:3 . -.
2&6”.an
.

1.... 2.9 n
.. W3 Pu

.bnr...

1?. Hr ..,

u?! 1...}

95
Si.
J 3.....1
re .

1:. .n: uh”
. . Vard.’ 1!.

\ .51“... hr... 1%.. Pp

rd. ‘ fa.

tin
a. 5.1:: ‘2
In. (1:...

 

 

 

1'}. 2.2-: T

(

 

9000 111 LIBRARY
Michigan State l
University

 

 

 

This is to certify that the

dissertation entitled

A MODEL OF CONSUMER EXTERNAL PRICE SEARCH BEHAVIOR
IN AN ELECTRONIC MARKETPLACE (WORLD—WIDE-WEB)

A! _ _a .———_———~A-a———_..

presented by

James A. Ramos

has been accepted towards fulfillment
of the requirements for

Ph.D. Mass Media

degree in

 

 

 

 

Major professor l

' Dme August 5, 2003

 

MSU is an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

 

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.

 

DATE DUE DATE DUE DATE DUE

 

3923?: 8 3) 39GB

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/ClRCIOateDue.065-p.15

 

A MODEL OF CONSUMER EXTERNAL PRICE SEARCH BEHAVIOR
IN AN ELECTRONIC MARKETPLACE (WORLD-WIDE-WEB)

By

James A. Ramos

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Mass Media Ph.D. Program
Departments of Advertising, Journalism, and Telecommunication

2003

Tl
consumer
theoretit‘a'
determinu.
Schmidt J
theoretic:
informan.

Sc
311d inter:
instrumcnf
r€liabi 1m
made at it
Participant
COVanancc

|
model. and
ASSUmmO:

F02.

SUOngeSt i‘T

SUbjfictS pt:

 

 

ABSTRACT

A MODEL OF CONSUMER EXTERNAL PRICE SEARCH BEHAVIOR
IN AN ELECTRONIC MARKETPLACE (WORLD-WIDE-WEB)

By
James A. Ramos

This dissertation proposes and tests a causal model of the factors influencing
consumer external price search behavior in an electronic marketplace using an inclusive
theoretical perspective that incorporates both traditionally and online search
determinants. Drawing on the framework of consumer information search proposed by
Schmidt and Spreng (1996), the present model brings empirical verification to a broad
theoretical approach that integrates search costs economics, motivational theory, and
information processing theory.

Select items from preexisting scales were adapted to the context of price search
and intemet shopping to form the basis for the construction of the survey research
instrument. A pilot study using 127 student subjects showed scales to have acceptable
reliability. The main study employed a convenience sample of 587 students who had
made at least one purchase through the W in the three months preceding survey
participation. Maximum Likelihood estimation in LISREL 8.5 was used to analyze the
covariance matrix of observed variables in tests of the measurement model, structural
model, and overall model fit. Estimation of the proposed model adhered to the
assumptions of the general LISREL model.

Following testing and respecification, the model indicated that motivation was the
strongest predictor of online price search. The role of motivation was greatest when the

subjects perceived that prices varied considerably between WWW sellers and were

highly it:

strongly

 

were deg
shopping
T

online pr

COHIEXI 0

 

highly involved with the intemet as a task-aiding tool. The effect of motivation was
strongly decreased when the costs associated with online price search were high. Costs
were decreased for individuals knowledgeable about online search and increased when
shopping time was limited.

The structural model accounted for approximately one third of the variance in
online price search. This figure exceeds the amount of variance accounted for in earlier
price search research that exclusively used the costs vs. benefits framework, and is within
the range of the variance accounted for in prior price search research conducted in the

context of grocery shopping.

Copyright by
JAMES A. RAMOS
2003

To the Rev. Mr. Alfred J. Mehmel, Jr.

Amor Ordinem Nescit

number
possibii

researel

commm
chaheng
research.
research

PlOf. Chg
my tht‘or
Ft
apprecian

V'lanney S

 

the Rev. }.
Brennan.
attention :

Ti:
SimUiianC.

moods 0,-

 

 

ACKNOWLEDGEMENTS

It was my greatest fortune, over the long journey of my education, to have had a
number of extraordinary teachers fill my mind with stimulating ideas and endless
possibilities. They all have contributed, in part, to my successful completion of this
research and ultimately my doctorate. I thank them all for their care of me.

I owe particular thanks to the four individuals who formed my dissertation
committee. My Chairperson, Prof. Charles Salmon, in the subtlest manner, always
challenged me to think in broader terms, to search for deeper answers, and to do better
research. Prof Richard Spreng’s advise during the development and execution of this
research was invaluable. Prof. Stephen Lacy was of particular aide with methodology.
Prof. Charles “Chip” Steinfield helped me to see and understand the economic roots of
my theoretical objectives. Gentlemen, I thank you all.

For her stern and uncompromising standards that gave me my earliest
appreciations of order and discipline I thank Sister Loretta Ann Flynn of the St. John
Vianney School. I thank the faculty and staff of Fordham Preparatory School, especially
the Rev. Mr. Alfred J. Mehmel, Jr., Rev. Russell J. Sloun, SJ, and Rev. Pierce A.
Brennan, S.J., for instilling a drive to excel and a special appreciation for meticulous
attention to details.

This degree is an accomplishment for my parents as much as it is for me. Having
simultaneously raised a family and attended college, Miguel and Aida Ramos remain

models of perseverance for me, and I thank them for this and many years of sacrifice for

vi

my educ
for her p
for being
will suce

there for

years, ha
Harding
my life. a
his car at
has OTCht
a friends?
relieving
fTTCndS iii
mOmlngs
Susan Ch
academic
both hard
he’ll and }
\t‘tekendE

mOre bear

 

 

my education. My sister Denise has been my strongest ally and friend, and I thank God
for her presence in my life and her contributions to my sanity. I thank my brother Louis
for being real. My brother Anthony, like me, struggles with life, and I pray, that he also
will succeed over his personal trials. Finally, I thank my aunt, Anna Velazquez, for being
there for both my family and me.

I also am especially blessed to have a wide spectrum of friends, who, over the
years, have offered countless words of encouragement and gestures of support. Joseph
Harding has served as a voice of reason in my head. He has brought needed laughter into
my life, and directed me toward countless serious thoughts. I am so fortunate to have had
his ear and understanding, and I am thankful for his continued friendship. Jerry Lanoue
has orchestrated many a pick-me-up sessions and lent me many words of encouragement,
a friendship has never sounded better. Mark Klajman hosted many weekends of stress
relieving fun in Toronto. Andrew Mimnaugh, and his mother Veronica, have been loving
friends through this long process. Andrew kept me company though many unmotivated
mornings, and reminded me often that I could finish the work I began. Andy Lee and
Susan Chang, my officemates, acted as my screens against unwanted calls and as my
academic comrades. I will miss them. Rosanna Garcia showed me that a dissertation is
both hard and rewarding work. Iris Andriessen, Mimi Kim, and Yany Gregoire offered
help and laughter through structural modeling statistics camp. Joseph Cintron kept the
weekends filled with interesting detours that made the work of the week all that much
more bearable. Jason Brewer was always a firm friend. I thank them all for their

contribution to my successful completion of this research. Finally, no one has made my

vii

time in
eyes to 1
l
comprel
joy and L
P
Jackawa
me throu
my gradi
friend. I
gratitude
F
Alfred 1_
because i
he Would
My faihe
did not m

 

Pray that

and love I

 

time in Lansing more unforgettable than Doug Christensen. I thank him for opening my
eyes to the simple pleasures of Michigan.

I thank Stephen F uhrmann for living with me through the completion of my
comprehensives and the conceptualization of this research. He was the source of much
joy and helped me to imagine a happier life when I was finished with this work.

No one has supported my endeavors in graduate education more than Gwenyth
Jackaway. In fact, on more occasions than I can count, Gwenyth has been there to help
me through mental challenges. The patience she showed with me at the earliest stages of
my graduate work undoubtedly qualifies her for canonization. She has been my dearest
friend. Though I simply say thank you here Gwenyth, please know that I have more
gratitude than might ever be expressed in words.

Finally, there is a need to acknowledge the seminal figure in my life, the Rev. Mr.
Alfred J. Mehmel, Jr. of F ordham Preparatory School. I jokingly called him “Daddy”
because of a mistaken identity incident my freshman year. I had no idea at the time that
he would become a father figure for me. My mother always told me I could do anything.
My father always told me that I could do better. But Al, well, he always told me that it
did not matter what I did, because he had all faith that I would do it well. It was this faith
that gave me the patient strength and personal conviction to ascend toward this dream. I
pray that this achievement makes him proud. My eternal thanks for your faith, wisdom

and love Daddy. Amor ordinem nescit.

viii

LIST Ol
LlST OI
LIST OI

C HAPTI
TNTROI
Onlii
\"alu

[

C

E

Stud;
Orga

CHAPTl
LITE ILA,

Intro.
Theo

TABLE OF CONTENT

LIST OF TABLES .................................................................................................. xii
LIST OF FIGURES ................................................................................................. xv
LIST OF APPENDICES ........................................................................................ xvi
CHAPTER 1
INTRODUCTION .................................................................................................... 1
Online Price Search ............................................................................................. 1
Value of Further Investigation ............................................................................. 3
Dearth of Online Price Search Research ......................................................... 3
Online Applicability of Traditional Determinants .......................................... 5
Expanding Theoretical Perspective ................................................................ 6
Study Purpose ..................................................................................................... 8
Organization of Subsequent Chapters .................................................................. 9
CHAPTER 2
LITERATURE REVIEW ........................................................................................ 10
Introduction ....................................................................................................... 10
Theoretical Foundations of Information Search ................................................. 10
Introduction ................................................................................................. 10
Economic Approach .................................................................................... 10
Marketing Approach .................................................................................... 12
Comprehensive Frameworks Explaining Information Search ............................. 17
Model Specification .......................................................................................... 22
Online Price Search Behavior ............................................................................ 24
Perceived Ability to Online Price Search ........................................................... 24
Education .................................................................................................... 25
Subjective Knowledge of Internet Search ..................................................... 26
Perceived Motivation to Online Price Search ..................................................... 27
Internet Involvement .................................................................................... 29
Need for Cognition ...................................................................................... 30
WW Shopping Enthusiasm ...................................................................... 3O
Perceived Costs of Online Price Search ............................................................. 31
Subjective Knowledge of Internet Search ..................................................... 32
Time Pressure .............................................................................................. 33
Ease of Access to Online Information .......................................................... 34
Perceived Benefits of Online Price Search ......................................................... 35
Benefits vs. Costs of Online Search ............................................................. 36
Perceived Price Dispersion .......................................................................... 37

ix

 

"1 ' K,

I Us «1’

 

Sirr

C HAP";
METI-II
lntr
Dat
Pan
Me;
Dat.

C HAPI

PILOT
Resr
R65:

C HAPT
.‘I’IAIN 5
Intro
Data
AICB:
Mod.

 

 

Perceived Budget Constraints ...................................................................... 38

Trust in W Sellers ................................................................................. 39
Simultaneous Equations .................................................................................... 40
CHAPTER 3
METHODS ............................................................................................................. 42
Introduction ....................................................................................................... 42
Data Collection Procedures ............................................................................... 42
Participants ........................................................................................................ 44
Measurement ..................................................................................................... 45
Data Analysis Strategy ...................................................................................... 58
CHAPTER 4
PILOT STUDY RESULTS ..................................................................................... 60
Results .............................................................................................................. 60
Research Instrument Revisions .......................................................................... 64
CHAPTER 5
MAIN STUDY RESULTS ...................................................................................... 74
Introduction ....................................................................................................... 74
Data Screening .................................................................................................. 74
Measurement Model Assessment ....................................................................... 75
Model Revision ................................................................................................. 89
Scale Adjustment ......................................................................................... 89
Model Respecification ................................................................................. 91
Model Estimation .............................................................................................. 92
Overall Fit Assessment ...................................................................................... 95
Structural Model Assessment ............................................................................ 98
Model Modifications ....................................................................................... 108
Modifications ............................................................................................ 108
Modification Results .................................................................................. 109
CHAPTER 6
SUMMARY AND CONCLUSION ....................................................................... 115
Introduction ..................................................................................................... l 15
Objectives of the Study .................................................................................... 115
Research Design .............................................................................................. 117
Method of Analysis ......................................................................................... 117
Summary of Main Study Results ..................................................................... 118
Measurement Model Assessment ............................................................... 118
Overall Fit Assessment .............................................................................. 119
Structural Model Assessment ..................................................................... 120
Model Modifications ................................................................................. 122
Contributions of the Study ............................................................................... 123
Limitations of the Study .................................................................................. 125
Future Research Directions .............................................................................. 127

 

 

 

 

 

APPEN

REFER

APPENDICES ...................................................................................................... 133

REFERENCES ..................................................................................................... 175

xi

Table 1.1

Table 3.1
Table 3.2
Table 3.3
Table 3.4.
Table 3.5
Table 3b
Table 3,?
Table 3.54
Table 3.9

Table 3.]

 

Table 3.1
Table 3.],
T351631
Table 3.1.

Table 3.1

Table 41

Tame 4.2

LIST OF TABLES

Table 1.1 — Number of WW Sites Visited Before Purchase .................................... 2
Table 3.1 — Ability to Online Price Search Indicators .............................................. 47
Table 3.2 — Benefits of Online Price Search Indicators ............................................ 48
Table 3.3 — Costs of Online Price Search Indicators ................................................ 49
Table 3.4 — Ease of Access to Online Information Indicators ................................... 50
Table 3.5 — Education Indicators ............................................................................. 50
Table 3.6 — Internet Involvement Indicators ............................................................ 51
Table 3.7 — Motivation to Online Price Search Indicators ........................................ 52
Table 3.8 — Need for Cognition Indicators ............................................................... 53
Table 3.9 — Online Price Search Indicators .............................................................. 54
Table 3.10 — Perceived Budget Constraints Indicators ............................................. 54
Table 3.11 — Perceived Price Dispersion Indicators ................................................. 55
Table 3.12 —- Perceived Time Pressure Indicators ..................................................... 56
Table 3.13 —- Subjective Knowledge of Internet Search Indicators ........................... 56
Table 3.14 — Trust in WWW Sellers Indicators ........................................................ 57
Table 3.15 — WWW Shopping Enthusiasm Indicators ............................................. 58
Table 4.1 — Ability to Online Price Search Pilot Results .......................................... 65
Table 4.2 —- Benefits of Online Price Search Pilot Results ........................................ 66

xii

Table 4.
Table 4.
ka4.
Table 4.
Table 4.
Table 4..
Table 4.
Table 4i
Table 4.1
Table 4.1
Table 4.‘

Table 4.

Table 4,

Table 5‘
Table 5
Table 5
Table 5
Tame 5
Table 1
Table 2
Table

Tab“ 8

Table 4.3 — Costs of Online Price Search Pilot Results ............................................ 66

Table 4.4 — Ease of Access to Online Information Pilot Results ............................... 67
Table 4.5 — Internet Involvement Pilot Results ........................................................ 67
Table 4.6 — Motivation to Online Price Search Pilot Results .................................... 68
Table 4.7 — Need for Cognition Pilot Results ........................................................... 68
Table 4.8 — Online Price Search Pilot Results .......................................................... 69
Table 4.9 — Perceived Budget Constraints Pilot Results ........................................... 69
Table 4.10 -— Perceived Price Dispersion Pilot Results ............................................. 70
Table 4.11 — Perceived Time Pressure Pilot Results ................................................ 70
Table 4.12 — Subjective Knowledge of Internet Search Pilot Results ....................... 71
Table 4.13 — Trust in W Sellers Pilot Results ................................................... 71
Table 4.14 — WWW Shopping Enthusiasm Pilot Results ......................................... 72
Table 4.15 — Pilot Study Correlations Between Latent Variables ............................. 73
Table 5.1 — Ability to Online Price Search Measurement Model Results ................. 78
Table 5.2 — Benefits of Online Price Search Measurement Model Results ............... 79
Table 5.3 — Costs of Online Price Search Measurement Model Results .................... 80

Table 5.4 - Ease of Access to Online Information Measurement Model Results ...... 80

Table 5.5 — Internet Involvement Measurement Model Results ................................ 81
Table 5.6 — Motivation to Online Price Search Measurement Model Results ........... 82
Table 5.7 — Need for Cognition Measurement Model Results .................................. 83
Table 5.8 —- Online Price Search Measurement Model Results ................................. 84
Table 5.9 — Perceived Budget Constraints Measurement Model Results .................. 85

xiii

Table I

Table 1

Table 5

Table 5
Table 5
Table 5
Table 5
Table 5
Table 5.
Table 5.
Table 5.1

Table 5.2

 

Table 5.10 — Perceived Price Dispersion Measurement Model Results .................... 85

Table 5.11 — Perceived Time Pressure Measurement Model Results ........................ 86
Table 5.12 — Subjective Knowledge of Internet Search Measurement Results .......... 86
Table 5.13 — Trust in W Sellers Measurement Model Results ........................... 87
Table 5.14 — WWW Shopping Enthusiasm Measurement Model Results ................. 87
Table 5.15 — Main Study Correlations Between Latent Variables ............................ 88
Table 5.16 — Model Overall Fit Indices ................................................................... 95
Table 5.17 — Structural Model Results ................................................................... 100
Table 5.18 — Latent Variable Notation ................................................................... 101
Table 5.19 — Hypothesized Relationships and Empirical Findings ......................... 102
Table 5.20 — Squared Multiple Correlations for Structural equations ..................... 103
Table 5.21 — Empirical Results for Initial vs. Modified Model ............................... 112

xiv

LIST OF FIGURES

Figure 2.1 — Model of Consumer External Information Search ................................ 21
Figure 2.2 — Model of Consumer Price Search in an Electronic Marketplace ........... 23
Figure 5.1 — Revised Model of Consumer Online Price Search ................................ 93
Figure 5.2 — Results of Online Price Search Model .................................................. 97
Figure 5.3 — Results of Modified Online Price Search Model ................................ 111

XV

APPEX

 

APPENI

APPENI

APPENI

LIST OF APPENDICES

APPENDICIES ..................................................................................................... 133
APPENDIX A — Pilot Research Instrument ........................................................... 134
APPENDIX B — Modifications to Pilot Research Instrument ................................. 149
APPENDIX C — Research Instrument ................................................................... 159

xvi

0nhneI

7/)

unedace
and pm";
instantar.
mmum
IHTOI'Inar

C0115111115-

1999j

 

R

l
have ShOl

TIITOrmati
in its \w
L‘Sabinty
users fep(
Iheymler
searched 1

pnCCISO?

Chapter 1

INTRODUCTION

Online Price Search

Since its rapid diffusion, the world-wide-web (WWW), the intemet’s graphic
interface, has come to serve as a vast electronic marketplace, linking the general public
and private firms to numerous sellers of a wide range of products and services. By
instantaneously connecting consumers to a multitude of sellers, the W has the ability
to maximize access to product information while minimizing the costs of acquiring this
information (Bakos, 1991; 1997). It is this ability, some researchers argue, that will drive
consumer participation in this new electronic marketplace (Alba et a1., 1997; Keeney,
1999)

Recent figures appear to confirm this proposition. Several large WWW surveys
have shown that online shoppers have a strong tendency toward conducting prepurchase
information search with a particular focus on comparison price shopping. For example,
in its WWW User Survey (10th wave/4th Quarter 1998) the Graphic, Visualization, and
Usability Center (GVU) at Georgia Institute of Technology found that 61% of WW
users reported searching several times a month for information about different products
they intended to purchase on or off the WWW (GVU, 1998). The top reasons users
searched the WWW included obtaining detailed product information (92%), comparing

price (80%), and checking on product availability (76%).

 

 

researcl'I
that 78“
sites, an
and five
“WW 1
notexpr
search. I

of respo

purchase

Table 1.1

Number (if

Ufthase

TWO

TOtaI

 

 

Another example is a 1998 survey of 5,000 regular WWW shoppers by market
research firm Jupiter Communication (Steinberg, 2000; Wilder, 1998). The survey found
that 78% of WW purchases occurred after the shopper had visited at least two seller
sites, and 46% of WW purchases occurred after the shopper had visited between three
and five seller sites (Steinberg, 2000). Table 1.1 provides a breakdown of the number of
WW merchant sites visited before making a purchase. Though the Jupiter survey did
not expressly investigate WWW purchasers’ motivations for engaging in extended
search, price comparison may have been a driving force, since over three-fourths (77%)
of respondents reported that they had already decided on the exact product they wanted to

purchase before using the W to search multiple seller sites.

Table 1.1

Number of World-Wide-Web Seller Sites Visited by Intended Purchasers Before Purchase

 

 

 

 

 

 

 

% of intended % of intended

purchasers visiting purchasers visiting
Number of WW seller 2+ WWW seller 3+ WWW seller
sites visited before % of intended sites before sites before
mirchase purchasers purchasing purchasing
One 22% -- --
Two 25% 25% --
Three to five 46% 46% 46%
Six or more 7% 7% 7%
Total 100% 78% 53%

 

Source: Jupiter Communication (Steinberg, 2000)

 

eBates.
169-3 of
single" ;
toward r.
previou~
groups. '

of the “1“

that the}

‘r,

toward p
mar ketin
examinir

0f detem

 

CTECITOTIII

Searcli.

