. (2001). Building service brands via social
identity: Lessons from the sports marketplace. Journal of Marketing Theory and
Practice, 9, 1-13.

US Census Bureau (2005). Statistical abstract of United States: 2004. Washington D. C .:
US Census Bureau.

Wahab, S., Crampon, L. J., & Rothfleld, L. M. (1976). Tourism marketing. London:
Tourism International Press.

Wang, K., Hsieh, A., Yeh, Y., & Tsai, C. (2004). Who is the decision-
maker: The parents or the child in group package tours? Tourism Management.
25, 183-194.

Woodside, A. G., & Lysonski, S. (1989). A general model oftraveler destination choice.
Journal of Travel Research, 27(1), 8-14.

115

Woodside, A. G., 8: MacDonald R. (1994). General system framework of customer
choice processes oftourism services. In R. V. Gasser & K. Weiermair (Eds),
Spoilt for choice. Decision making processes and preference change of tourist:
Intertemporal and intercountry perspectives (pp. 30-59). Thaur, Germany:
Kulturverlag.

Woodside, A. G., & Lysonski, S. (1989‘). A general model oftraveler destination choice.
Journal of Travel Research, 27(4), 8-14.

,.
.
Illt

  
     

 

HIG U

11111111111111llllillllllllllll

3 1293 O 845 8093