/U

T“;
Smith. 1
Price Sear
(GVU, Ig

scholars 11

A more recent example is a survey of 3,000 WWW consumers commissioned by
eBates, a retail portal, and conducted by Harris Interactive in June 2000. In this survey,
16% of participants clustered into a group the authors labeled “hooked, online, and
single” and 20% of participants were labeled “hunter—gathers” (Whelan, 2001). A drive
toward price search and a propensity to purchase from the low price leader, regardless of
previous knowledge or experience with that seller, characterized members of both
groups, which together constituted slightly more than one-third of participants. Members
of the “hunter gatherers” cluster differed from the “hooked, online, and single” group in
that they were older in age, newer to W shopping, and even more sensitive to price.

Value of Further Investigation

A broader understanding of the factors influencing online consurners’ propensity
toward price search behavior would be of value to researchers of electronic commerce,
marketing, and consumer behavior for three reasons. First, there is a dearth of research
examining online price search. Second, there is a need to assess the online applicability
of determinants known to influence consumer external information search in non-
electronic markets. Third, there is a need for more expansive theory on consumer online
search.

Dearth of Online Price Search Research

There is conspicuous lack of research that has investigated consumer online price
search. This dearth of online price search research exists despite apparent evidence that
price search is a common behavior among a considerable portion of online consumers
(GVU, 1998; Steinberg, 2000; Whelan, 2001; Wilder, 1998) and clear interest among

scholars in understanding consumer behavior in computer-mediated shopping

 

 

 

 

enviror.
Bellmal
study h.
with \V
This stt

sellers r

underst.

search. I
that can
number

CODSUmt

 

LOhse, I
examine
2001: \V
inflllenc
resm'cte‘
Kiel, & I
understa
13€r3pi.ct
Lukoslu:
COnSUme

environments (Alba et al., 1997; Deighton, 1997; Hoffman & Novak, 1996; Lohse,
Bellman, & Johnson, 2000; Peterson, Balasubramanian, & Bronnenberg, 1997). A single
study has examined online price search, looking at consumer choice behavior when faced
with W sellers offering marginally different prices (Smith & Brynjolfsson, 2001).
This study, however, focused on consumers’ choice behavior given familiar/unfamiliar
sellers post online price search. Further empirical research is needed to develop
understanding of the factors influencing the amount online price search itself.

While there is a dearth of research specifically concerning consumer online price
search, there is literature examining information search in computer-mediated markets
that can be adapted to aide in the development of online price search knowledge. A
number of studies have investigated the effects of reduced buyer search costs on
consumer choice behavior in computer mediated shopping environments (Hoque &
Lohse, 1999; Lynch & Ariely, 2000; Ward & Lee, 2000), and some research has
examined the intemet as a consumer information source (Ratchford, Talukdar, & Lee,
2001; Ward & Morganosky, 2000). There even has been some study of the factors
influencing the amount of consumer online information search, though it has been
restricted to the theoretical level without empirical testing. For example, Hodkinson,
Kiel, & McColl-Kennedy (2000) have proposed a diagrammatic approach to
understanding consumers’ online search, applying the “wayfinding” spatial orientation
perspective from geography to human-computer studies. Another example is the work of
Lukosius, Hyman, & Stratemeyer (2001), who have augmented a theoretical model of
consumer external information search proposed by Schmidt and Spreng (1996), adding

constructs they believe are relevant in a computer-mediated shopping environment.

 

 

 

 

present
consurr
and ifs
electror
traditioi
researcl
deterrm
Situatior
Situatior
COnSUmi
efforts 0
”When

I

 

 

 

 

eleCtrOni
011 the (It
tEndenC)
marketin
pUrchaS“
i“Count”;
1965; Be

1968. 19-

Online Applicability of Traditional Determinants

Research investigating online price search also would be of value because it
presents an opportunity to empirically verify if numerous factors found to influence
consumer search in non-electronic markets are applicable to an electronic marketplace,
and if so, to what extent. Further, research might uncover determinants unique to an
electronic shopping environment that influence consumer online price search. Both
traditional and unique determinants deserve the attention of researchers. Yet almost no
research has empirically verified the applicability of traditional information search
determinants to electronic shopping environments or investigated the existence of unique
situational factors. The Sundaram and Taylor (1998) study of “in-home shopping
situations” is the single exception. This research, however, did not exclusively assess
consumers online information search. The study’s focus, rather, is on the external search
efforts of purchasers who used home-based shopping formats, namely mail order,
telephone, or the W.

In addition, future attempts to create frameworks within which to understand
electronic market consumer behavior in broad terms will necessitate empirical research
on the determinants of online information search, particularly given online shoppers’
tendency toward price search. Consumer external search is acknowledged throughout the
marketing and consumer behavior literature as possessing a significant role in the
purchasing decision process. This is exemplified in the fact that external search is
incorporated into several of the leading consumer buying behavior models (Andreasen,
1965; Bettrnan, 1979; Engel, Blackwell, & Kollat, 1978; Engel, Kollat, & Blackwell,

1968, 1973; Howard, 1977; Nicosia, 1966).

 

 

theoreti
online I
Lee, 20
explain
greatly

search f

 

predictii
0n elect:
IOWef [h
behavio
Welfare
ec01mm

marker;

l:Ully ex
COmple
marl-let;
Value‘r

indit-id

Expanding Theoretical Perspective

A broader understanding of online price search would be of value because one
theoretical approach, search costs economics, currently dominates research on consumer
online search (Bakos, 1991, 1997; Hoque & Lohse, 1999; Lynch & Ariely, 2000; Ward &
Lee, 2000). This approach has proven to be promising, yet inadequate, at fully
explaining the online search phenomenon. Understanding of online search would benefit
greatly from a combination of this approach with the extensive literature on consumer
search from the fields of marketing and consumer behavior.

Smith, Bailey, and Brynjolfsson (2000), through a literature review, investigated
predictions, based on search cost economics, of online consumer behavior and its effects
on electronic market structure. They found that price dispersion on the W was no
lower than in conventional markets. This suggests that predictions of electronic buyer
behavior and market dynamics (e.g., price wars, profit erosions, and increased consumer
welfare (Bakos, 1991, 1997)) are inadequate, when based solely on search costs
economics, and that there remain incentives for retailer to participate in electronic
marketplaces (Alba et al., 1997).

There are three major reasons for the inadequacy of search cost economics to
fully explain the online search phenomenon. First, the theory assumes consumers are
completely rational and optimizing shoppers. Yet there is broad recognition in the
marketing and consumer behavior literature that consumers are not purely goal-oriented
value-maximizing entities (Babin, Darden, & Griffin, 1994; Parsons, 2002). In fact,

individuals have been shown to shop as a result of functional motives, non-functional

 

 

 

 

motive

that re

techno
of the I
on low
costs 0‘

search i

has bee _
Consum
WWW

"on‘trit

dlSCUSSi
Consumt
COnSumf
that are 1
reasons 1
convent)
vendOI’s (
the WWI

Clem 01m

 

motives, and a mix of functional and non-functional motives (Westbrook & Black, 1985)
that result in both hedonic and utilitarian outcomes (Babin et al., 1994).

Second, the theory’s application to intemet shopping generally has taken a
technologically determinist perspective (Chandler, 2001). It is predicated on the potential
of the W to facilitate specific human behaviors (i.e., search and comparison) based
on lower search costs. But because consumers place different values on the benefits and
costs of search, it is unrealistic for one to expect all consumers to engage in extended
search behavior.

Third, search cost economics is an incomplete perspective because the theory as it
has been applied to electronic marketplaces (Bakos, 1991, 1997) also holds that
consumers can become aware of all possible vendors at minimal costs. In reality, the
W is a vast marketplace, gaining full knowledge of which would require any buyer
non-trivial allocations of time resources.

An interdisciplinary perspective benefiting from the extensive literature
discussing the psychological aspects of consumer/vendor exchanges (i.e., marketing and
consumer behavior) might bring a more complete understanding of electronic market
consumers’ price search behavior and may help to explain some of the empirical findings
that are inconsistent with the theory. For example, Smith et al. (2000), proposed several
reasons for why they found price dispersion on the W to be no lower than in
conventional markets. One reason was that intemet shoppers often will turn to the
vendors of whom they are aware, because there remain costs to search and compare on
the WWW, however minimal they appear compared to the costs of search in non-

electronic markets.

Study Purpose

The price search behavior of consumers in an electronic market, thus, appears to
be an area especially ripe for empirical investigation, as well as further theoretical
explication. To contribute research in this area, this dissertation proposes and tests a
causal model of the factors influencing consumer external price search behavior in an
electronic marketplace, using an inclusive theoretical perspective that incorporates both
traditionally non-electronic and uniquely online search determinants. This study focuses
specifically on price search because of its prominence in the search behavior of online
consumers. The proposed model will be operationalized and validated in the context of
the W as it is currently the most pervasive consumer electronic marketplace.

The use of a causal model is viewed as an opportunity to assess the applicability
of traditional determinants of consumer search to an electronic shopping environment and
to test constructs believed to be uniquely salient to consumers’ online search for lower
prices. The prime advantage of estimating a causal model lies in this method’s ability to
show simultaneously the structural relationship of different constructs on consumer
electronic marketplace price search while also demonstrating the influence of these
constructs on each other (MacKenzie, 2001). The theoretical model of consumer external
information search proposed by Schmidt and Spreng (1996) was employed as a
framework within which to specify the relationship between these traditional and unique
constructs. In its specific adaptation of the mediating variable structure proposed by
Schmidt and Spreng, the present model will empirical verify a broad theoretical
perspective on the online price search phenomena that combines search costs economics

with both motivational and information processing theory.

 

 

 

theoret
the me"
results
empirit
its find

Stud)"s

 

Organization of Subsequent Chapters
Chapter 2 reviews the consumer search literature to present a comprehensive
theoretical background before specification of the proposed model. Chapter 3 discusses
the methods used to operationalize and test of the proposed model. Chapter 4 presents
results of the pilot study used to test the research instrument. Chapter 5 presents the
empirical results of model’s estimation. Chapter 6, the final chapter, recaps the study and
its findings, and draws conclusions. The dissertation ends with an assessment of the

study’s contributions and shortcomings, and with directed calls for further research.

 

 

 

 

discuss

inform;

 

second
differer
these t1

SPCCIfie

m

Chapter 2

LITERATURE REVIEW

Introduction

Given the model’s origins in consumer search theory, the first part of this chapter
discusses the development of different theoretical perspectives on consumer external
information search and determinants of search that derive from these approaches. The
second section examines models that have attempted to bring synthesis between these
different theoretical perspectives on and determinants of information search. Based on
these theoretically unifying efforts, the proposed online price search model is then
specified in section three.

Theoretical Foundations of Information Search Research

Introduction

In the broadest terms, consumer information search has been studied from the
perspectives of both economics and marketing. The economics perspective has centered
on a costs vs. benefits framework, while the marketing perspective has focused on the
development of consumer behavior models.
Economics Approach

The application of economic theory to the study of consumer search is rooted in
the costs vs. benefits framework that was pioneered by Nobel Laureate George Stigler

(1961). Stigler revolutionized economics when he challenged the assumption that buyers

10

 

 

 

actxsi
offinir
vary n
costs ;

aQaIns

beCaust
The cer:
pr013051 ‘
search 1
10w 5921?

being 31.

 

“33mm,

COHIEXt.

 

act with perfect knowledge of the marketplace, knowing all sellers and all seller
offerings. He instead proposed that there were economics to information, in that buyers
vary in how much information they possess because they place different values on the
costs and benefits of search for information about the marketplace, weighing the two
against each other.

Though Stigler’s (1961) model has been amended (Butters, 1977; Rothschild,
1974; Salop & Stiglitz, 1977; Stiglitz, 1979; Wilde & Schwartz, 1979), and many of the
model’s original dimensions have been modified (Kohn & Shavell, 1974; Ratchford,
1982; Telser, 1973; Weitzman, 1979), its basic structure is still discemable. The model
examines search for the lowest price under conditions of price dispersion in the
marketplace. At its root, the model holds that a buyer acting to optimize his or her search
engages and continues search only so long as the expected marginal return of one unit of
search exceeds the expected marginal cost of one unit of search.

The economics of information approach (Stigler, 1961) has been criticized
because it is not possible to empirically verify many of its propositions (Miller, 1993).
The central contribution of this approach to the information search literature is the
proposition that the costs of search a buyer incurs are related negatively to the amount of
search that that buyer will conduct, ceteris paribus. Stated positively, consumers with
low search costs will conduct more search than consumers with high search costs, all else
being equal.

As expressed in this proposition, the costs vs. benefits framework has been tested
as a stand-alone mechanism to explain consumer external information search. In this

context, it has received limited empirical verification with weak support at best (Goldman

11

 

 

 

 

 

 

& Job
3 basil

while

fully e

 

Signific
Underst
e118mm
TIICOI‘po
directed

emphas.

faCIOIS I

behai'lOr

 

 

& Johansson, 1978; Urbany, 1986). This may be explained by noting that on its own, as
a basic mechanism, the proposition presents a simple and parsimonious structure. Yet
while possessing an obvious intuitive appeal, this simple framework lacks the ability to
fully explain a process as sophisticated as consumer external information search. The
costs vs. benefits framework, however, is often acknowledged in the information search
literature of marketing, as it presents an excellent context within which to explain other
psychological elements that contribute to the amount of consumer external information
search. One will notice, therefore, its consistently prominent position in the development
of marketing research pertaining to information search (Srinivasan, 1990).

Marketing Approach

Consumer external information search serves a function in a larger consumer
behavior process. Because of this, theoretical perspectives on consumer behavior have
significantly influenced research into external information search. A general
understanding of the marketing approach to the study of search, thus, may be gained from
examining the theoretical development of leading consumer behavior models that have
incorporated information search into the frameworks. The role of information search, as
directed by the predominant theoretical approach of these models, therefore, is
emphasized in the following discussion.

It should be noted that while each successive viewpoint on the consumer behavior
process is built upon one dominant theoretical approach, each model also is influenced by
factors rooted in the theoretical perspectives of earlier models. In this way, it will be
shown how successive researchers have built a fuller understanding of consumer

behavior, and external search in the process, using a dominant theoretical framework

12

 

 

 

supplt

earlies
infom
search

perspe

solvini
was th
uncert.‘
would
in the c

t0 ther

behavir
PTOduc.‘
and Son
Perceiv.
Mauve
amOum
may be .
Consllmt
Opinion

assumpt:

 

 

supplemented by earlier theoretical perspectives. One also will note that even in the
earliest of these research pieces, and throughout their development, the limited
information processing ability of consumers is recognized as an influencer of information
search and choice behavior, until finally it emerges as the dominant theoretical
perspective.

To give a short overview, early researchers conceptualized shopping as problem
solving (Bauer, 1967; Howard & Sheth, 1969). From this early perspective motivation
was the driving force behind information search (Srinivasan, 1990) with prepurchase
uncertainty reduction being the consumer’s goal (Newman, 1977). Later researchers
would continue to acknowledge problem recognition and motivation as key components
in the consumer decision process (Engel et al., 1978), but would give greater recognition
to the role of information processing ability (Bettrnan, 1979; Engel et al., 1978).

To begin the more detailed discussion, Bauer (1967) proposed that consumer
behavior be understood as risk taking, “...in the sense that any action of a consumer will
produce consequences which he cannot anticipate with anything approximating certainty,
and some of which are likely to be unpleasant” (Bauer, 1967, p. 24). To reduce their
perceived risk, consumers develop strategies that enable them to make choices with a
relative sense of confidence, particularly when there is deficient information. The
amount of external information search and the degree of perceived risk reduction, thus,
may be expected to correlate highly. Mitigating this relationship, however, are
consumers’ risk alleviation strategies, which including: relying on brand loyalty, seeking
opinion leaders, and adopting social norms. Though not expressly stated, the author’s

assumption that consumers create strategies to deal with the enormity of pre-decision

l3

 

 

 

 

infom

proces

proces
that is
behavr
behavi
enviro:
learnin
ftpeati
increas
Simplif
leanin g

and w“

however
restricts.

“”9 Wii‘r—

 

information, what might otherwise be understood as heuristics, hints at the limited
processing capabilities of consumers, and how this mediates external information search.

Howard and Sheth (1969) built an integrated model of the purchasing decision
process based on learning theory. In their model, consumers move along a learning curve
that is divided into sequential segments, ranging from extensive problem solving
behavior, to limited problem-solving behavior, and finally to routinized response
behavior. The authors proposed that it is the need of consumers to simplify their
environment that motivates them to problem solve. Problem solving occurs through
learning. Learning comes as a result of consumers being in similar buying situations,
repeating purchasing decisions, and acquiring new information. As consumer learning
increases they move along the curve. In other words, consumers, motivated by a need to
simplify their environment, attempt to make purchasing decisions routine through a
leaning process that occurs as a result of previous experience in the shopping situation
and with the product. Information search decreases as consumers learn about products
and their preferences towards particular brands (i.e., move along the learning curve).
Thus, they have less need to acquire new information. Overt search in this model also is
affected by several constructs, which include: stimulus ambiguity, confidence, attitude,
and motives.

The Howard-Sheth model (1969) is another instance where assumptions about
motivations, in this case having to do with environment/choice simplification, hint,
however unknowingly, at the limited processing capacity of consumers, and how this
restricted ability leads to streamlined decision behavior to manage stimulus-response. In

line with a trend in consumer behavior research to view the consumer as a limited

I4

 

 

 

 

 

inforr

(1969'

ofthe
sequei
Once I
solvni'
l3ehef
inform:

menu

_——_

throng

recepri

devorc‘
hlsnnp
JOhnsm
conglm1
or her n
”Omen

BUSIOr‘

0n infOI’j

FITSL CC)“

 

 

information processor, Howard (1977) would later revise the basic Howard-Sheth model
(1969) to take memory (i.e., the information processing mechanism) into account.

Engel, Blackwell and Kollat (Engel et al., 1968, 1973) proposed a model (EKB)
of the consumer decision process that begins with a problem recognition stage and moves
sequentially through information search, alternative evaluations, choice, and outcomes.
Once again, motivation was the driving force of the model’s approach with problem
solving as the specific goal for consumers to engage in external information search.
Beliefs and attitudes, derived from product brand evaluations, also were said to affect the
information search process. A later version of the model (Engel et al., 1978) included
information processing in its design. Information search then could occur externally
through stimuli or internally through active memory (i.e., exposure, attention, and
reception).

Acknowledging the role of motivation, researchers of information search have
devoted considerable attention to involvement and its ability to affect information search.
In simplest terms, involvement has to do with pertinence and relevance (for reviews see,
Johnson & Eagly, 1990; McQuarrie & Munson, I992; Mittal & Lee, 1989). The greater a
consumer’s involvement with a purchasing decision, it is supposed, the greater will be his
or her motivation to search for the product that best achieves his or her goals (Beatty,
Homer, & Kahle, 1988; Beatty & Smith, 1987; Mittal, 1989; Salmon, 1986; Smith &
Bristor, 1994).

Bettrnan (1979) built a pioneering model of consumer choice based specifically
on information processing theory. The model is predicated on two major assumptions.

First, consumers are viewed as limited in their ability to process information. The basic

15

 

 

 

 

efiect
deveh
requir
consur
ofthe
consur
proces
inchuh

mepm

phonh
In10rm
kanun
bonini
b€fore.
10 Tiller
“den.

percent

has led r

111161713 I

 

familiar;

551's L;

 

effect of depicting consumers as limited capacity processors is that they are assumed to
develop heuristics that allow them to manage complicated circumstances, so as not to
require more processing capacity than is available. The second major assumption is that
consumers are goal oriented. Achieving the larger shopping goal(s), goal(s) at each stage
of the decision process, and sub-goals that arise during the decision process motivates
consumer behavior. One should note how Bettman, though advocating information
processing as the central phenomenon in the route to decision, still acknowledges and
includes the role of motivation as a guiding factor in his model, and in doing so builds on
the preceding theoretical approaches.

Consumers navigate the choice process through a goal hierarchy function that
prioritizes these goals at each stage of the model. The stages of the model are attention,
information acquisition and evaluation, decision processing, and consumption and
learning processes. In the information acquisition and evaluation stage, search occurs
both internally (i.e., memory) and externally. A complete internal search is not necessary
before an individual may decide to initiate an external search, and consumers may return
to internal search as they acquire new pieces of information. In this way, internal search
and external search are an intertwined process. Total search continues until an individual
perceives that he or she possesses sufficient information to a make a decision.

The information processing approach to the study of external information search
has led researchers to explore the influence of memory in the relationship between
internal and external search. Research in this vein has considered prior knowledge,
familiarity, and experience as related determinants of external search (Brucks, 1985;

Fiske, Luebbehusen, Miyazaki, & Urbany, 1994; Johnson & Russo, 1984; Lee, Herr,

16

 

 

 

 

Kard L
deten

appro

Hutch

literai
pnorl
betwer
these r
be und

I’Brucl.

Iheorer
Inform;
based p
decade.
determ1

|
|
I for m I
|

8: Lehn.

fiHed “I

ftsearch

 

Kardes, & Kim, 1999; Punj & Staelin, 1983; Rao & Sieben, I992; Urbany, 1986). Other
determinants of external search that have derived from the information processing
approach include prior beliefs (Duncan & Olshavsky, 1982) and expertise (Alba &
Hutchinson, 1987; Punj & Srinivasan, 1989).

Of these determinants, prior knowledge has been examined extensively in the
literature. Early research showed conflicting data between consumers low and high in
prior knowledge. Later research demonstrated an inverted “U” shape relationship
between prior knowledge and the amount of information search. In an effort to address
these results, recent conceptualizations of prior knowledge have proposed that knowledge
be understood as two distinct constructs: objective knowledge and subjective knowledge
(Brucks, 1985; Park, Mothersbaugh, & Feick, 1994; Spreng & Olshavsky, 1989).

Comprehensive Frameworks Explaining Information Search

The models of information search and consumer behavior discussed above lent
theoretical direction for research into determinants of the amount of consumer external
information search. From these theoretical approaches, as well as other non-theoretically
based perspectives, researchers studying consumer information search over the past four
decades have demonstrated a relationship between information search and more than 40
determinants, the major and most relevant of which were detailed in the preceding section
(for reviews see, Beatty & Smith, 1987; Bettman, 1979; Guo, 2001; Miller, 1993; Moore
& Lehmann, 1980; Newman, 1977).

Despite this wide—ranging literature, consumer information search remains an area
filled with unanswered questions and theoretical contradictions. The difficulties

researchers have encountered attempting to operationalize search and create valid

17

 

 

 

 

measr
of the
1989 I
there
anent:
work
on the

relatio

 

 

 

 

comp]
Olshai

asapr

comp,
Unden
I973;
e.‘Iami
dlrEct
Shucn
eXtern
Punj a
mean
bfinhei

measures to represent theoretically based determinants have contributed to the existence
of these unsettled issues (Newman & Lockeman, 1975; Urbany, Dickson, & Wilkie,
1989). Yet beyond these challenges, and notwithstanding an otherwise broad literature,
there is a conspicuous lack of comprehensive frameworks within which researchers have
attempted to explain the relationship of the different hypothesized determinants as they
work concomitantly in the information search process. In fact, so much of the research
on the determinants of consumer information search has focused on bivariate
relationships, which have not proven robust when tested simultaneously, that any
complete understanding of the search process has eluded our knowledge (see Duncan and
Olshavsky, 1982, p. 32 for a discussion of this issue). This deficiency has been described
as a pressing research need (Wilkie & Dickson, 1991).

A limited number of researchers have attempted to build broader based and more
comprehensive frameworks within which consumer information search might be
understood more fully through the application of covariance structural analysis (Joreskog,
I973; Joreskog & Sorbom, 1981). The advantage of this method lies in its ability to
examine simultaneously the relationships among determinant constructs and reveal both
direct and indirect paths of association (MacKenzie, 2001). Research employing this
structural modeling approach has significantly contributed to our understanding of
external information search (Blodgett, Hill, & Stone, 1995; Maute & Forrester, 1991;
Punj & Staelin, 1983; Srinivasan & Ratchford, 1991; Sundaram & Taylor, 1998). This
research has gone beyond earlier studies that merely measured bivariate relationships
between a number of different determinants and search. It has demonstrated the affect

determinants have on each other in relation to search. This has enabled the bringing

18

 

 

togetl

create

offers
Schm
detem

above

perspe

CXpres
compl-

COnSUr

 

 

together of disparate yet related ideas on the determinants of consumer external search to
create broader understanding.

Still broader perspectives on consumer search have been proposed in research that
offers theoretically integrated approaches (Moorthy, Ratchford, & Talukdar, 1997;
Schmidt & Spreng, 1996). Frameworks in this vein not only incorporate multiple
determinants of consumer search (e. g., involvement or knowledge), as the studies noted
above have done, but also integrate constructs representing the major theoretical
perspectives on the study of search (e. g., motivation or ability).

Moorthy, Ratchford, and Talukdar (1997) propose a framework that is intended
expressly to bring together the economic and psychological approaches to create a more
complete understanding of external information search. In this model, the authors view
consumers as rational and adaptive decision makers faced with a brand choice.
Consumers develop search strategies by weighing the costs against the benefits of search.
“The benefit of search is driven by how a consumer perceives the uncertainty in her
choice environment (problem framing), the importance she gives to the product category
(what is traditionally referred to in the behavior literature as “involvement”), and her risk
aversion” (Moorthy et al., 1997, p. 264). This framework highlights the role of problem
framing, which the authors believe is the key to explaining why highly involved
consumers with low search cost typically conduct little search. They find that the need to
search develops only when the consumer does not possess a perception of which brand is
best, what the authors term relative brand uncertainty.

Schmidt and Spreng (1996) propose the most extensive theoretical model of

external information search. The advantage of the Schmidt and Spreng model, presented

19

 

 

 

in F i g
these
The n

|
motit

1980i

Cacioi

Effect,

search

 

Present
and ecc
these r

Ollilne

in Figure 2.1, is that it incorporates nearly 20 determinants of search and proposes that
these determinants affect search through one of four mediating psychological constructs.
The mediating constructs that directly affect search are perceived ability to search and
motivation to search, based on the information processing literature (Bettman & Park,
1980) with additional support from the Elaboration Likelihood Model (ELM) (Petty &
Cacioppo, 1986). The two other mediating constructs are perceived benefits of search
and perceived costs of search, borrowed from the economics approach. The determinants
mediated through these four constructs derive from the theoretical perspectives of
cognitive psychology based in information processing, of social psychology based on
motivational research, and of economics based on the costs vs. benefits framework.
These determinants include, for example, knowledge, risk, and involvement.

The four mediating variable structure of the Schmidt and Spreng model (1996), in
effect, unifies the three major theoretical approaches examining consumer information
search discussed previously. In addition, its broad inclusion of determinants of search
presents a framework within which to specify the causal relationship of individual, social,
and economic constructs shown to affect the amount of consumer external search. For
these reasons, the model of consumer price search behavior in an electronic marketplace
proposed in this study, which seeks to empirically verify a broad understanding of the

online price search phenomenon, will be based on this mediating variable structure.

20

 

IA—rkuvsrv\ 11- l—Irqi--il!‘ intct.u¢s:t~—r.i— to: Ice oarswsv:'—ns~h F—uruasn‘U a.\;\.—Lflh

 

N .N 043%.;

 

I lili-

 

eozmcmou SM @002

«$089,735: “—0602:

Emfimsficm wafiaonm 33>?

 

       

:Baom no: 0530 we
squad 8283

 

598$ 35 25:0 8

    

5.3 333 a can...

  

3:85.800 Howe—Sm 32080;

sigma out.— 33023

     

lot: maria 2 328.5;

 

 

  
 

seam 85 255 a

flaw 3388A

com—98.32 oEoboo—m
8_>Eom ; . m

 

 

        

noun—EOLE
25:0 9 mmooo< .3 8mm

8:38“— oEF nozoobm

camom BEBE

 

5.86m 35 25:0 2

3% 8285

 

 

 

EL 839832. 6. 305.85 an E Hogan—om noumom ootm _mfioaxm 685280 mo 6on meme 0.5

we ammo—399 3:835

 

3 05mm

21

 

 

preset
influe
10 vai
variab
Arron
also h.
positix
follou
on the
Strucru
reason;
PITCe 3(
Search.

(e.g‘ “I

 

 

 

Model Specification

Building on the above discussion, this study proposes a structural equation model,
presented in Figure 2.2. The model is composed of 14 determinant constructs believed to
influence price search in electronic marketplaces. The structure of the model is such that
10 variables are hypothesized to influence online price search through four mediating
variables. Each line in Figure 2.2 represents a relationship between constructs.
Arrowheads at the end(s) of each line indicate the direction of the relationship(s). Lines
also have been labeled to show the hypothesized sign of their path coefficient (i.e.,
positive or negative). Each line corresponds to a hypothesis that will be detailed,
following the literature and reasoning that supports it. The proposed model relies heavily
on the Schmidt and Spreng model (1996) with regards to its four mediating variables
structure. Therefore, only moderate modifications have been made to the original
reasoning explaining the relationships between these four mediating variables and online
price search behavior. One should note that the proposed model examines general price
search, and thus does not explore external price search for any specific product class

(e.g., new automobiles or digital cameras).

22

 

 

 

N .N 0.5.4.:

 

EmstIPZm ZOHEZOOU H2m2m>qo>z~
OZELOIm «Om Qmmz OZEDDZM

 

   
   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Manama: £5
+ + + momfim AIIH11_ Ezaammooac zofi<2¢0az~ _
no $80 A]
828%... + 3:84.38 5395 _
+ m5 Em 820$ _
- i +
20563 22220 mo; 9:me
282% oh
292252 + 20563 :52 OH 952 L
+ + mmozmamta 5:99: amZmoxma
+ In i
+ :83; I + “ 85:0 mo". 92592 292320qu _
no $5me
828%... Emzm>no>a $29235 _
>:>:o< + +
:umfim + Vim 828mg
20.33285

n<zmmfixm - moaaoé $62542: Samoa”: _
. zofiufimfifi _

+ +
moomnaozz mZBmEDm _

zomfim +
oe 53:2 “ men—33020. mZSmEO _

828mg +
295028 _

 

 

:2 m as “28an
Z DEE

23

 

 

 

 

":5

\, ,

exenT
shopp
dfisxx
behaV
beforJ
I“0P01

condu

 

definn
Whfle'
Houa

inflonq

Online Price Search Behavior

Electronic marketplace price search is defined as the degree of effort a buyer
exerts to acquire and compare the price offerings of sellers in a computer mediated
shopping environment (adapted from Urbany, Dickson, & Kalapurakal, 1996, p. 92). In
this way, online price search is conceptualized in a broad manner that recognizes the
behavioral aspect of search and the conscious psychological process a buyer may engage
before deciding to comparison price shop. One will note from this definition that the
proposed model particularly concerns the amount of online price search an individual
conducts prior to making a purchase on the WWW. Further, it should be noted from this
definition that the proposed model specifically examines external price search, which,
while having been conceptualized in many ways (Bettman, 1979; Engel et al., 1968;
Howard & Sheth, 1969; Nicosia, 1966), has consistently been viewed as the product of
information seeking, gathering, and processing from non-memory based sources
(Srinivasan, 1987).
Perceived Ability to Online Price Search

Ability to price search online is defined as an individual’s perceived cognitive
capability of searching for and processing price information. (adapted from Schmidt &
Spreng, 1996). This definition is meant to encompass not only consumers’ perceived
cognitive processing aptitude toward online price search, but also their prior knowledge
of methods for online price search and of sources of information. This conceptualization
is consistent with Maclnnis, Moorman, and Jaworski (1991), who, in studying

consumers’ ability to process brand information from advertisements, viewed ability as

24

 

 

perta'

role c

inforr
exper
group
ability
Olsha
to era
condu

(1996

 

edUCaI;
Udell. ‘

6550“,

 

pertaining to a consumer’s skill in interpreting information, while noting the important
role of prior knowledge in processing proficiency.

A number of researchers have linked cognitive processing ability with external
information search behavior. Bettman and Park (1980) in explaining the results of an
experiment that manipulated prior knowledge and information search theorized that
group differences were the result of individuals’ varying levels of information processing
ability. In a study of the external search behavior of television purchasers, Duncan and
Olshavsky (1982) found that consumers with greater perceived confidence in their ability
to evaluate product and brand information, what they termed “ability to judge”,
conducted more search. Based on this evidence, and consistent with Schmidt and Spreng

(1996), the following is hypothesized.

H1: Perceived ability to price search on the WWW will be positively related to

online price search behavior.

Two antecedent variables are believed to influence consumers’ perceived ability
to online price search. These are consumers’ educational level and their subjective
knowledge of intemet search.

Education. Numerous studies have found an association between higher levels of
education and increased amounts of consumer information search (Claxton, Fry, & Portis,
1974; Katona & Mueller, 1955; Kiel & Layton, 1981; Schaninger & Sciglimpaglia, 1981;
Udell, 1966). While methodological differences have produced varying levels of

association, the positive relationship between education and search in these studies is

25

 

C0115]
pOSSC
“1651

to fin

conce
result:
1987;
subjeg
Specif;
There:
aSSCSSr:

acquis

 

 

consistent. In addition, other research has shown that households with members
possessing higher levels of education search with greater efficiency (Michael, 1972).
These studies suggest that individuals with more education will have an increased ability

to find, recognize, and process pertinent data in an information task like price search.

H2: Higher levels of education will be positively related to perceived ability to

price search on the WWW.

Subjective knowledge of intemet search. Subjective knowledge typically is
conceptualized as an individual’s perceived knowledge of a particular product class
resulting from his or her experience with that category of products (Alba & Hutchinson,
1987; Brucks, 1985; Spreng & Olshavsky, 1989). For the purposes of this research,
subjective knowledge is adapted to assess an individual’s perceived knowledge of a
specific task resulting from his or her experience with that task (i.e. intemet search).
Therefore, subjective knowledge of intemet search is defined as an individual’s
assessment of his or her proficiency at utilizing the W to perform information
acquisition tasks.

Subjective knowledge is the result of experience and involves confidence
(Schmidt & Spreng, 1996). Knowledge is generated through experience as consumers
engage in product usage. Usage triggers a learning process (i.e., a knowledge acquisition
process) whereby the user moves from a state of familiarity to expertise (Alba &
Hutchinson, 1987). For the purposes of this research, product usage is exchanged for

task performance. From this perspective, one may conceptualize the knowledge

26

 

 

 

 

 

acqu

relatt

 

 

 

|
thatt

subjc
levell

Cons

COIlfh

 

acquisition process as one in which individuals, through task performance, develop task
related knowledge structures (i.e., skills) that lead to increase cognitive ability to conduct
that task. Confidence is also a significant factor because in the process of reporting their
subjective knowledge, individuals are not only gauging the adequacy of their knowledge
level, but are also indicating their self-confidence in that knowledge level (Brucks, 1985).
Consumers who considered themselves knowledgeable shoppers perceive greater
confidence in their ability to evaluate product and brand information (Duncan &
Olshavsky, 1982). Subjective knowledge, thus, is most accurately a measure of an
individual’s confidence in what he or she believes they know. It is hypothesizes,
therefore, that subjective knowledge of intemet search and perceived ability to online

price search should have a positive association.

H3: Subjective knowledge of intemet search will be positively related to

perceived ability to price search on the WWW.

Perceived Motivation to Online Price Search

Motivation to online price search is defined as an individual’s perceived desire to
exert effort gathering and processing price information in an electronic marketplace,
including both the direction and the intensity of the effort (adapted from Schmidt &
Spreng, 1996). The direction of the motivation refers to the different types of
information a consumer might consider before exerting effort to gather and process price
information. The intensity of the motivation refers to the varying level of an individual’s

drive toward goal fulfillment.

27

Goal orientation is a central component in this definition of motivation. This is
consistent with Park and Mittal (1985), who view motivation, in terms of consumer
involvement, as “goal directed arousal”. Bettman (1979) also purports a goal-oriented
perspective to motivation. In his theory of consumer choice, Bettman highlights the role
of motivation in goal fulfillment, noting that, “choices are made to achieve certain
purposes, or accomplish some goals” (Bettman, 1979, p. 18). From Bettman’s
perspective motivation influences behavior in (1) that its intensity affects actions (e.g.,
how much time is allocated to complete a specific task) and (2) that it directs particular
choices over others.

In the proposed model, motivation to online price search is hypothesized to have a
direct positive affect on online price search behavior, while also acting as a mediating
variable for a number of distinct constructs, which are believed to contribute to
motivation’s direction and its intensity. Schmidt and Spreng (1996) note research that
suggests motivation to search be viewed as a sub-goal within the context of a specific

consumer behavior episode (Olshavsky, 1985; Spreng & Olshavsky, 1989).

H4: Perceived motivation to price search on the WWW will be positively

related to online price search behavior.

This model proposes that motivation to online price search is influenced by a
number of distinct constructs, which contribute to its direction and its intensity: intemet
involvement, need for cognition (NFC), WWW shopping enthusiasm, benefits of online

price search, and costs of online price search.

28

 

(199%
$313
cons
actth
indei
unern

useH’

motiv
percei
goals

percei
Zakifi

dEgrec-

Internet involvement. Internet involvement is a construct first proposed by Salam

 

(1998) to assess the varying levels of value consumers place on the intemet in their lives.
Salam believes that this value level is a crucial component in understanding online
consumer behavior because the extent of an individual’s use of the intemet is a deliberate
act that at a base level is motivated by an individual’s belief that the medium will be of
aide in achieving goals or desired states. He argues that through an understanding of
intemet involvement researchers will be able to begin assessing the affect of the medium
itself on individuals’ behavior in relation to electronic commerce.

Salam (1998) defines intemet involvement as, “an unobservable state of
motivation of a person regarding the Internet or World Wide Web and is his or her
perceived relevance related to the Internet based on inherent needs, values, interests,
goals and objectives” (p. 45). Central to Salam’s concept of intemet involvement is
perceived personal relevance (Barki & Hartwick, 1989; Petty & Cacioppo, 1981;
Zaichkowsky, 1985). The intemet becomes personally relevant to a consumer to the
degree it functions as an instrument involved in executing his or her objectives or
achieving his or her goals. One should expect an individual to have feelings of greater
personal relevance with the intemet (i.e., higher intemet involvement) the more features
of the WWW become an integral part of processes to execute objectives or achieve goals
(e. g., checking the weather, paying bills, or shopping). It is reasoned that consumers with
high intemet involvement recognize the W as a multi-purpose tool that facilitates
shopping tasks, like price search. Therefore it is hypothesized that intemet involvement

will be positively associated with motivation to online price search.

29

H5: Internet involvement will be positively related to perceived motivation to

price search on the WWW.

Need for cognition. Cacioppo and Petty define need for cognition (NFC) as “. . .a
tendency for an individual to engage in and enjoy thinking” (Cacioppo & Petty, 1982, p.
116). NFC has been shown to be a factor in information acquisition in non-electronic
markets (Inman, McAlister, & Hoyer, 1990; Verplanken, Hazenberg, & Palenewen,
1992). Jones and Vijayasarathy (1998), studying differences in the opinions of subjects
using a print vs. a WWW version of a catalog, found that high NFC subjects placed a
greater emphasis on using information in their choices and preferred both print and online
catalog shopping. The authors note that this implies high NFC individuals should possess
more positive attitudes toward in-home shopping media because of their information
richness. This research suggests that NFC is a motivational antecedent of external
information search, and will have a positive association to consumers’ motivation to

online price search.

H6: Need for cognition will be positively related to perceived motivation to

price search on the WWW.

WWW shopping enthusiasm. WWW shopping enthusiasm is defined as the
enjoyment an individual feels for the task of collecting and processing online information
about products (adapted from Schmidt & Spreng, 1996). Consumers are not purely goal-

oriented value-maximizing entities (Babin etal., 1994; Parsons, 2002). Individuals have

30

 

been
of fu:
hedo:
Chara
(Bab:
becat
shOpr
Freed
lounr
Ch”):
consu

‘WWAWJ

asseSsr
SaCfif‘K
from S
incOtis

”996)

 

been shown to shop as a result of functional motives, non-functional motives, and a mix
of functional and non-functional motives (Westbrook & Black, 1985) that result in both
hedonic and utilitarian outcomes (Babin et al., 1994). Shopping enthusiasts are
characterized as consumers motivated by hedonic values, like escape or increased arousal
(Babin etal., 1994). Product purchase for this group is incidental in the shopping process
because they enjoy the experience of shopping. There is evidence suggesting that
shopping enthusiasts seek lower priced goods as an outcome (Babin et al., 1994; Darke &
Freedman, 1995). In addition WWW shopping enthusiasts have been shown to seek
lower priced goods as an outcome of their online browsing behavior (Wolfinbarger &
Gilly, 2001). Based on this research it is hypothesized that W shopping enthusiasts,
consumers who are high in experiential/hedonic motivation when they shop on the

WWW, will also possess greater motivation to online price search.

H7: WWW shopping enthusiasm will be positively related to perceived

motivation to price search on the WWW.

Perceived Costs of Online Price Search

Perceived costs of online price search are defined as a consumer’s subjective
assessment of the monetary expenditure, time sacrifice, physical effort, and psychological
sacrifice that he or she expends searching for price information on the WWW (adapted
from Schmidt & Spreng, 1996). Studies have linked higher search costs with decreases
in consumer search (Punj & Staelin, 1983; Srinivasan, 1986). Schmidt and Spreng

(1996) theorized that the perceived costs of information search affect search activity

31

through the motivation to search construct, reasoning that as consumers perceive their
search costs to be increasing they will posses less motivation to search (Bettman, 1979;
Stigler, 1961). Likewise it is hypothesized in the proposed model that the relationship
between perceived search costs and online price search behavior is mediated through the

motivation construct.

H8: Perceived costs of price search on the WWW will be negatively related to

perceived motivation to price search on the W.

A number of distinct constructs are hypothesized to influence costs of online price
search: subjective knowledge of intemet search, perceived time constraints, and ease of

access to online information.

Subjective knowledge of intemet search. A negative relationship is expected to
exist between subjective knowledge of intemet search and the perceived costs of price
search. This is because experience has been shown to develop a consumer’s knowledge
base, allowing the individual to draw upon that information in similar situations, and thus
freeing cognitive ability to be directed toward task processing (Alba & Hutchinson,
1987). Since individuals highly experienced in online search presumably would possess
a level of familiarity with the task, and thus consider themselves high in subjective
knowledge of online search, it is anticipated that they will execute online price search
with reduced cognitive effort, lowering their search costs. For example, Ward and Lee

(2000) found that less experienced WWW users to be less proficient at online

32

COSIS

the m
CV'ldC‘l
Muel‘;
1971;
PRESS].
COHSU
is mm
t0 ach
COnSUl
an Clet
in this
OPPOn
time Cal

COHSUYI

 

information search, with (1) lower search success rates, (2) less time spent searching, and
(3) shorter time before search was terminated. They reasoned that this lack of online
search proficiency was the result of users’ search being more costly, verses the search

costs of more experienced (more search proficient) users.

H9: Subjective knowledge of intemet search will be negatively related to

perceived costs of price search on the WWW.

Time pressure. Time pressure refers to a consumer’s perception of the amount of
the time he or she has available to make purchasing decisions. Despite mixed empirical
evidence (Beatty & Smith, 1987; Blodgett et al., 1995; Claxton etal., 1974; Katona &
Mueller, 1955; Moore & Lehmann, 1980; Newman & Staelin, 1972; Newman & Staelin,
1971; Putrevu & Ratchford, 1997; Wilkie & Dickson, 1991), it is believed that time
pressure to purchase will be an important factor weighed against the costs of search when
consumers are deciding to engage in online price search. The extent to which a consumer
is motivated to search for a lower price depends upon the net gain that he or she expects
to achieve from price search (Ratchford, 1982; Stigler, 1961). To achieve this gain a
consumer considers the costs of search, since search it is not a costless activity, even in
an electronic marketplace where search costs are greatly reduced. Time is a central factor
in this consideration of costs to search because time is a limited resource that represents
opportunity for the consumer (i.e., opportunity costs of time). Once allocated to search,
time cannot be spent in other activities that might generate value in other ways for the

consumer. Time is oflen conceptualized in this manner, that is, as a resource a consumer

33

expc
pCIC;
lngh

0min:

defin
uibe
Consu
thatir
“Upor
organ
50Urc;
1982;
Starch
hfibrn
delay |

d0“n1

 

wnhel

dday ‘.

 

expends as part of the cost of the search, the value of which is determined by its
perceived availability. Therefore, when a consumer shopping on the WWW perceives a
high degree of time pressure, it is expected he or she will experience higher costs of

online price search.

H10: Perceived time pressure will be positively related to perceived costs of

price search on the WWW.

Ease of access to online informption. Ease of access to online information is
defined as the extent to which a consumer seeking information on the WWW perceives it
to be available with minimal delay. Ease of access as discussed here is a matter related to
consumer perceptions of the time it takes to load WWW pages. Bettman (1979) notes
that information’s accessibility, as much as its availability, needs to be considered as an
important environmental factor affecting consumer external search. Studies from
organizational communications indicate that the perceived accessibility to information
sources is positively associated with the use of those sources (Culnan, 1983; O'Reilly,
1982). Sundaram and Taylor (1998) propose a structural model of external information
search for “in-home shopping situations” in which they found ease of access to sources of
information positively affected search. In terms of the WW and specifically perceived
delay accessing information, Rose, Lees, and Meuter (2001) found that increasing the
download time for a WWW page resulted in users prematurely aborting the loading page
with greater frequency. The authors also found that as attitudes toward page-loading

delay worsened, the likelihood increased of users prematurely aborting the loading pages.

34

(llatc
searcl
t0 58L:
PUI’C 1’:
3an
mebu
decist
ShOUl (.
ackno.
cred”,
CODSU;

gained

 

It is proposed that consumers will view WWW page downloading time as part of the
costs of online price search, and that delays are expected to increase the perceived costs
of this search. It is expected that consumers who experience less delay loading WWW
pages (i.e., high ease of access to information) will conduct more online price search

because they will realize the decreased costs of price search.

H11: Ease of access to online information will be negatively related to

perceived costs of price search on the WWW.

Perceived Benefits of Online Price 36%

Search in this model is viewed as a process whose intended result is a better buy
(Ratchford & Srinivasan, 1993), which, given the proposed model’s focuses on price
search, makes the better buy necessarily one at a lower price. There are financial benefits
to search (Ratchford & Srinivasan, 1993), however online consumers have been shown to
purchase from sellers who are not the low price leaders of their search (Smith &
Brynjolfsson, 2001). The perceived ability of an online seller to facilitate and complete
the buyer-seller exchange appears to be a key factor in online consumer’s purchasing
decision. A conceptualization of the perceived benefits of online price search, therefore,
should not be limited to the monetary return of search. Rather, it also should
acknowledge the less apparent role of the buyer’s perceived confidence in a seller’s
credibility. Perceived benefits of online price search, therefore, is defined as a
consumer’s perceived utility gain potential and exchange self-confidence. The utility

gained in this case would be monetary (i.e., amount saved). The exchange self-

35

cont
lrorr
fort
ther
cons
beyc
benc
rese.
andt
Sunt
hypo

onhr

numl
dlSpe.

dlSCU

in {OI-T

flam

 

confidence would be a matter of a consumer’s belief that the utility gain can be obtained
from a particular online seller. One should note that this definition is multi-dimensional
for the purpose of explaining the theoretical position of benefits of online price search in
the proposed model. Its operationalization, however, may be uni-dimensional, since
consumers are unlikely to consciously acknowledge price search as having benefits
beyond potential monetary saving. This is consistent with other conceptualizations of
benefits to search (Srinivasan, 1987; Srinivasan & Ratchford, 1991). A number of
researchers have found direct positive associations between perceived benefits to search
and degree of search (Duncan & Olshavsky, 1982; Srinivasan & Ratchford, 1991;
Sundaram & Taylor, 1998). Consistent with the Schmidt and Spreng model (1996), it is
hypothesized that the relationship between perceived benefits of online price search and

online price search behavior is mediated through the motivation construct.

H12: Perceived benefits of price search on the WWW will be positively related

to perceived motivation to price search on the W.

This model proposes that benefits of online price search is influenced by a
number of distinct constructs: perceived costs of online price search, perceived price
dispersion, perceived budget constraints, and trust in W sellers. The following is a
discussion of each construct and its link to the benefits construct.

Benefits vs. costs of online price search. As noted previously, the economics of
information approach is rooted in the costs vs. benefits framework (Stigler, 1961). This

framework is ofien incorporated into marketing and consumer behavior studies of

36

information search (Goldman & J ohansson, 1978; Ratchford, 1982; Srinivasan, 1990;
Srinivasan & Ratchford, 1991; Urbany, 1986). According to the framework a buyer
engages and continues search only so long as the expected marginal return of one unit of
search exceeds the expected marginal cost of one unit of search. Therefore benefits of
online price search and costs of online price search are expected to have an inverse

relationship.

H13: Perceived costs of price search on the WWW will be negatively related to

perceived benefits of price search on the WWW.

Perceived price dispersion. Perceived price dispersion is a consumer’s subjective
assessment of the variance of prices in an electronic marketplace. According to
traditional search research (Duncan & Olshavsky, 1982; Goldman & Johansson, 1978;
Urbany, 1986; Zimmermann & Geistfeld, 1984) buyers estimate the variance of available
prices in a market (i.e., perceived price dispersion) to assess the potential financial return
from search effort. Consumers then weigh this estimate against the costs of search to
determine if they should engage in search, and if so, to what degree. In reality,
consumers typically possess limited information about actual prices, even for frequently
purchased products like grocery items/package goods (Goldman, 1977; Urbany, Dickson,
& Sawyer, 2000; Wakefield & Inman, 1993). Goldman and Johansson (1978) note that
consumers, given this common state of imperfect price knowledge, likely develop rough
price approximations and use these in calculating if their search will yield returns with

value in excess of the effort (e.g., Nystrom, Tamsons, & Thams, 1975). Thus consumers

37

 

 

wil

bel:

infc
015
the:
in t
eXp

dis;

sub;
fina
EXP!
€ng
bece
relat

Sean

will engage in more search when they perceive significant price differences because they
believe potential monetary gains exist in a market.

With exception (Urbany et al., 1996), support for this part of the classic
information economics theory can be regarded as weak (Claxton et al., 1974; Duncan &
Olshavsky, 1982; Goldman & Johansson, 1978). However, it has been suggested that
these results would benefit from perspectives that integrate non-economic based factors
in the consumer decisions process (Urbany, 1986). Given the proposed model’s
expansive theoretical perspective and its focus on price, it is believed perceived price

dispersion will possess a positive relationship to perceived benefits of online search.

H14: Perceived price dispersion will be positively related to perceived benefits

of price search on the WWW.

Perceived budget consflints. Perceived budget constraints are a consumer’s
subjective assessment of his or her general financial state. Consumers who view their
financial state as weak are believed to benefit more from search, and therefore are
expected to price search more often. When an item(s) being purchased represents a large
expenditure in relation to a buyer’s budget the buyer will search more for lower prices,
because there exists the potential for greater relative savings (Stigler, 1961). The
relationship between size of expenditure, typically measured as purchase price, and
search behavior has been examined in a number of studies (Bucklin, 1966; Dommermuth,
1965; Dommermuth & Cundiff, 1967; Katona & Mueller, 1955; Kiel & Layton, 1981;

Newman & Staelin, 1972; Udell, 1966). Urbany et al. (1996) note however that while

38

 

_— L__.V_ .—

budg
expe:
hous
“8W1
ivoul
FoHc
pnce
ages

COHS

“D
O
{/1
(I)

decn

budget constraints’ effects on price search can be measured using the size of the
expenditure, other factors (e. g., household size) cause expenditures to vary across
households. Likewise Schmidt and Spreng (1996) believe price alone is a particularly
flawed antecedent measure of search benefits. A more critical examination, in their view,
would account for the relativity of price given a consumer’s financial constraints.
Following Urbany et a1. (1996), rather than focus on the size of the expenditure (i.e.,
price), the perceived budget constraints construct in the proposed model is meant to
assess consumers’ general perceptions of monetary constraints. It is thus reasoned that
consumers using the WWW who perceive their monetary resources as constrained
possess greater incentive to search online for lower prices, valuing the potential for

decreased expenditure as a benefit of search.

H15: Perceived budget constraints will be positively related to perceived

benefits of price search on the WWW.

Trust in W sellers. Trust has been noted as an “order qualifier” for purchase
decisions (Doney & Cannon, 1997). In other words, a consumer must trust a seller before
he or she places an order. Trust is the buyer’s expectation that the seller can be relied
upon to fulfill their part of the exchange and that the seller will not exploit the buyer’s
vulnerability (Geyskens, Steenkamp, & Kumar, 1996). Research suggests that lower
prices alone are insufficient to persuade some consumers to trust unknown WWW sellers
(Van den Poel & Leunis, 1999; Whelan, 2001). Even in those cases where online

consumers conduct price search, buyers do not always purchase from the low price

39

 

lead

300?

\VhO
shot
H331
one:
hkel
Bust
the):

reah.

leader, and they often use brand as a proxy for retailer credibility (Smith & Brynjolfsson,
2001)

It is clear that trust in WWW sellers is still an issue, even for those consumers
who have overcome their initial fears of transaction and privacy risk to become online
shoppers. Trust in W sellers appears to vary in individuals, and this variance appears
to affect consumers’ online choice behavior. Based on the previous mentioned research,
one should expect that consumers with high levels of trust in W sellers will be more
likely to patronize a broader spectrum of online sellers. It is reasoned that these high
trust consumers will also perceive the benefits of online price search to be high, because
they are inclined to purchase from a broader spectrum of sellers and therefore more ofien

realize actual savings.

H16: Trust in W sellers will be positively related to perceived benefits of

price search on the WWW.

Structural Etmations

A system of five structural equations emerges from the preceding model
specification, each based on one of the five endogenous variables in the proposed model.
These endogenous variables are (1) price search behavior in an electronic marketplace,
(2) perceived ability to online price search, (3) perceived motivation to online price
search, (4) perceived benefits of online price search, and (5) perceived costs of online
price search. The following five equations represent the system of structural equations

comprising the proposed model of consumer external price search behavior in an

40

elect

ther

electronic marketplace. The signs enclosed within brackets denote the expected sign of

the relationship. All constructs in the proposed model are treated as latent variables.

1)

2)

3)

4)

5)

Price Search in an
Electronic
Marketplace

Perceived Ability to
Online Price Search

Perceived Motivation
to Online Price
Search

Perceived Benefits of
Online Price Search

Perceived Costs of
Online Price Search

f(

f(

f(

f(

f(

perceived ability to online price search (+),
perceived motivation to online price search (+) )

education (+),
subjective knowledge of intemet search (+) )

intemet involvement (+),

need for cognition (+),

W shopping enthusiasm (+),

perceived benefits of online price search (+),
perceived costs of online price search (—) )

perceived price dispersion (+),
perceived budget constraints (+),
trust in W sellers (+) )

perceived benefits of online price search (—),
subjective knowledge of intemet search (—),
perceived time pressure (+),

ease of access to online information (—) )

41

Chapter 3

METHODS

Introduction

This chapter presents the methods used for data collection and analysis. It also
outlines the sampling procedures employed and the development of the research
instrument. The survey method, involving a convenience sample of students from a large
Midwestern university, was used for data collection. Data collection occurred in two
phases: (1) data were collected specifically to pilot test the research instrument; (2) data
were collected for estimation and analysis of the full causal model. Results of both
phases of the data collection are presented in subsequent chapters.

Data Collection Procedures

Data were collected for the pilot study in early February 2003. Students from two
upper level marketing undergraduate courses were recruited for the pilot, which was
administered afier class time. Students received class credit in exchange for their
voluntary participation. A total of 127 qualified students completed the survey. A
participant was considered qualified if he or she had made at least one purchase through
the W in the three months preceding questionnaire administration.

Data for the main study were to be collected using a survey that participants
accessed via the WWW. Technical issues developing the WWW version, however,

delayed its deployment. In the meantime, surveys were begun to be collected in paper in

42

March 2003. Within the month, a sufficient number of qualified surveys (N=728) were
collected in paper form to test the causal model and establish statistical power
(MacCallum, Browne, & Sugawara, 1996). With no need for further data, the decision
was made to abandon plans to administer the survey via the W. This decision also
was believed to be advantageous to the data’s integrity, since it was observed during the
pilot that, unless a proctor was present, participants tended to give the survey cursory
treatment.

To recruit participants for the main study, professors in a College of
Communication and a Department of Romance Languages were first contacted to request
access to the various sections of their classes. A standard protocol was established
whereby the researcher arrived five minutes prior to the end of a class. The researcher
was presented to the students at the end of class time. The researcher then introduced the
study and explained its purpose to the class. Students also were made aware of the
monetary incentive for participation in the form of a raffle for a $100 cash prize. In
addition, nearly all professors offered some form of extra credit for participation.
Afterwards, students were asked for their voluntary participation to complete the survey.
Students who chose not to participate in the study were asked to leave to the classroom
before surveys were distributed. As with the pilot, a qualified participant had to have
made at least one purchase through the W in the three months proceeding
questionnaire administration. This process produced 728 qualified surveys.

Since this study employed a convenience sample, it should be acknowledged that
this is, in actuality, a non-probability study. The results, therefore, are limited in that they

truly cannot be generalized across the whole of the WWW shopping population.

43

 

 

 

 

Valid
nude

709-,

repor

Year.

aCCet

 

However, the study’s results and the conclusion drawn from these results retain
suggestive, if not directive, value to further develop online price search knowledge.
Participants

A convenience sample of students from a large Midwestern university was used
for both the pilot study and the main study. The pilot sample of 127 qualified students
contained slightly more females (54%) than males (46%). The average age of
participants was 20 years old, with 90% ranging in age fiom 20 to 23. Most participants
(71%) were in their second or third year of undergraduate education. The majority of
participants (62%) reported being employed, with 53% earning under $10,000 per year,
and 19% earning between $10,000 and $20,000. All participants had purchased at least
one item through the W in the three months preceding the survey administration,
with 26% having made their last purchase in the two weeks preceding the survey, 19% in
the 4 weeks preceding, and 19% in the 6 weeks preceding.

In the main study, the vast majority of participants came from classes in a College
of Communication, particularly a Department of Advertising. In total, 728 qualified
students completed the questionnaire. After data screening, 587 surveys were retained as
valid. Of the sample obtained from these 5 87 surveys, 72% of participants were
undergraduates and 18% were graduate students. The ratio of females to males was
70%/30%, and 20 was the average age of participants. A considerable proportion (65%)
reported being employed, with the vast majority (85%) earning less than $10,000 per
year.

Members of this sample overwhelmingly owned their own computer (97%),

accessed the WWW daily (93%), searched the WWW for information daily (61%), and

44

 

 

 

CO

pet
onr
the

the

prk
con
last
and
half
they

had

dew
modi
and (
QUESI
COns:

three

 

whkfi

 

conducted their WWW showing from their home computer (95%). The majority (50%)
also reported making at least one WWW purchase every three months. Spending in this
period is typically (72%) under $200 in total. While all participants had made at least
one purchase through the W in the three months preceding the survey to qualify for
the study, 70% had made their last purchase in the 8 weeks preceding, 44% had made
their last purchase in the 4 weeks preceding, and 23% had made their last purchase in the
2 weeks preceding. Almost half of the sample (49%) indicated that they had compared
prices online prior to their last WWW purchase. On average, members of this group
compared 4 prices before making their purchase. The majority of participants made their
last WWW purchase for themselves (65%). These purchases typically were for clothing
and accessories (26%), airline tickets (14%), books (12%), or event tickets (10%). While
half of participants (50%) reported having made their last WWW purchase fiom a seller
they had bought from before, the other half (50%) purchased from a seller whom they
had previously never used.
Measurement

This section discusses the operationalization of each construct and explicates the
development of the scale items used for their measurement. Since almost all items were
modified from preexisting scales for applicability to the context of WW shopping
and/or online price search, the entire research instrument was pilot tested. The pilot
questionnaire contained 90 items, 61 of which pertained to the measurement of the15
constructs in the proposed model. Semantic differential type items were used to measure
three constructs. The remaining 12 constructs were measured using statements about

which survey participant indicated their level of agreement on a 7 point Likert type scale.

45

 

 

 

 

SL‘

1111

SC:
alr
edr

cap
COg
lOCE
mee
to It
moc'

one

1
t

 

The pilot research instrument is presented in Appendix A. Results of the pilot study and
subsequent revisions to the research instrument are reported in the next chapter.

Two points should be noted about the development of these measures. First, all of
the constructs in the proposed model were considered latent unobservable variables. The
indicators, therefore, define the domain of the construct they measure. Second, whenever
possible, indicators were modified from preexisting scale items. The use of preexisting
scales was based on their conceptual link to the construct’s operational parameters. In
almost all cases, select items were used from preexisting scales. With the exception of
education, all constructs were measured using multiple indicators.

Abilig to Online Price Search

Ability to price search online is defined as an individual’s perceived cognitive
capability of searching for and processing price information. It pertains to individuals’
cognitive structures (i.e., skills) and is manifested in their confidence in their ability to
locate lower prices when shopping on the WWW. Ability to online price search was
measured using multiple items asking consumers to rate their confidence in their ability
to locate lower prices when shopping on the WWW. Items for this construct were
modified from Urbany, Bearden, Kaicker, and Smith-deBorrero (1997). In addition,

original items were created.

46

Table 3.1

Ability to Online Price Search

 

Pilot statement

 

Please describe your ability to search for lower prices when shopping on the world-wide-web:
1. Low vs. High
2. Inadequate vs. Superior
3. Weak vs. Strong
4. Incapable vs. Capable
5. Not confident in my ability vs. Confident in my ability

6. Uncomfortable with my ability vs. Comfortable with my ability

 

Benefits of Online Price Search

Benefits of online price search are a consumer’s perceived utility gain and
exchange self-confidence resulting from a search for a WWW seller with the lowest
price. Conceptualization of the benefits of online price search is not limited to monetary
return. However, the construct’s operationalization focuses exclusively on the financial
return expected from search, since consumers are unlikely to consciously acknowledge
price search as having benefits beyond potential monetary saving. Benefits of online
price search were measured using select items from the search benefits scale developed

by Srinivasan and Ratchford (1991).

47

Table 3.2

Benefits of Online Price Search

 

Pilot statement

 

1. By searching around on the world-wide-web for price information, 1 am certain of making the
best buy.

2. It doesn’t pay to shop around before buying on the world-wide-web. ( R )
3. Shopping around at more than one web site helps me to find the lowest price.
4. When shopping on the world-wide-web, there is too much to lose by being ignorant about prices.

5. By rushing into a purchase when shopping on the world-wide-web, one is bound to miss a good
deal.

 

Costs of Online Price Search

Costs of online price search are a consumer’s subjective assessment of the
monetary expenditure, time sacrifice, physical effort, and psychological sacrifice that he
or she expends searching for price information on the WWW. It is a measure of the lose
in time and money a consumer expects to incur as a result of online price search and an
assessment of the sacrifice in psychological energy and physical effort he or she expects
to exert in the process. The items used to measure perceived costs of online price search
focus on the perception of time and effort sacrificed to price search online. With the
exception of one indicator, adapted from Donthu and Gilliland (1996), all items are

original.

48

Table 3.3

Costs of Online Price Search

 

Pilot statement

 

1. When shopping on the world-wide-web, it feels like a chore to search for lower prices.

2. When shopping on the world-wide-web, I feel it is a large sacrifice of time to search for lower
prices.

3. When shopping on the world-wide-web, I usually don’t have the patience to search for lower
prices.

4. When shopping on the world-wide-web, it takes too much effort to find lower prices.

5. When shopping on the world-wide-web, 1 hate spending time to gather information on products.

 

Ease of Access to Online Information

Ease of access to online information is the extent to which a consumer seeking
information on the WWW perceives it to be available with minimal delay. The focus of
this construct is on a consumer’s attitude toward the typical amount of time it takes for
WW pages to load on the computer he or she usually uses to shop on the WWW. Ease

of access to online information was measured using an original scale.

49

Table 3.4

Ease of Access to Online Information

 

Pilot statement

 

Please think about the location from where you do most of your shopping
on the world-wide-web when answering the following. The speed at which web
pages load from this location is:

l. Intolerable vs. Tolerable
2. Unsatisfactory vs. Satisfactory

3. Slow vs. Fast

 

Education

Education refers to the level of formal higher education (e.g., undergraduate or
graduate) an individual has completed. Since this study uses college level students as
subjects, this measure asked participants their level in years (e. g., 14 for sophomore or 15

for junior).

Table 3.5

Education

 

Pilot statement

 

1. Please indicate the number of years of formal education you have completed.
13 : l4 : 15 : l6 : l7 : 18 : 219
Undergraduate Graduate

 

50

Internet Involvement

Internet involvement is an unobservable state of motivation of a person regarding
the WW and is his or her perceived relevance related to the Internet based on inherent
needs, values, interests, goals and objectives. It is an assessment of an individual’s use of
the W in relation to his or her experience with its task aiding and goal fulfillment
ability. Salam (1998), who originally developed this construct, based his scale of intemet
involvement on the personal involvement inventory (PII) scale developed by
Zaichkowsky (1985). This study modified select items from the abbreviated PII scale

developed by Zaichkowsky (1994).

Table 3.6

Internet Involvement

 

Pilot statement

 

To me the world-wide-web is:
l. Unimportant vs. Important
2. Mean nothing to me vs. Means a lot to me
3. Irrelevant vs. Relevant
4. Worthless vs. Valuable

5. Not needed vs. Needed

 

Motivation to Online Price Search
Motivation to online price search is as an individual’s perceived desire to exert
effort gathering and processing price information in an electronic marketplace, including

both the direction and the intensity of the effort. It is characterized by an individual’s

51

goal orientation, which in this case is locating a desired product at a lower price using the
WWW. Therefore it is viewed as a measure of an individual’s perceived motivation to
search the W, with location of a lower price being the principal goal. Perceived
motivation to online price search was measured using original items and select items

adapted from the Holbrook (1986) scale of motivation to engage in activities.

Table 3.7

Motivation to Online Search

 

Pilot statement

 

1. It is important for me to get the best price when shopping on the world-wide-web.

2. It really doesn’t occur to me to search for lower prices when shopping on the world-wide-web.

( R)
3. 1 am motivated to search for lower prices when shopping on the world-wide-web.
4. I am enthusiastic about searching for lower prices when shopping on the world-wide-web.
5. I feel driven to find the best price, when shopping on the world-wide-web.

6. I really want to find the best prices, when shopping on the world-wide-web.

 

Need for Cognition

Need for Cognition is the tendency of an individual to engage in and enjoy
thinking. This measure focuses on an individual’s preference toward general thinking.
Therefore, need for cognition (NFC) was measured using items from the abridged NFC

scale developed by Cacioppo, Petty, and Kao (1984) that deal with thinking in general.

52

Table 3.8

Need for Cognition

 

Pilot statement

 

1. The notion of thinking abstractly is appealing to me.

2. Thinking is not my idea of fun. ( R )

3. I prefer my life to be filled with puzzles that I must solve.
4. I only think as hard as I have to. ( R )

5. Learning new ways to think doesn’t excite me too much. ( R )

 

Online Price Search

Consumer external price search behavior in an electronic marketplace is the
degree of effort a buyer exerts to acquire and compare the price offerings of sellers in a
computer mediated shopping environment. This construct was measured using the
number of online prices compared prior to last WWW purchase, recalled estimates of
typical WWW price search behavior, and overall attitude toward online price search.
Typical WWW price search behavior and attitude toward WWW price search in general
were measured using select items from scales of comparison shopping (Hawes &
Lumpkin, 1984; J. E. Urbany et al., 1996) and price consciousness (Lichtenstein,

Ridgway, & Netemeyer, 1993).

53

Table 3.9

Online Price Search

 

Pilot statement

 

1. How many prices did you compare on the web before making your last purchase on the world-
wide-web?

2. When shopping on the world-wide-web, “Never buy the first one you look at” is a good motto.
3. When shopping on the world-wide-web, 1 am willing to go to extra effort to find lower prices.

4. I make it a rule to visit more than one web retailer or use a shop-bot to compare prices before I
buy on the world-wide-web.

5. I would never shop at more than one web site just to find low prices before I buy on the world-
wide-web. ( R )

6. When shopping on the world-wide-web, 1 often compare the prices at more than one web retailer
before deciding where to buy from.

 

Perceived Budget Constrfla

Perceived budget constraints are a consumer’s assessment of his or her general
financial state. It refers to a consumer’s perception of having sufficient monetary funds
to satisfy his or her needs or wants. The budget constraints construct was measured using

the budget constraints scale of Urbany et al. (1996).

Table 3.10

Perceived Budget Constraints

 

Pilot statement

 

l. I frequently have problems staying within my budget.
2. My budget is always tight.

3. 1 often have to spend more money than 1 have available.

 

54

Perceived Price Dispersion

Perceived price dispersion is a consumer’s assessment of the variance of prices in
an electronic marketplace (i.e., the WWW). It is a measure of a consumer’s belief that
W sellers’ prices for similar products vary considerably. Perceived price dispersion

was measured using select items from the price dispersion scale of Urbany et al. (1996).

Table 3.11

Perceived Price Dispersion

 

Pilot statement

 

1. Some web sites have prices that are much lower than others.

2. Prices of individual items may vary between web sites, but overall, there isn’t much difference in
the prices between web sites. ( R )

3. The price of individual items oflen varies a lot between web sites.

 

Perceived Time Pressure

Time pressure refers to a consumer’s perception of the amount of the time he or
she has available to make WWW purchasing decisions. It measures a consumer’s belief
that his or her shopping on the WWW generally is rushed due to a lack of time. Time

pressure was measured using the time pressure scale developed by Putrevu and Ratchford

(1997).

55

Table 3.12

Perceived Time Pressure

 

Pilot statement

 

1. When I shop on the world-wide-web, 1 find myself pressed for time.

2. When 1 shop on the world-wide-web, l have more than enough time to complete my shopping.

(R)

3. When 1 shop on the world-wide-web, I finish my shopping fast because 1 have other things to do.

 

Subiective Knowledge of Internet Search

Subjective knowledge of intemet search is an individual’s assessment of his or her
proficiency at utilizing the W to perform information acquisition tasks. It is a
measure of an individual’s confidence in his or her ability to locate information using the
WWW. Subjective knowledge of intemet search was measured using select items

adapted from the Flynn and Goldsmith (1999) scale of subjective knowledge.

Table 3.13

Subjective Knowledge of Internet Search

k

Pilot statement

1. I feel very knowledgeable about searching for information on the world-wide-web.

1 Among my circle of friends, I’m one of the “experts” on searching for information on the world-
wide-web.

3- 1 know how to search for information on the world-wide-web.

P

I know more about searching for information on the world-wide—web than most people.

X

56

Trust in W Sellers

Trust in W sellers is the buyer’s expectation that the seller can be relied upon
to fulfill his or her part of the exchange and that the seller will not exploit the buyer’s
vulnerability. Trust is regarded as a risk reliever. Therefore, this is a measure of the
degree of a consumer’s desire to avoid assuming unnecessary risk in conducting a
purchase through the WWW. Trust in W sellers was measured using items

concerning risk aversion from Donthu and Gilliland (1996) and Moorthy et a1. (1997).

Table 3.14

Trust in W Sellers

 

Pilot statement

 

1. I would buy from a web site 1 never heard of before.

2. I am not concerned about a potential financial loss when buying from a web retailer I never heard
of before,.

3. I don’t consider it risky to buy from a web site I have never heard of before.

 

WWW Shopping Enthusiasm

WWW shopping enthusiasm is the enjoyment an individual feels for the task of
collecting and processing online information about products. Shopping enthusiasts are
characterized as consumers motivated by hedonic values, like escape or increased arousal
(Babin et al., 1994), and measure of this construct focus on these values. WWW
shopping was measured using adapted items from the hedonic portion of the Babin et a1.

(1994) scale assessing hedonic and utilitarian shopping values.

57

Table 3.15

WWW Shopping Enthusiasm

 

Pilot statement

 

1. Shopping on the world-wide-web feels like an escape.

2. Compared to other things 1 could do, the time spent shopping on the world-wide-web is truly
enjoyable.

3. I enjoy shopping on the world-wide-web for its own sake, not just for the items I may purchase.

 

Data Analysis Strategy

The pilot study was used to assess the validity and reliability of the indicators and
scales used to represent the constructs of interest. Confirmatory factor analysis, using the
LISREL 8.5 statistical software package, was used to determine the validity of the scale
items used to measure each construct. Scale reliability was assessed using Cronbach
Alpha reliability coefficient and principal components analysis figures produced in the
SPSS 11.0 statistical software package. Discriminaant validity was assessed through an
examination of the correlations between latent variables. Scale items that compromised
the reliability of their respective scales were noted for possible elimination. Participants
also were asked to comment on the questionnaire. These comments were used to refine
the design of the research instrument to ensure instructions were clear, questions were
understandable, and participants had a general ease with the survey. Since the pilot study
employed a student sample from the same university as the sample used for the main
study testing the full causal model, it was expected that there was a fair degree of external
validity related to the pilot and the main study’s sample population. Results of the pilot

study are reported in the next chapter.

58

The two-step approach to model building was employed for the main study
(Anderson & Gerbing, 1988; James, Mulaik, & Brett, 1982). This process emphasizes
the analysis of the measurement model before examination of the structural model. In
this way the fit of the observed variables to the latent variables is assessed independently
and prior to estimating the fit of the latent variables on themselves. Pioneers in the
development of structural equation modeling and creators of the LISREL statistical
software program, Joreskog and Sorbom (1993, p. 113) describe the value of this two-
step approach in stating:

The testing of the structural model, i.e., the testing of the initially specified

theory, may be meaningless unless it is first established that the

measurement model holds. If the chosen indicators for a construct do not
measure that construct, the specified theory must be modified before it can

be tested. Therefore, the measurement model should be tested before the

structural relationships are tested.

To analyze the measurement model, as in the pilot study, scale reliability was assessed
using Cronbach Alpha reliability coefficient and principal components analysis, and
discriminant validity was assessed through an examination of the correlations between
latent variables. Maximum Likelihood estimation in the LISREL 8.5 statistical program
was used to fit the observed covariance matrix against the predicted covariance matrix in

analysis of the measurement model, structural model, and overall fit indices.

59

Chapter 4

PILOT STUDY RESULTS

This chapter discusses the results of the pilot study used to establish the validity
of the items and the reliability of the scales measuring the latent variables in the proposed
model. It also presents revisions to the research instrument that were the product of this
preliminary investigation.

The ranges of all items were checked for improper input and univariate outliers.
Rare instances of missing data were addressed by mean substitution. The results of tests
for measurement validity and reliability are summarized in Tables 4.1 through 4.14.

Each table contains (1) the statements for the pilot tested scales, (2) the single factor
loadings of each item, (3) the total correlation for each item, and (4) the Cronbach Alpha
reliability coefficient for each scale. The symbol ( R) following any statement denotes a
reverse worded scale item. Table 4.15 is a correlation matrix of the latent variable used
to determine discriminant validity.

Of the 61 items used to measure the 15 latent variables in the proposed model,
five items had unacceptably poor single item factor loadings (< .50) and unsatisfactory
item-total correlations (< .50). These five indicators were: item 5 of the online price
search scale, items 1 and 3 of the need for cognition scale, item 2 of the benefits of online
price search scale, and item 2 of the perceived price dispersion scale. The Cronbach

Alpha reliability values of all scales were above the conventionally recommended 0.6.

60

Most scales appeared to be unidimensional, measuring only one construct. To cross
validate this conclusion, a principal components analysis was conducted on each scale.
With the exception of the need for cognition scale, a single component was extracted
from each scale, confirming the unidimensional of all other scales.

A pattern was found between the five low loading items. Three of the five items
were reverse worded statements. The other two items were part of a five-item scale
measuring need for cognition, of which the other three items were reverse worded. It was
then perhaps not surprising that, when the need for cognition scale was subjected to a
principal components analysis, these two items loaded onto one factor, while the other
three (reverse worded) items loaded onto another factor.

An examination of the latent variable correlation matrix to determine discriminant
validity, presented in Table 6.15, found that, while the signs of the hypothesized
relationships between constructs were in predicted directions, the correlations between a
number of latent variables were high (r >.40).

Given the strong theoretical association between some of theses constructs, a
certain degree of correlation may be expected. It is reasonable to suppose, for example,
that involvement with, knowledge of, and a facility with the W are interrelated
concepts that explain the correlations between ease of access to online information and
intemet involvement (r=.40), intemet involvement and subjective knowledge of intemet
search (r=.52), and subjective knowledge of intemet search and WW shopping
enthusiasm (r=.48). Likewise, subjective knowledge of intemet search and ability to
online price search (r=.55) are both constructs related to aspects of online search.

Perceived price dispersion and benefits of online price search (r=.65) both tap into

61

individuals’ perception of potential savings that can be realized in an electronic
marketplace. Both perceived time pressure and costs of online price search (r=.60) deal
with time as a resource, the lack of time in the former construct and the value of time
being a large part of the latter construct. Irrespective of these strong theoretical links, the
existence of these high correlations is less than optimal.

Because five of the 15 constructs in the proposed model act as mediating
variables and deal with theoretically different aspects of price search (i.e., ability,
motivation, costs, benefits, and price search behavior), particular attention was focused
on the relationships between these constructs. While strong correlations existed between
all of these constructs, exceptionally high correlations were found between online price
search and motivation to online price search (r=.84), between online price search and
benefits of online price search (F85), and between benefits of online price search and
motivation to online price search (r=.95). These exceptionally high correlations were
expected to have considerable affect on the estimation of the structural model, since they
indicated extremely strong linear dependency between benefits, motivation and price
search.

It also should be noted that prior to administering the pilot survey there was
concern that the questionnaire’s length would affect responses because considerable time
would be required to complete the survey. However, this did not appear to be an issue.
Students completed the lengthy questionnaire in approximately ten minutes. The speed
with which students completed the questionnaire shifted concern away the length of the
survey, and placed attention on the amount of attention participants may have given to

each question and the effects this treatment had on the results of the of the pilot study.

62

For example, the low factor loadings in the tests of indicator validity were all
related to reverse worded statements. This pattern suggests that in general participants
did not recognize the reverse wording of items. The speed with which the surveys were
completed could explain these results. Thus, the five items were retained, despite their
low factor loadings.

An examination of between item-correlation matrices for each scale showed high
inter-item correlations. A visual check of surveys revealed a consistent pattern of near
identical responses between related items. This pattern suggests that the scale items are
either highly redundant and/or that participants’ responses were biased through a priming
effect coupled with a generally quick treatment of the surveys. Given multiple
participants’ written comments regarding repetitive questions and the speed with which
participants generally completed the survey, it is suspected that this pattern in responses
is the product of both influences to a certain degree, and that each acted to confound the
other. Since no means are available for partitioning of these effects, it was concluded
that both should be accepted as plausible explanations for the results.

Likewise, it was difficult to conclude definitively if the high correlations among
the benefits, motivation and price search constructs were the product of gross redundancy
between their respective indicators or the result of respondents not giving adequate
consideration to the differences between scale items. These results also might suggest
that the differences in the five price search constructs, which appear appropriate on a
theoretical level, may, in actuality, have relationships of extreme subtle nuance. If that is
the case, a measurement method requiring individuals to consciously differentiate

between these five aspects (i.e., survey) may be inappropriate. An examination

63

comparing and contrasting the indicators for these constructs lead to the conclusion that
they appear, on the surface, sufficiently different and that particular attention should be
focused on the relationship between the price search constructs in the main study’s
analysis of the measurement model.
Research Instrument Revisions

In the revised version of the research instrument, used for the main study, the
items measuring the latent variables were completely randomized, since it was believed
that participants’ generally brief treatment of the pilot questionnaire influenced the results
of the pilot study. This means no scale item was preceded or followed by a question
related to the same construct, and all efforts were made to separate questions from related
constructs. In this manner, participants were encouraged to exert more effortful cognitive
consideration of the questionnaire. One should note that because six of the 15 constructs
dealt with some aspect of online search, five having to do with facets of online price
search specifically, a considerable proportion of the scale items were by their nature
search oriented. These items, therefore, were almost always separated by only one non-
search statement. To ensure all construct indicators were included in this randomization
process, new scales were constructed for the three constructs measured with semantic
differential type items (i.e., ease of access to online information, ability to online price
search, and intemet involvement).

To address redundancy issues, and thereby, it was hoped, deal with extreme
construct correlations, a review of item wording was initiated, and refinements were
made where appropriate. Where possible, scales were overestimated to ensure that they

contained a sufficient number of items to withstand a process of item elimination, while

64

 

 

 

 

ll

.1,

1’2

1,

.w

 

maintaining their ability to reliably estimate their respective construct. Therefore, in a
number of cases, items were added to scales. Appendix B presents tables with construct
scales used in the pilot and revisions that resulted from this modification process.
Appendix C presents the revised research instrument used in the main study, including a
key noting the construct relationship of each item. The following chapter discusses

results of the main study to estimate and assess the full model.

Table 4.1

Ability to Online Price Search

 

Statement Factor Item-total Cronbach
loadings Correlation Alpha

 

Please describe your ability to search for lower
prices when shopping on the world-wide-web:

1. Low vs. High .83 .76
2. Inadequate vs. Superior .92 .86
3. Weak vs. Strong .93 .88
4. Incapable vs. Capable .84 .84
5. Not confident in my ability vs. Confident in my .83 .84

ability
6. Uncomfortable with my ability vs. Comfortable .79 .80

with my ability

Six items .94

 

65

Table 4.2

Benefits of Online Price Search

 

 

 

 

 

 

Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. By searching around on the world-wide-web for .54 .48
price information, I am certain of making the best
buy.
2. It doesn’t pay to shop around before buying on the .32 .28
world-wide-web. (R)
3. Shopping around at more than one web site helps .58 .52
me to find the lowest price.
4. When shopping on the world-wide-web, there is too .69 .53
much to lose by being ignorant about prices.
5. By rushing into a purchase when shopping on the .78 .60
world-wide-web, one is bound to miss a good deal.
Five items .79
Table 4.3
Costs of Online Price Search
Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. When shopping on the world-wide-web, it feels like .80 .75
a chore to search for lower prices.
2. When shopping on the world-wide-web, I feel it is a .76 .72
large sacrifice of time to search for lower prices.
3. When shopping on the world-wide-web, I usually .83 .78
don’t have the patience to search for lower prices.
4. When shopping on the world-wide-web, it takes too .90 .85
much effort to find lower prices.
5. When shopping on the world-wide-web, 1 hate .72 .65
spending time to gather information on products.
Five items .90

 

66

Table 4.4

Ease of Access to Online Information

 

Statement Factor Item-total Cronbach
loadings Correlation Alpha

 

The speed at which web pages load from this

 

 

 

 

location is:
1. Intolerable vs. Tolerable .86 .82
2. Unsatisfactory vs. Satisfactory .92 .86
3. Slow vs. Fast .90 .85
Three items .92
Table 4.5
lntemet Involvement
Statement Factor Item-total Cronbach
loadings Correlation Alpha
To me the world-wide-web is:
l. Unimportant vs. Important .79 .65
2. Means nothing to me vs. Means a lot to me .70 .53
3. Irrelevant vs. Relevant .77 .62
4. Worthless vs. Valuable .54 .57
5. Not needed vs. Needed .55 .56
Five items .85

 

67

 

 

 

 

t)

')J

Table 4.6

Motivation to Online Price Search

 

 

 

 

 

 

Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. It is important for me to get the best price when .85 .84
shopping on the world-wide-web.
2. It really doesn’t occur to me to search for lower .58 .58
prices when shopping on the world-wide-web. (R)
3. I am motivated to search for lower prices when .86 .83
shopping on the world-wide-web.
4. I am enthusiastic about searching for lower prices .90 .83
when shopping on the world-wide—web.
5. I feel driven to find the best price, when shopping .95 .89
on the world-wide-web.
6. I really want to find the best prices, when .85 .83
shopping on the world-wide-web.
Six items .93
Table 4.7
Need for Cognition
Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. The notion of thinking abstractly is appealing to me. .47 .44
2. Thinking is not my idea of fun. (R) .70 .54
3. I prefer my life to be filled with puzzles that I must .31 .31
solve.
4. I only think as hard as l have to. (R) .60 .45
5. Learning new ways to think doesn’t excite me too .75 .57
much. (R)
Five items .70

 

68

Table 4.8

Online Price Search

 

 

 

 

 

 

Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. How many prices did you compare on the web 0.51 .49
before making your last purchase on the world-
wide-web?
2. When shopping on the world-wide-web, “Never 0.74 .61
buy the first one you look at” is a good motto.
3. When shopping on the world-wide-web, I am 0.70 .60
willing to go to extra effort to find lower prices.
4. I make it a rule to visit more than one web retailer 0.83 .60
or use a shop-bot to compare prices before I buy on
the world-wide-web.
5. I would never shop at more than one web site just to 0.27 .23
find low prices before I buy on the world-wide-web.
(R)
6. When shopping on the world-wide-web, I often 0.87 .79
compare the prices at more than one web retailer
before deciding where to buy from.
Six items .81
Table 4.9
Perceived Budget Constraints
Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. I frequently have problems staying within my .76 .62 ‘
budget.
2. My budget is always tight. .47 .42
3. l oflen have to spend more money than I have .90 .67
available.
Three items .74

 

69

Table 4.10

Perceived Price Dispersion

 

 

 

 

 

 

Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. Some web sites have prices that are much lower .82 .65
than others.
2. Prices of individual items may vary between web .35 .30
sites, but overall, there isn’t much difference in the
prices between web sites. (R)
3. The price of individual items often varies a lot .75 .45
between web sites.
Three items .65
Table 4.1 1
Perceived Time Pressure
Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. When I shop on the world-wide-web, I find myself .68 .51
pressed for time.
2. When I shop on the world-wide-web, l have more .77 .55
than enough time to complete my shopping. (R)
3. When I shop on the world-wide—web, I finish my .48 .40
shopping fast because I have other things to do.
Three items .67

 

70

Table 4.12

Subiective Knowledge of lntemet Search

 

 

 

 

 

 

Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. I feel very knowledgeable about searching for .81 .71
information on the world-wide-web,
2. Among my circle of friends, I’m one of the .76 .72
“experts” on searching for information on the
world-wide—web.
3. I know how to search for information on the world- .81 .72
wide-web.
4. I know more about searching for information on the .76 .70
world-wide-web than most people.
Four items .87
Table 4.13
Trust in WWW Sellers
Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. I would buy from a web site I never heard of before. .52 .46
2. I am not concerned about a potential financial loss .60 .51
when buying from a web retailer 1 never heard of
before.
3. I don’t consider it risky to buy from a web site 1 .95 .66
have never heard of before.
Three items .72

 

71

Table 4.14

 

 

 

WWW Shopping Enthusiasm
Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. Shopping on the world-wide-web feels like an .82 .76
escape.
2. Compared to other things I could do, the time spent .89 .82
shopping on the world-wide-web is truly enjoyable.
3. I enjoy shopping on the world-wide-web for its own .87 .80
sake, not just for the items I may purchase.
Three items .89

 

72

 

 

 

dd._ vod wvd 2d- Ed mad odd 2d 3d and Sd ddd- 3d end 505055”,— mefinosm £03 3
do; 2d mmd- Sd- de mod 3 d- _dd Ed 2 d «Nd- odd and €0=0m 33>? 5 SEE Q
dd; 3.? dmd 3d vmd wmd 9d and 3d dad- de mmd :080m do 0300—329 058035 E
do; wdd- vod wmd- mmd- mmd- mmd- dmd- odd dmd- dmd- 0.50005 05,—. 32085 I
do; vod- 3d nod- 9.0 2d 2d Rd- mod mmd :55035 05.5 35083 2

do." Ed. 2d. mdd- cod- 8d dmd 2d 3d $505280 flowcsm “0050005 a

do; odd- «wd 2d mod cod- nwd dvd noaom 05.5 0550 w

do; 5.? M: .d mod- 5.? mod. odd :oEcmoU 8.5 @002 N.

do.“ 2 .o :6 Nod- 36 mmd zoumombotm 0550 8 558352 d

cog ovd Ed- Ed dvd 5050203: “050:: m

do; 2.? m _ .d mvd 5305085 0550 9 m0000< do 000m v

2: mod- med- 580% 0055 0550 do 980 m

dd._ mmd :050m 055 0550 do 5.35m m

2: goaom SE 8:5 2 be? 2

E m _ m. I d. d w n d w v m N _ 035.2; 520..—

 

 

26 030,—.

73

Chapter 5

MAIN STUDY RESULTS

Introduction

This chapter first discusses the screening of the data. It then presents an
assessment of the measurement model. Results of this analysis indicated the necessity to
reconsider the proposed model’s measurement and specification before proceeding to its
full estimation. After model revisions, estimation procedures are outlined. Following
this discussion, assessment is made of the overall model fit and structural model.
Directed modifications to the model and corresponding improvements achieved are then
presented.

Data Screening

In total, 728 surveys were gathered from qualified participants for the main study.
The response values of all reversed worded items were reversed to facilitate
interpretation. The ranges of all variable values were examined for inaccurate data entry,
allowing for the location and correction of the few instances where this error existed.
Rare instances of missing data were dealt with through mean substitution.

In an analysis for univariate normality, the skewness and kurtosis of each latent
variable were translated into z-scores, with significant values indicating distribution
issues. This procedure, however, is sensitive to large sample sizes (e.g., > 500). A visual

inspection for skewness and kurtosis was then initiated using histograms and distributions

74

curves. There were clear indications that a number of variables required transformation.
Transformations were performed where needed. The actual size of variables’ skewness
and kurtosis afier transformation, as close to zero as possible, acted as the arbiter of
normal distribution.

The transformation process appeared to correct instances of univariate outliers.
Multivariate outliers were detected using Mahalanobis distance. In total, 86 cases
containing multivariate outliers were discovered through this process. These cases were
eliminated from the analysis because of regression’s sensitivity to outliers, and the
expected effects these outliers would bring to the estimation of the sample covariance
matrix. Because multivariate outliers tend to mask other multivariate outliers, this
procedure was repeated three more times: 33, 16, and six outliers were discovered in the
second, third, and fourth procedures respectively. These outliers also were removed from
the final data set. Statistical power was not compromised as a result of these procedures,
since 587 cases remained for full estimation of the proposed model (MacCallum et al.,
1996).

Measurement Model Assessment

Assessment of the measurement part of the model is principally concerned with
the relationships between the latent variables and their indicators to ensure valid and
reliable measurement. Results of tests for indicator and scale validity and reliability are
summarized in Tables 5.1 through 5.14. Each table contains (1) the item statement, (2)
the single factor loadings of each item, (3) the total correlation for each item, and (4) the

Cronbach Alpha reliability coefficient for each scale. The symbol ( R) following any

75

statement denotes a reverse worded item. Table 5.15 shows the correlation matrix of the
latent variables used to assess discriminant validity.

The single factor loadings for each item and their respective error variances were
all significant as indicated by t-values in excess of 1.96 in absolute terms. The Cronbach
Alpha reliability values of all scales were above the recommended 0.6. Ten items,
however, had generally weak single factor loadings (< .60) and unsatisfactory item-total
correlations (< .50). With the exception of item one of the perceived price dispersion
scale, these items were eliminated from their respective scales. The single exception was
made because retaining item number one of the perceived price dispersion scale (K = .58)
allowed for the scale to remain just identified without compromising its reliability.
Unlike the pilot, no discemable pattern was found between items with low factor
loadings.

Principal components analyses found all scales, except those of need for cognition
and perceived budget constraints, to be unidimensional. The removal of items discussed
above corrected for these issues. It should be noted that as a result of this item
elimination process the perceived budget constraints scale was reduced to a single item
measure. Perceived budget constraints’ item number one was selected as its single
indicator, since its wording most accurately captured the construct and it contained
considerable variance.

An examination of the latent variable correlation matrix, Table 5.15, found that,
with one exception, the signs of all hypothesized relationships were in predicted

directions. Although a positive association between perceived budget constraints and

76

benefits of online price search had been hypothesized, the results indicated a negative
relationship between these two constructs.

Addressing discriminant validity, the correlations between a number of latent
variables were found to be high (r > .40). As discussed in the results of the pilot study,
some strong associations were expected given the close theoretical relationship between
certain constructs. For example, 97% of survey participants reported owning a computer,
93% said that they accessed the WWW daily, and 61% noted that they searched the
W for information daily. It then was of little surprise to find strong correlations
between intemet involvement and subjective knowledge of intemet search, since both
constructs deal with individuals living a wired lifestyle. Given this study’s focus on
consumer search for the lowest price, it was likewise not surprising to see the strong
association between the participants’ perception of price dispersion on the WW and
their sense of the benefits and motivations to engage in online price search behavior.

Because the five endogenous latent variables in the proposed model deal
specifically with theoretically different aspects of price search (i.e., ability, motivation,
costs, benefits, and price search behavior), particular attention was focused on the
relationships between these constructs. Correlations between the endogenous constructs
were found to be considerably high (r > .60). However, unacceptably high correlations
were found between online price search and motivation to online price search (r = .99),
between online price search and benefits of online price search (r = .93), and between
benefits of online price search and motivation to online price search (r = .93). These
relationships clearly indicated multicollinearity between the scales and implied that the

constructs are so closely linked that participants did not discern their differences. Given

77

their extreme values, these correlations were expected to have considerable affect on the

estimation of the structural model to the point of precluding the generation of reliable

parameter estimates. The model, therefore, was revised to address this issue through a

combination of scale adjustment and model respecification as discussed in the following

section.

Table 5.1

Ability to Online Price Search

 

 

Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. When it comes to searching for lower prices on the .789 .754
world-wide-web, I am an expert.
2. Searching for lower prices on the world-wide-web .763 .730
is something I can do with great ease.
3. When it comes to searching for lower prices on the .797 .757
world-wide-web, 1 am better at it than most people.
4. When I shop on the world-wide-web, I have .834 .792
superior techniques for locating lower prices.
5. I am extremely effective at locating lower prices .862 .816
when I search for them on the world-wide-web.
6. I am very confident in my ability to search for .812 .769
lower prices on the world-wide-web.
Six items .919

 

78

Table 5.2

Benefits of Online Price Search

 

 

Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. When shopping on the world-wide-web, I am .742 .664
certain of making the best buy by searching around
for price information.
2. Shopping around at more than one web site helps .759 .686
me to find the lowest price.
3. Before buying on the world-wide-web, a person can .746 .666
save a lot of money if they compare prices from
different websites.
4. When shopping on the world-wide-web, there is too .588 .550
much to lose by being ignorant about prices.
5. It pays to shop around before buying on the world- .783 .686
wide—web.
6. When shopping on the world-wide-web, one is .632 .584
bound to miss a good deal by rushing into a
purchase.
7. Before buying on the world-wide-web, there are .876 .766
definite benefits to comparing the prices at different
websites.
Seven items .889

 

79

Table 5.3

Costs of Online Price Search

 

 

 

 

 

 

Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. When shopping on the world—wide—web, it feels like .610 .575
a chore to search for lower prices.
2. When shopping on the world-wide-web, I feel it is a .765 .691
large sacrifice of time to search for lower prices.
3. When shopping on the world-wide-web, 1 usually .832 .635
don’t have the patience to search for lower prices.
4. When shopping on the world-wide-web, it takes too .831 .648
much effort to find lower prices.
5. When shopping on the world-wide-web, 1 hate .601 .509
spending time to gather information on products.
Five items .849
Table 5.4
Ease of Access to Online Information
Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. The speed at which web pages load on the computer .806 .741
where I do most of my web shopping is tolerable.
2. I am completely satisfied with the speed at which .878 .787
web pages load on the computer where I do most of
my web shopping.
3. Web pages load quickly on the computer where 1 do .833 .760
most of my web shopping.
Three items .876

 

80

Table 5.5

lntemet Involvement

 

 

Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. The world-wide-web plays an extremely important .718 .612
role in my life.
2. It means a lot to me to have daily access to the .884 .753
world-wide-web.
3. The world-wide-web is a very valuable tool in my .843 .694
life.
4. The world-wide-web is a relevant part of my daily .726 .567
life.
5. I need the world-wide-web. .805 .732
6. The day isn’t complete unless I have used the .624 .690
world-wide—web.
7. Daily access to the world-wide-web is something I .719 .671
could not live without.
8. I feel like I use the world-wide-web for everything. .648 .623
Eight items .910

 

81

Table 5.6

Motivation to Online Price Search

 

 

Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. When I shop on the world-wide-web, it is important .772 .735
for me to get the best price.
2. When I shop on the world-wide-web, it really .753 .712
doesn’t occur to me to search for lower prices. ( R )
3. When I shop on the world-wide-web, I am .909 .862
motivated to search for lower prices.
4. When I shop on the world-wide-web, I am .778 .753
enthusiastic about searching for lower prices.
5. When I shop on the world-wide—web, I feel driven .887 .843
to find the best price.
6. When I shop on the world—wide-web, I really want .852 .811
to find the best prices.
Six items .927

 

82

Table 5.7

Need for Cognition

 

 

Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. The notion of thinking abstractly is appealing to me. .356 .341
2. Thinking is not my idea of fun. ( R ) .725 .546
3. I prefer my life to be filled with puzzles that 1 must .294 .324
solve.
4. I only think as hard as I have to. ( R ) .728 .490
5. Learning new ways to think doesn’t excite me too .754 .554
much. ( R)
6. I would rather do something that requires little .794 .615
thought than something that is sure to challenge my
thinking. ( R)
7. It’s enough for me that something gets the job done: .532 .396
I don’t care how or why it works. ( R )
Seven items .794

 

83

Table 5.8

 

 

Online Price Search
Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. How many prices did you compare on the web .660 .637
before making your last purchase on the world-
wide-web?
2. When shopping on the world-wide-web, “Never .620 .599
buy the first one you look at” is a good motto.
3. When shopping on the world—wide-web, I am .782 .750
willing to go to extra effort to find lower prices.
4. Before I buy on the world-wide-web, I make it a .883 .840
rule to compare the prices of different web sites.
5. When I shop on the world-wide-web, I often .885 .839
compare the prices between web site before
deciding whom to buy from.
6. I would not buy anything on the world-wide-web .883 .843
without first comparing the price at another web
site.
Six items .906

 

84

Table 5.9

Perceived Budget Constraints

 

 

 

 

 

 

Statement Factor Item-total Cronbach
loadings Correlation Alpha
I frequently have problems only spending what I am .376 .480
suppose to for the week.
My spending budget is usually tight. .539 .192
I ofien feel I have to spend more money than I have .393 .506
available.
I don’t usually feel like I have enough money to .939 .473
spend during the week.
Four items .652
Table 5.10
Perceived Price Dispersion
Statement Factor Item-total Cronbach
Loadings Correlation Alpha
No matter what I am shopping for on the world- .517 .458
wide-web, I believe that some web site is selling it
at a much lower price than others.
When shopping on the world-wide-web, one should .819 .625
expect the price of an item to really differ between
web sites.
For almost any item one can purchase on the world- .786 .610
wide-web, the price often varies a lot between web
sites.
Three items .743

 

85

Table 5.11

Perceived Time Pressure

 

 

 

 

 

 

Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. When I shop on the world-wide-web, I find myself .812 .723
pressed for time.
2. When I shop on the world-wide-web, I have more .541 .481
than enough time to complete my shopping. ( R )
3. When I shop on the world-wide-web, I finish my .522 .411
shopping fast because I have other things to do.
4. When I shop on the world-wide-web, I am in a .777 .652
hurry.
5. When I shop on the world-wide-web, I only have a .828 .646
limited amount of time to finish my shopping.
Five items .820
Table 5.12
Subiective Knowledge of lntemet Search
Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. I feel very knowledgeable about searching for .731 .622
information on the world-wide—web.
2. Among my circle of friends, I’m one of the .795 .715
“experts” on searching for information on the
world-wide-web.
3. Search for information on the world—wide-web is .806 .696
something I know how to do really well.
4. I know more about searching for information on the .834 .768
world-wide-web than most people.
Four items .870

 

86

Table 5.13

 

 

 

 

 

 

Trust in WWW Sellers
Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. If I have never heard of a web site, I will not buy .638 .647
from it. ( R)
2. When buying from a web site I have never heard of .718 .563
before, I am concerned about a potential financial
loss. ( R)
3. I consider it risky to buy from a web site I have .905 .760
never heard of before. ( R )
4. When buying on the world-wise—web, I need to .775 .734
have heard of the web site before I purchase
anything. ( R )
Four items .842
Table 5.14
WWW Shopping Enthusiasm
Statement Factor Item-total Cronbach
loadings Correlation Alpha
1. Shopping on the world-wide-web feels like an .719 .582
escape.
2. Compared to other things I could do, the time spent .753 .632
shopping on the world-wide-web is truly enjoyable.
3. I enjoy shopping on the world-wide-web for its own .628 .541
sake, not just for the items I may purchase.
4. While I shop on the world-wide-web, I am able to .686 .591
forget my problems.
Four items .789

 

87

 

 

 

00: 80- 00.0 8.0- .20 2.0 20 Q0- :0 2.0 00.0 000 0.00- 0.0 2.0 5302000 0200000 003 2
00: :00- 000. 00.0- 000- 0.0- 00.0 _00 00.0- 00.0 00.0 :0. 00.0 00 20000333205 3
00: :0- 20 00.0 .30 :0 20 :0 000 000 2.0. 2.0 00.0 02320 0002305: 308.3% 2
00: 8.0 20 20. £0. 20. ~00- 2.0 3.0- 000 :0- 30. 2032.: 220002800 2
00: 00.0 00.0 8.0- 00.0 :0 :0 _00 20. ~00 00.0 85000585 002800 :
00: :0- :0- 2.0- :0 :0 :0- S0 :0- 000. 02000000300000.2805 0:
2: 20 00.0 2.0 00.0 20 E0- 80 E0 00an 8F: 80000
2: :0 0.0.0 000 :0 2.0- 000 3.0 80200080 0820
do; de bod 2d end- Mdd mud £050m 05.5 0550 9 555382 N.
00.: 00. 20 0.0. R0 000 000202250835 0
00: 2.0- _00- _00 00.0 808000 0
dd._ dmd- _Nd mad sedan—.085 0550 8808.050 000m v
00.: 00.0- $0. fiaom 85 005500 280 m
dd.“ mud noaom 055 0550 .5 0:50:35 N
00: €an SE 2200 9 200.4. _
2 z 2 2 : 0. 0 0 a 0 m 0 m N _ 20233023

 

 

26 055.

88

Model Revision

Results of the analysis for discriminant validity indicated the necessity to
reconsider the proposed model’s measurement and specification before proceeding to its
full estimation. Revision of the proposed model focused on the relationships between
benefits, motivation, and price search, because it was the exceptionally high correlations
among these constructs that compromised reliable model estimation. Revision to the
proposed model consisted of both scale adjustment and model respecification.
§9§le Adjustment

The largest and most detrimental correlation was between the scales measuring
online price search, the principal dependant variable of the model, and motivation to
online price search (r = .99). Correlations between the individual items measuring these
constructs also were generally high (r > .50). Item number one and number two of the
online price search scale possessed the smallest correlations with items of the motivation
scale. Item one specifically asked participants about the number of prices they compared
online prior to their last WWW purchase. Item two asked about participants’ general
tendency toward online price search. While it was hoped that using a combination of
behavior and tendency items would generate a more robust measure of search, it was the
goal of the study ultimately to understand behavior. It was decided, therefore, to use only
item number one as the measure of online price search. The correlation between online
price search and motivation was dramatically reduced as a result of this procedure (r =
.61).

Price Search Behavior in an Electronic Marketplace was defined as the degree of

effort buyers exert to acquire and compare the price offerings of sellers in a computer

89

mediated shopping environment. This broad conceptualization of online price search,
however, no longer appeared appropriate, given the removal of tendency indicators from
its measurement. Price Search Behavior in an Electronic Marketplace, therefore, was
redefined as the degree of price comparison conducted by buyers prior to purchasing in a
computer mediated shopping environment. This more narrow conceptualization of online
price search, and its operationalization as the number of prices an individual compared
online immediately before making his or her last WWW purchase, was consistent with a
number of past studies that viewed external price search as the number of stores a buyer
visited to compare prices (Carlson & Gieseke, 1983; Kujala & Johnson, 1993; Urbany,
1986; Urbany et al., 2000). Additionally, this conceptualization maintained the study’s
core focus on the amount of price search behavior and it, in fact, addressed concerns that
the previous conceptualization and operationalization of the construct were not pure
behavioral assessments.

It should be noted that measuring online price search with a single indictor, while
acceptable, was restricting, in that it precluded the estimation of error. It required the
presumption that the measure was error free (i.e., the error variance was fixed at 0.0).
The reliability of this measure, therefore, was completely dependant upon participant’s
memory of price search prior to their last WWW purchase. There was a degree of
confidence in using this single indicator, since participants’ memories were primed in
preliminary questions asking them to recall the exact item they last purchased on the
WW and when it was purchased. Additionally, while all participants were qualified as

having made at least one purchase on the W in the three month preceding the survey,

90

70% made their last WWW purchase in the 8 weeks preceding the survey, 44% in the 4
weeks preceding, and 23% in the 2 weeks preceding.
Model Respecification

An exceptionally high correlation also was found between the benefits and
motivation scales (r = .93). Correlations between the individual items of these scales also
were generally high (r > .40). The conceptualization and operationalization of these
constructs were reexamined to discern the probable cause of these results.

Consistent with Ratchford and Srinivasan (1993), search in this study was viewed
as a process whose intended result was a better buy. Since this research focused on price
search, the better buy was necessarily one at a perceived lower price. The construct,
therefore, was operationalized as unidimensional to recognize finding the lowest/best
price as the principal benefit of price search. This was consistent with other search
research (Srinivasan, 1987; Srinivasan & Ratchford, 1991). As in the research of Park
and Mittal (1985) and Bettman (1979), goal orientation was the central component in this
study’s conceptualization of motivation to search. Given this study’s primary interest in
price search, this construct was operationalized to capture the intensity of an individual’s
motivation toward the particular goal fulfillment of finding the lowest/best price. Thus,
in comparing the items measuring these two constructs (c.f., Table 5.2 vs. Table 5 .6), one
will note that both scales viewed the location of the lowest/best price as the
desired/undesired outcome. These operationalizations appeared completely appropriate
when one considers this study’s specific focus on price search, as opposed to a more
general investigation of information search, which possesses benefits and motivations

that extend beyond a financial orientation.

91

Though not a relationship hypothesized a priori, it was an intriguing finding that,
when dealing with price search specifically, as opposed to information search in general,
the financial savings participants viewed as the benefit of price search was so strongly
related to their motivation to engage in price search. Certainly few would argue that in
general consumers are motivated to engage in behavior that does not yield benefits. In
the particular case of price search, where both reaping benefits and firlfilling goals (i.e.,
motivation) are rooted in the same product and predicated on achieving the same
outcome (i.e., finding the lowest price), it appeared redundant to view the benefits of and
motivation to conduct this type of search as separate constructs. The motivation
construct, therefore, was reconceptualized as inherently including benefits of online price
search as an implicit element of its composition.

The proposed model, shown in Figure 2.2, thus, was respecified to reflect this
new conceptualization of motivation to online price search. The respecified proposed
model is presented in Figure 5.1. One will note the absence of the benefits construct as it
now was considered part of the motivation construct. Accordingly, the determinant
variables associated with benefits (i.e., Perceived Price Dispersion, Perceived Budget
Constraints, and Trust in W Sellers) were redirected as influencers of motivation.

Model Estimation

Before assessment of the overall model fit and the structural model, the following
outlines the specifics of the model’s estimation. The revised causal model, presented in
Figure 5.1, was tested in its entirety. LISREL 8.5 statistical software package was used
for estimation. The covariance matrix of the observed variables was used as the input

matrix. Maximum Likelihood was used as the method of estimation.

92

    

      

 

83350.0 020095 333

 

.832on .8.“ @002 808029,:— 6885

                 
  

anyhow 03.5 0530 8
is. 3085

 

3580050 Howe—gm 3380a;

500.6905 ouch 33088

 

     

cough—om:—
oEEO 9 308.2 .«o 08m

     

05330

8598082 22585 noaom 8F— oEEO mo

flaw 33023

 

 

83.005 25,—. 320802“

   

nocaum BEBE
mo 0302305— 0500.35

   
     

seesaw out; 25:0 8
Mam—Ea 32083

  

33>? 82 8202 088805 :0 E Brazom :88m 025 :2:me 8:50:26 me 3902 3

026m
:0 202.0

 

93

The model’s input coding was free of any specifications intended to resolve
operational difficulties, and it adhered to the assumptions of the general LISREL model.
In terms of input specification, 18 observed dependent variables were used to measure the
four endogenous latent variables and 35 observed independent variables were used to
measure the ten exogenous latent variables.

The measurement model for the dependent variables (LAMBDA-Y) and
independents variables (LAMBDA-X) were specified as full and fixed matrices. The
matrices assessing the relationships between the endogenous latent variables (BETA) and
between the exogenous latent variables and the endogenous latent variables (GAMMA)
also were specified as full and fixed. The covariance matrix of endogenous latent
variables (PHI) was specified as diagonal and free. The matrices of the error variance for
the indicators of the endogenous latent variables (THETA EPSILON), for the indicators
of the exogenous latent variables (THETA DELTA), and for the disturbance terms of the
endogenous latent variables (PSI) were specified as diagonal. In this manner, these error
terms were estimated as mutually uncorrelated, which is minimally assumed for the
specification of a LISREL model.

For identification purposes one indicator of each latent variable acted as the
reference indicator (i.e., set to unity at 1.0). Additionally, the latent variables measured
with a single indicator (i.e., online price search behavior, perceived budget constraints,
and education) were assumed to be error free (error variance fixed to 0.0). The reliability
of the online price search measure was discussed in the preceding section. The reliability

of the education measure was thought to be very high. There were, however, no

94

indications related to the reliability of the perceived budget constraint measure. Any

assumptions about its reliability, therefore, are arbitrary.

Overall F it Assessment

The degree to which the model as a whole is consistent with the empirical data is

assessed in an examination of the model’s overall fit. Overall fit indices are summarized

in Table 5.16. A diagrammatic representation of the empirical estimation of the model is

presented in Figure 5.2.

Table 5.16

Model Overall Fit Indices

 

 

 

 

Fit index Value
Degrees of Freedom (df) 1269
Chi-square (x2) 3876.246
Root Mean Square Error of Approximation (RMSEA) 0.059
Normed Fit Index (N FI) 0.935
Non-Normed Fit Index (NNFI) 0.953
Comparative Fit Index (CPI) 0.957
Root Mean Square Residual (RMR) 0.101
Standardized Root Mean Square Residual (SRMR) 0.085

 

95

The chi-square for the model was high and significant, x2 (df== 1269, N = 587) =
3876.246, p = 0.0. Based on these values, the model would have been rejected. The chi—
square statistic, however, is known to be sensitive to large sample sizes, and its basic
assumption that a model can fit perfectly to a data set makes it an unrealistic test. Thus,
the probability result of this test statistic was not unexpected, and the model was retained.

The overall fit of a model, alternatively, was assessed using more pragmatic
goodness of fit statistics. One of the common tests used is the root mean square error of
approximation (RMSEA). Considered one of the most revealing metrics of overall model
fit, RMSEA takes the error of approximation in the population into consideration, and
tests the model against a model with optimally chosen parameters. Values less than 0.05
indicate a good model fit, while values between 0.05 and 0.08 indicate reasonable model
fit. The model had a RMSEA of 0.059. The model was determined to have reasonable
fit to the data according to this metric.

A number of test statistics compare the model against a naive base model, usually
referred to as the independence model, which assumes that each indicator is error free
and not correlated to any other indicator. The normed fit index (NFI), non-normed fit
index (NNFI), and comparative fit index belong to this group. Values in excess of .90
indicate an acceptable model fit to the data. The model possessed a NFI of 0.935, a
NNFI of 0.953, and a CFI of 0.957. These values were very encouraging indicators that

the model overall fit the data well.

96

          

 

Emflmsficm mafiaonm 33 3 :02:on 08 0002 0:080:02: 006005

                 

muozom 333 E «08,—.

nohom 00: 25:0 9

.dulld; 30>: 003080;

 

$505280 003:5 002023

02009.5 025 0020802”
«036+

 

    

520808:—
0530 2 00000< «a 000m

       
   

_ :5-

   

  

23330
002900.002 088805

:0. Elllqudmjlcgé 0m m 00. m

     
    
    

noaom 00cm 25:0 .00
flqqu 003089—

     
       

0Bmmflm 08:. 00>m0000m

mmmd-

  

:080m 00:00::
00 0200—305— 0300039m

 

n0§0m 02.5 0515 8
20% 8520

5:00st

mmod...

20.0 u 203 .000 08082000 20800090 3865 an: 205 6
00008300 28800th 03200930 0.8 0025/ CV

 

97

The root mean square residual (RMR) assesses the average residual value
obtained from the fitting of the sample covariance matrix to the matrix of the proposed
model. However, these residuals are relative to the sizes of the observed variances and
covariance, and therefore they are better represented when based on the correlation
matrix through the standardized root mean square residual (standardized RMR).
Optimally these values should be small (i.e., < .05). The model’s root mean square
residual (RMR) was 0.101, and its standardized RMR was 0.085. These values were less
than optimal. They were considered, however, within an acceptable distance of zero.
Thus, it was concluded that the model had a just reasonable fit to the data according to
these fit indices.

Given the values of these indices, the model as a whole was not rejected. These
figures, however, indicated that there were areas of the model within which less than
optimal specification existed. A detailed analysis of the results of the structural model
was used to bring these areas of discrepancy into sharper relief.

Structural Model Assessment and Hypothesis Testing

Assessment of the structural part of the model focuses on the direction and
strength of the links between the model’s latent variables, both exogenous and
endogenous. This analysis principally is concerned with the extent to which the
empirical data support or refute the theoretically specified (i.e., hypothesized)
relationships between the latent variables and the amount of variance accounted for in the
model’s structural components (i.e., endogenous variables). A diagrammatic
representation of the empirical results is presented in Figure 5.2. The results of the

structural model are shown in Table 5.17. The notation used for the equations in Table

98

5.17 is outlined in Table 5.18. Table 5.19 presents a summary of the hypothesized vs.
observed relationships among the latent variables. The squared multiple correlations for
the structural equations are presented in Table 5.20.

Seven of the 14 structural relationships hypothesized were supported. In addition,
two other hypothesized relationships were statistically significant, however, they were
not in the predicted direction. The seven hypothesized structural links with empirical
support were: search knowledge on ability, motivation on search, intemet involvement on
motivation, costs on motivation, search knowledge on costs, time pressure on costs, and
price dispersion on motivation. Both need for cognition and trust in W sellers had
been hypothesized to have a positive influence on motivation to online price search,
however, they were found to have negative affects.

Ability had no significant effect on online price search ([3 = 0.004). Hypothesis
one, therefore, was not supported. Similarly, Srinivasan (1987) in an early articulation of
his model of external information search for automobiles found that ability to search did
not have a significant direct link to search. He did find, however, that ability had an
i ndirect relationship to search through benefits. Though no indirect relationship between
ability and price search was specified in the proposed model, an examination of
modification indices indicates that a positive relationship may exist through the

motivation construct, within which benefits has been incorporated.

99

Table 5.17

Structural Model Results

 

Ability to Online Price Search

111 = +0023 62 +0.791 g.
(0.739) (18.485)*

Costs of Online Price Search

712 = 0002 2;, +0483 0 -O.258 g,
(-0.056) (10.521)* (-5.950)*

Motivation to Online Price Search

713 = -O.7ll'r]2 +0.137 E3 -O.105 E4 -0.034E5
(-19.09)* (4.129)* (-3.216)* (-1.197)

+0.419 E6 -0.072 E9 -0.052 E10
(10.997)* (-2.421)* (-1.279)

Online Price Search Behavior

“4 = +0.004n1 +0.621n3
(0.096) (15.905)*

* Significant (alpha = .05)
Note: The coefficients in the equations shown are standardized regression estimates and the
values in parenthesis are corresponding t-values.

 

100

Table 5.18

Latent Variable Notation

 

 

 

 

Latent Variable Notation
Endogenous
Ability to Online Price Search 7]!
Costs of Online Price Search 712
Motivation to Online Price Search 713
Online Price Search Behavior 714

Exogenous
Ease of Access to Online Information
Education
lntemet Involvement
Need for Cognition
Perceived Budget Constraints
Perceived Price Dispersion
Perceived Time Pressure
Subjective Knowledge of Internet Search
Trust in W Sellers

WWW Shopping Enthusiasm

E1
E2
E3
E4
Es
Ea
E7
E8
E9
Em

 

101

Table 5.19

Hypothesized Relationships and Empirical Findings

 

 

 

 

Effect of
Standardized
Hypothesized Observed Regression Significance
H# Construct l on Construct 2 Sign Sign Coefficient (alpha < 0.05)
Hl Ability Price Search Positive Positive 0.004 Not Sig.
H2 Education Ability Positive Positive 0.023 Not Sig.
Subjective
Knowledge of
H3 lntemet Search Ability Positive Positive 0.791 Significant
H4 Motivation Price Search Positive Positive 0.612 Significant
H5 lntemet Involvement Motivation Positive Positive 0. 137 Significant
H6 Need for Cognition Motivation Positive Negative -0. 105 Significant
WWW Shopping
H7 Enthusiasm Motivation Positive Negative 0052 Not Sig.
H8 Costs Motivation Negative Negative -0.71 1 Significant
Subjective
Knowledge of
H9 lntemet Search Costs Negative Negative -0.258 Significant
H10 Time Pressure Costs Positive Positive 0.483 Significant
Ease of Access to
H l 1 Online Information Costs Negative Negative -0.002 Not Sig.
Perceived Price
H14 Dispersion Motivation Positive Positive 0.419 Significant
Perceived Budget
H 15 Constraints Motivation Positive Negative 0034 Not Sig.
H16 Trust in Sellers Motivation Positive Negative -0.072 Significant

 

Note:

Table is based on the revised proposed model presented in Figure 5.1.

Hypothesis 12 and 13 were eliminated as a result of model revisions.

Hypothesis 14, 15, and 16 were revised to affect motivation.

102

Table 5.20

Squared Multiple Correlations for Structural Equations

 

Squared multiple correlation (R2)

 

 

 

Endogenous latent variable for structural equation
Ability to Online Price Search 0.628
Costs of Online Price Search 0.349
Motivation to Online Price Search 0.743
Online Price Search 0.376

 

Hypothesis two was not supported. Level of education did not have a significant
influence on individuals’ perceived ability to online price search (0 = 0.023), due perhaps
to the particular sample employed. However, hypothesis three was supported. The effect
of subjective knowledge of intemet search on perceived ability to online price search was
significant, positive, and strong (B = 0.791). Individuals knowledgeable about
information search on the WWW appear to view online price search as an easy
application of their prior search experience to this particular type of search task. The
strength of this relationship was also evident in the considerable amount of ability’s
variance explained for by the structural model (R2 = 0.628).

Hypothesis four was supported. Motivation to online price search was found to
have a significant positive affect on online price search behavior ([3 = 0.612). As
specified, motivation was the sole significant determinant variable to have a direct affect

on online price search, and the strength of this relationship was considerable. Seven

103

constructs were hypothesized to affect motivation to online price search, and five of these
links were significant. The structural model accounted for nearly three fourths of the
variance in motivation to online price search (R2 = 0.743). The antecedent latent
variables of motivation clearly have strong explanatory power.

Internet involvement had a significant positive link with motivation to online
price search (B = 0.137). Hypothesis five, therefore, was supported. Individuals highly
involved with the intemet view the W as an important part of their lives, helping
them to accomplish objectives and achieve goals. Through involvement with the
medium, individuals recognize the W as a tool that facilitates tasks, and they view
price search as one of the many tasks that are made easier. In this regard, price search is
not a matter of something that they know how to do well, but something that they know
can be done easier.

Hypothesis six was not supported. Need for cognition (NFC) had a significant
negative effect on motivation to online price search (B = -.105), though this relationship
had been hypothesized to be positive. Individuals high in NFC showed a tendency to
acquire more information in experimental search tasks (Verplanken et al., 1992). It
appears, however, that there is not a direct translation to the specific acquisition of price
information in an electronic marketplace. The use of a severely reduced form of the
classic NFC scale may have contributed to this finding.

Hypothesis seven was not supported. Though hypothesized to have a positive
effect, WWW shopping enthusiasm did not show a significant link to motivation to
online price search (B = -0.052). It is perhaps not surprising that W shopping

enthusiasm had such a weak negative effect on motivation to online price search.

104

Finding lower prices, at best, can be only a side benefit for hedonic shoppers, who by
definition primarily are seeking escape and distraction in their consumer behavior.

Costs of online price search had a significant negative influence on motivation to
online price search (B = -0.711). Hypothesis eight, therefore, was supported.
Individuals’ motivation to search for lower prices appears to be reduced considerably
when the costs associated with online search are high. High search costs clearly act as a
disincentive to online search through severely decreased motivation.

Hypothesis nine was supported. Subjective knowledge of intemet search had a
significant negative effect on perceived costs of online price search (B = -0.25 8).
Individuals who believe themselves to be highly knowledgeable about intemet search see
searching on the W as a task familiar to them. They believe that they can engage in
and complete online price search with reduced effort and time, because of their task
knowledge, and they, therefore, perceive fewer costs associated with this type of search
task.

Time is a particularly strong influencer of individual’s perception of the costs
associated with online price search. Perceived time pressure had a significant positive
effect on costs of online price search (B = 0.483). Hypothesis ten, therefore, was
supported. When time is limited, the costs of online price search increases. In such
instances, individuals’ value of time, a limited commodity, tends to outweigh their
perception of potential gain in savings from online price search.

Hypothesis 11 was not supported. It was hypothesized that the increased speed
with which individuals access the WWW would reduce their perception of the costs

associated with online price search. Ease of access to online information, however, did

105

not show a significant link to costs of online price search (B = 0002). Access speed and
search costs may not be related as originally reasoned from the research of Rose, Lees,
and Meuter (2001).

Just over one-third of the variance in costs of online price search was accounted
for in its structural model (R2 = 0.349). This is an encouraging finding, though it appears
that the latent variables not specified to have a direct link and/or other factors not
specified in the model make significant contributions to the perception of online search
costs.

Hypothesis 12 and 13 were eliminated as a result of the reconceptualization of the
motivation construct to include benefits as an inherent part of its composition. For the
same reason, Hypothesis 14, 15, and 16, initially specified to affect benefits, were revised
to influence motivation.

Hypothesis 14 was supported. Perceived price dispersion had a significant
positive effect on motivation to online price search (B = 0.419). The belief that prices in
the online shopping environment vary considerably from seller to seller appears to act as
a strong incentive for individuals to search for lower prices. The expectation of savings
is a driver for online buyers’ motivation to compare the prices of WW sellers.

Hypothesis 15 was not supported. Perceived budget constraints did not show a
significant link to motivation to online price search (B = -0.034). Budget constraints may
not have acted as a motivator for the student sample used for the study, 80% of which
reported having some type of employment.

Trust in W sellers was hypothesized to have a positive effect on motivation

to online price search. A significant relatively weak negative link was found instead

106

(B = -0.072). Hypothesis 16, therefore, was not supported. Buyers who conduct online
price search have shown a preference to purchase from branded WWW sellers, rather
than sellers they know not to be the low price leader (Smith & Brynjolfsson, 2001). It is,
perhaps, this same buyer trepidation that accounts for this frnding. High transaction trust
individuals may be biased toward familiar online sellers, which in turn accounts for their
high trust. One might expect, therefore, diminished motivation to compare WWW sellers
on any number of characteristics, including price. Ultimately, online transaction trust’s
strongest and most relevant influence on online consumer behavior may have less to do
with search and more to do with choice behavior (Smith & Brynjolfsson, 2001).

Just over one-third of the variance in online price search behavior was accounted
for in its structural model (R2 = 0.376). This is a very acceptable finding. It exceeds the
amount of variance accounted for in earlier price search research that exclusively used
the costs vs. benefits framework (R2 = 0.140) (Goldman & Johansson, 1978). It also
exceeds the amount of variance accounted for in research that specifically examined price
search for durables (R2 = 0.152) (Zimmermann & Geistfeld, 1984). In addition, it is
within the range of the variance accounted for in price search research conducted in the
context of grocery shopping (R2 from 0.120 to 0.520) (Beme, Mugica, Pedraja, & Rivera,
2001; Carlson & Gieseke, 1983; Lichtenstein et al., 1993; Putrevu & Ratchford, 1997;
Urbany et al., 1996).

Yet, despite the strong link of motivation on search (B = 0.612), it cannot be
concluded that motivation is the single principal factor driving online price search
behavior, since just over 60% of its variance remains unexplained. It appears then that

there are other factors influencing online price search, either not specified to affect search

107

directly and/or not specified in the model. For example, modification indices indicate
that perceived costs of online price search may possess some direct explanatory power.
Model Modifications

Based on the assessment of the over model fit, the causal model should not be
rejected. However, the preceding detailed examination of the structural model revealed
that the model’s specification was less than optimal. Through modifications
improvements were sought in the model’s overall fit and the amount of variance
accounted for in the model’s structural components.

Modifications

Three modifications were made to the measurement model. Modification indices,
ranging from 36.304 to 109.712, suggested that three items were related strongly to other
latent variables. Therefore, item three of the costs scale, item two of the motivation
scale, and item three of the ability scale were eliminated from the analysis. The removal
of these items did not compromise the reliability of their respective scales.

A number of modifications were made to the structural model. The path from
ability to online price search behavior was eliminated because it was not found to be
significant. A relationship between ability and search, however, still was believed to
exist, given the substantive theory and testing of this relationship (Bettman, 1979;
Bettman & Park, 1980; Duncan & Olshavsky, 1982). A modification index of 48.263
indicated that specifying a path between ability and motivation would improve model fit.
Ability then would have then an indirect effect on search though motivation. This was
consistent with the Srinivasan (1987) model of information search for automobiles, in

which ability was shown to have an indirect affect on search though benefits, which in

108

the current model has been incorporated into the motivation construct. Therefore, a path
was freed from ability to motivation with the expectation of a positive relationship.

An exceptionally high modification index (226.412) also suggested that a path
should be specified from ability to costs. Given that individuals high in ability have a
sense of how to price search, they were expected to perceive fewer costs associated with
this type of search. Thus, a negative relationship was expected.

Specifying a path from ability to costs, however, would produce an improper
estimate (B > 1.00). This would occur because, as specified, ability and costs would
share a predictor variable, subjective knowledge of online search, which has a strong
linear dependency with ability (r = .78). The link between subjective knowledge and
costs, therefore, would attenuate the relationship between ability and costs. The path
between subjective knowledge and costs was eliminated to overcome this improper
solution. The relationship of subjective knowledge on costs was maintained as an
indirect effect through ability.

To ensure the model was not restricting the effect of costs on search, by affixing it
solely as a mediated relationship through motivation, a direct path was freed from costs
to search and a negative relationship was expected. Costs were now specified to affect
search directly and through motivation.

Modification Results

The modified model represented an overall better fit to the data. A diagrammatic
representation of the modified model results is presented in Figure 5 .3. For contrastive
purposes, Table 5.21 presents empirical results for the initial model vs. the modified

model. The chi-square value was reduced from x2 (df= 1269, N = 587) = 3876.246, p =

109

0.0. to x2 (df= 1118, N = 587) = 3149.146, p = 0.0. The RMSEA was decreased from
0.059 to 0.055, bringing it closer to the 0.05 indicating a good model fit. The NFI, NNFI,
and CFI, while virtually unchanged, remained comfortably above the 0.90 indicating
acceptable model fit. The RMR was reduced from 0.101 to 0.073. Likewise, the
standardized RMR was reduced from 0.085 to 0.065, bringing both values closer to the
preferred 0.05.

As a result of modifications, a better fit to the data also was observed in the
structural model. Seven of the ten exogenous variables were found to have a significant
relationship with one of the mediating variables (i.e., ability, costs, or motivation). The
three exceptions were ease of access to online information, education, and perceived
budget constraints. These seven exogenous variables also had significant indirect effects
on online price search behavior. In rank order, the standardized total effects of the
exogenous variables on price search were: subjective knowledge of intemet search
(0.281), perceived price dispersion (0.184), perceived time pressure (-0. 172), WWW
shopping enthusiasm (-0.091), intemet involvement (0.052), trust in W sellers (-
0.048), and need for cognition (0043). Three of these seven relationships, need for
cognition, trust in W sellers, and WW shopping enthusiasm, were not the

predicted direction.

110

 

 

             

 

8006050m 900025 333 0200on 00.5 000 Z 000000202: 00:00:05

 

mwod-
086-

t \

30800000 00w0sm 002085
08.0.

 

   
       

09:00m 005 00:00 9
0Io||00>q0 2| 02000005
0: nd+

02000005 005 00200005

 

 

   

     

 

  
    

  

 

036+
306+
0000035
2333:
80:9...va 6:08:80 semfimfiuflmw 2 300+
00 E 806000 083m umum . . . 00000m 0000005
.00 030—3002 0>0o030m
«00.0-

mmvd-

    
    

00.80m 00:5 00:00 .50 0.50005 0005. 0030005

mm_ .o- flared. 00300005

320+

 

    

000000055
00:00 9 00000< .00 03M
300+

20.0 n 20:00 :02: 08560000 20800060 38:00: 08:: 0:00; 6
000000000 02000000 000000080 000 00003 A:

 

111

Table 5.21

Empirical Results for Initial Model vs. Modified Model

 

 

 

 

 

Initial Modified
Test statistic model model
Overall Model Fit Assessment
Degrees of Freedom (df) 1269 1118
Chi-square (x2) 3876.246 3149.146
Root Mean Square Error of 0.059 0.055
Approximation (RMSEA)
Normed Fit Index (NFI) 0.935 0.936
Non-Normed Fit Index (NNFI) 0.953 0.956
Comparative Fit Index (CFI) 0.957 0.960
Root Mean Square Residual (RMR) 0.101 0.073
Standardized Root Mean Square 0.085 0.065
Residual (SRMR)
Measurement Model Assessment
(Squared Multiple Correlations for Structural Equations)
Ability to Online Price Search (R2) 0.628 0.566
Costs of Online Price Search (R2) 0.349 0.508
Motivation to Online Price Search (R2) 0.743 0.711
Online Price Search (R2) 0.376 0.361

 

112

One change in the relationship of the exogenous variables was found. WWW
shopping enthusiasm, which prior to modifications was found to have a non-significant
negative effect on motivation (B = -0.052), now exhibited a significant negative link (B =
-0. 177), albeit a relatively weak one. A positive relationship had been predicted, based
primarily on research using multiple focus groups that found price search was a regular
practice for both hedonic and utilitarian online shoppers (W olfinbarger & Gilly, 2001).
This finding, however, did not translate into a relationship where hedonic online shoppers
directly seek lower prices. Rather, it first must be acknowledged that finding lower
prices, at best, can be only a side benefit for hedonic shoppers, who by definition
primarily are seeking escape and distraction in their consumer behavior.

Ability to online price search was found to have a significant positive effect on
motivation (B = 0.422) and a significant negative effect on costs (0 = 0462). Through
its relationship to motivation and costs, ability had a total effect on price search of 0.375.
Though the amount of variance accounted for in the ability structural model decreased
(R2 from 0.628 to 0.566), it remained at a respectable level. The amount of variance
accounted for in the costs structural model increased (R2 from 0.349 to 0.508). Costs also
were found to have a significant direct negative link to online price search behavior (B = -
0.123). In conjunction with its relationship to motivation, costs had a total effect of -
0.341 on price search behavior. While the variance accounted for in the modified
structural model for the mediating variables were all in a respectable range after
modifications (> .50), no improvement was seen in structural model for online price
search behavior. The amount of variance accounted for in the online price search

structural model was virtually unchanged (R2 from 0.376 to 0.361). Overall the modified

113

model should not be rejected, however, just over 60% of the variance in the structural
model of online price search behavior remained unexplained. One should expect,

therefore, that there are variables with explanatory power not specified in the initial or

modified model.

 

114

Chapter 6

SUMMARY AND CONCLUSION

Introduction

This chapter first summarizes the research study, reviewing the project’s
objectives, design, and methods. It then presents an overview of the assessment of the
measurement model, overall model fit, and structural model. A discussion of this
research’s contributions is then presented, followed by a report of its limitations. The
dissertation concludes with an agenda for future research on the topic.

Objectives of the study

Price search is a particularly common prepurchase behavior among a substantial
portion of online consumers. The proliferation and routinization of price search behavior
among online buyers should be of critical concern to online sellers. Online sellers’
pricing positions potentially are compromised in an electronic marketplace made
extremely efficient through a population dominated by a breed of better-informed
consumers. It follows from this perspective to envision destabilization effects ensuing as
price search becomes a standard prepurchase practice among online consumers, and, in
doing so, seeps seller incentive to participate in electronic commerce.

A more reasonable perspective, however, requires the acknowledgement that no
marketplace operates with complete efficiency. The degree of price search behavior will

continue to vary between online consumers. It is with an understanding of this variation

115

that online sellers, depending on their customer orientation, will be able to either
facilitate this type of search or build business models and strategies that account and even
compensate for this type of online consumer behavior.

How then can this variation be understood? There is a conspicuous lack of
research specifically examining consumer online price search behavior. The research that
does exist on the topic leans heavily on theories of reduced buyer search costs adapted
from economics. Yet, a purely economic explanation appears inadequate. Online price
search behavior clearly is facilitated through an electronic marketplace’s ability to
maximize access to product information, while minimizing the costs associated with
acquiring this information. These costs, however, are never eliminated completely,
regardless of their reduction relative to search costs in non-electronic marketplaces.
Consumers’ circumstances and characteristics also must be considered to generate a
robust understanding of the determinants of the amount of price search that a consumer
engages in prior to purchase in an electronic marketplace.

To contribute research in this area, this dissertation proposed and tested a causal
model of the factors influencing consumer external price search behavior in an electronic
marketplace. The theoretical model of consumer external information search proposed
by Schmidt and Spreng (1996) was employed as a framework within which to specify the
relationship between factors. In its specific adaptation of the mediating variable structure
proposed by Schmidt and Spreng, the present model attempted empirical verification of a
broad theoretical perspective on the online price search phenomena that combines search
costs economies with both motivational and information processing theory. The use of a

causal model also was viewed as an opportunity to assess the applicability of traditional

116

determinants of consumer search to an electronic shopping context and to test constructs

believed uniquely salient to consumers search for lower prices in a computer mediated

environment. The proposed model was operationalized and validated in the context of

the WWW as it is currently the most pervasive consumer electronic marketplace.
Research Design

Select items from preexisting scales were adapted to the context of price search
and intemet shopping and used in combination with original items to form the basis for
the construction of the survey research instrument. A pilot study was conducted to assess
the validity of the indicators and the reliability of the scales used to measure the 15 latent
variables in the proposed model. Using 127 student subjects, scales showed acceptable
reliability.

For the main study, a total of 728 surveys were collected from qualified student
participants, of which 587 were usable after data screening. Qualified participants had
made at least one purchase through the W in the 3 months preceding survey
administration. Almost half of participants (49%) indicated that they had compared
prices online prior to their last WWW purchase. On average, members of this group
compared 4 prices before making their purchase.

Method of Analysis

In a pilot study, confirmatory factor analysis, using the LISREL 8.5 statistical
software package, was conducted to determine the validity of the scale items used to
measure each construct. Scale reliability was assessed using Cronbach Alpha reliability

coefficient and principal components analysis figures produced in the SPSS 11.0

117

statistical software package. Discriminaant validity was assessed through an examination
of the correlations between latent variables.

The two-step approach to model building was employed for the main study
(Anderson & Gerbing, 1988; James et al., 1982). This process emphasizes the analysis of
the measurement model before examination of the structural model and the overall model
fit. To analyze the measurement model, as in the pilot study, indicator validity was
assessed using confirmatory factor analysis. Scale reliability was assessed using
Cronbach Alpha reliability coefficient and principal components analysis, and
discriminant validity was assessed through an examination of the correlations between
latent variables. Maximum Likelihood estimation in the LISREL 8.5 statistical program
was used to fit the observed covariance matrix against the predicted covariance matrix in
analysis of the measurement model, structural model, and overall model fit. Estimation
of the proposed model adhered to the assumptions of the general LISREL model.

Summary of Main Study Results
Measurement Model Assessment

Assessment of the measurement model revealed that all indicators and their
corresponding error variances were significant. Nine items were deleted because of weak
single item factor loadings. As a result, the perception of budget constraints was
measured using a single item. All multiple items scales were found to have sufficient
reliability coefficient values. Unacceptably high correlations were found between price
search and motivation and between benefits and motivation. These figures indicated

multicollinearity between the scales.

118

The model’s measurement and specification were revised to address issues of
severe linear dependency between price search and motivation and between benefits and
motivation. Online price search, which had been operationalized using a combination of
behavior and tendency items, was measured instead using a single behavior oriented item
that asked participants to indicate the number of prices they compared prior to their last
WWW purchase. The correlation between online price search and motivation was
dramatically reduced as a result of this procedure.

Through a closer examination of the relationship between benefits and
motivation, it was realized that that when dealing with price search specifically, as
opposed to information search in general, both reaping benefits and fulfilling goals (i.e.,
motivation) are rooted in the same product and predicated on achieving the same
outcome (i.e., finding the lowest price). It appeared redundant, therefore, to view the
benefits of and motivation to conduct online price search as separate constructs. The
motivation construct, thus, was reconceptualized as inherently including benefits of price
search as an implicit element of its composition. The initially proposed model, Figure
2.2, was respecified to reflect this new conceptualization, as presented in Figure 5.1.
Overall Model Fit Assessment

Collectively the overall fit statistics indicated that the model represented an
adequate fit to the data, but also that there were areas of the model within which less than
optimal specification existed. For example, the model’s NFI (0.935), NNFI (0.953), and
CFI (0.957) were encouraging indicators that the model overall fit the data well. Less

encouraging values were found for the model’s RMR (0.101) and standardized RMR

119

(0.085). Through an analysis of the results of the structural model, the areas in the
model’s specification responsible for these discrepancies were identified.
Structural Model Assessment

Before summarizing the results of the structural model, presented
diagrammatically in Figure 5.2, a brief overview of the model’s variables is provided.
The causal model of price search behavior in an electronic marketplace has four latent
endogenous variables and ten latent exogenous variables. The endogenous variables are:
Online Price Search Behavior, Ability to Online Price Search, Motivation to Online Price
Search, and Costs of Online Price Search. The exogenous variables are: Education,
Subjective Knowledge of lntemet Search, Perceived Time Pressure, Ease of Access to
Online Information, Perceived Price Dispersion, Perceived Budget Constraints, Trust in
W Sellers, lntemet Involvement, Need for Cognition, and WW Shopping
Enthusiasm.

Empirically, nine of the 14 structural relationships proposed in the model were
found to be statistically significant. Seven of these nine significant links were in the
direction hypothesized.

From the information processing perspective, one would expect ability to have an
effect on price search behavior. The direct effect, however, was not significant.
Education did not show a significant influence on ability. Subjective knowledge of
intemet search, individuals’ assessment of their general WWW search capability, did
have a significant positive link to perceived ability to price search.

Motivation to online price search, derived from the social

psychological/motivational perspective, was expected and did have a significant direct

120

positive effect on online price search. In rank order, the factors with significant direct
links on motivation were: costs of online price search, perceived price dispersion, intemet
involvement, trust in WWW sellers, and need for cognition. WWW shopping enthusiasm
and perceived budget constraints did not show significant effects on motivation.

Cost was the major determinant variable on motivation. Motivation was
negatively influenced by costs. The direction of this relationship also is consistent with
the costs vs. benefits framework, since motivation was conceptualized as possessing
benefits as an intrinsic part of its composition. A sense of time constraints when
shopping on the W was a strong positive influence on perceptions of the costs
associated with online price search. This was contrasted by individuals’ subjective
knowledge of intemet search, which had a negative effect on the perceived costs of
online price search. Surprisingly, easy access to intemet information showed no
influence on perceived costs.

Perceived price dispersion, individuals’ belief that prices noticeably vary between
WWW sellers, was the strongest determinant on motivation afier costs. While not as
influential on motivation as costs, perceived price dispersion’s effect was considerable.
Individuals’ intemet involvement, the personally relevance of the intemet in executing
objectives and/or achieving goals, also had a positive influence on motivation. Contrary
to expectations, neither need for cognition nor trust in W sellers positively affected
motivation, and their negative influence was relatively weak. The variance accounted for
in the structural models of ability (R2 = 0.628), motivation (R2 = 0.743), costs (R2 =

0.349), and price search behavior (R2 = 0.376) were all acceptable.

121

Model Modifications

To improve structural assessment values and global fit statistics, the model’s
specification was enhanced through modifications. Empirical results of the modified
model’s estimation are diagrammatically represented in Figure 5.3, and Table 5.21
compares results.

Four major modifications were made to the structural model. The path from
ability to online price search behavior was eliminated because it was not found to be
significant. Paths were specified from ability to motivation and from ability to costs.
Costs were specified to affect price search directly.

The modified model represented an overall better fit to the data. The modified
model’s RMSEA (0.055), RMR (0.073), and standardized RMR (0.065) were all closer to
the threshold value indicating a good model fit (0.05). A better fit to the data also was
observed in the structural model. Ability to online price search was found to have a
significant positive effect on motivation and a significant negative effect on costs. Costs
also were found to have a significant direct negative link to online price search behavior.
In addition, the amount of variance accounted for in the costs structural model increased
(R2 from 0.349 to 0.508). The amount of variance accounted for in the online price
search structural model, however, was virtually unchanged (R2 from 0.376 to 0.361).

The modified model of external price search behavior in an electronic
marketplace was found to have moderate explanatory power. In addition, it contained
considerable meaning with regards to the relationships between determinant variables.
The modified model, therefore, was not rejected. Yet despite modifications, just over

60% of the variance in online price search behavior remained unexplained by the

122

structural model. Thus, as with the initial model, one should expect that there are
variables with explanatory power not specified in the modified model.
Contributions of the Study

This study’s empirical verification of a causal model that specifically focused on
online price search is, in itself, a contribution toward a more complete understanding of
online consumer behavior. Prepurchase price search is a common behavior among a
considerable portion of online consumers. Previous research, however, has generally
taken a broader perspective, looking more at online information search than price search
in particular.

Much of consumer search research has focused on bivariate relationships, which
have not proven robust when tested simultaneously that this deficiency has been
described as a pressing research need (Wilkie & Dickson, 1991). This study advances
our knowledge of the search process through its structural explanation of the relationships
of different search determinants as they work concomitantly. For example, it had been
hypothesized, consistent with bivariate search research, that ability would have a positive
direct effect on online price search. While a significant bivariate relationship was found,
the link was not significant in the presence of motivation, thereby leading to the
conclusion, supported in multivariate search research (Srinivasan, 1987), that ability’s
effect on search is indirect. Likewise, the influence of subjective knowledge on search
was not direct, but rather completely mediated through ability. Another example is the
finding that perceived price dispersion had a strong affect on motivation to online price

search. With exception (Urbany et al., 1996), there had been generally weak support for

123

this part of the classic information economics theory (Claxton et al., 1974; Duncan &
Olshavsky, 1982; Goldman & Johansson, 1978).

This study also has empirically verified a theoretically integrated causal model of
consumer online price search behavior. In this regard, it has extended online search
research beyond its typical focus on search costs economics. Employing the mediating
variable framework proposed by Schmidt and Spreng (1996) as its base, the model has
unified theoretical perspectives and determinants from economics, social
psychology/motivation, and cognitive psychology/information processing.

This study’s contributions also include the testing of the applicability of numerous
determinants found to influence consumer search in non-electronic markets (e.g.,
perceived time pressure). In addition, a number of factors unique to an electronic
shopping environment (e.g., ease of access to online information) that had been expected
to influence consumer online price search were found to have weak, counter intuitive,
and/or no effect.

Marketing practitioners and online sellers may find the model to be useful in
efforts to design promotional campaigns and WW sites. Sellers and marketers of sites
that have value-pricing strategies should design their sites to facilitate search for price
information, given time limitations’ influence on the costs associated with search, and in
turn, costs strong negative effect on the motivation of comparison-price shoppers. In
promoting their site, value-leader marketers should highlight the price dispersion in the
online marketplace for their particular product(s), since comparison-price shoppers are
highly motivated to search when they believed that prices varied considerably between

sellers.

124

Limitations of the Study

As with all scientific endeavors, this study contains a number of limitations. This
study employed survey method for data collection. Some memory bias, therefore, exists
in the data, since it was collected post-hoc up to three months after a WWW purchasing
experience. Speculation on the extent of this effect is impossible. However, it should be
noted that 70% of participants had made their last purchase in the 8 weeks preceding
survey competition, 44% had made their last purchase in the 4 weeks preceding, and 23%
had made their last purchase in the 2 weeks preceding. This was considered encouraging
news for the recall reliability of participants (49%) who indicated that they had compared
prices online prior to their last WWW purchase.

This study used a convenience sample of students from a large Midwestern
university. While the use of a convenience sample was suitable for this study’s
exploratory tests of the model’s structure, it would be inappropriate to generalize the
results to the entire online consumer population. Further, the use of a student sample
composed heavily of undergraduate students undoubtedly had some influence on the
results, since this group was composed principally of individuals 18 to 23 years old. A
representative sample likely would produce some differences in the model’s structure.
For example, using a student sample, education had no effect on perceived ability to
online price search. A probability sample of typical consumers with more variance, it is
suspected, would yield the predicted result. That said, given the continued diffusion of
the WWW into the workplace, home, and all levels of education, regular access and use
of the WWW may be increasingly a more revealing variable than education in assessing

consumers’ ability to search for lower prices. The typical time constraints on adults with

125

families should be expected to manifest itself in the perceived time pressure construct,
and the strength of the relationship of time constraints on costs, therefore, should be
expected to increase. Likewise, the financial pressures of adults with families may
appear, through the perceived budget constraints construct, to act as a strong motivator of
online price search, though it did not show a significant effect on motivation in tests
using the present student sample.

The use of a large sample size reduces the possibility that the results are a
capitalization on chance. However, it is only through further development and
verification of the model’s measurement and structural components that more definitive
conclusions can be drawn with regard to this point.

While there was limited variance in the sample’s age, there was considerable
variance with regard to the types of products participants purchased prior to their online
price search. This heterogeneity in products resulted from the qualification of
participants based on their last WWW purchase, regardless of its product class. The use
of a non-product specific model surely affected its structure, and thus, the application of
the model to a single product class should be expected to produce varying results. For
example, if the model were to be tested against a class of products for which purchase
requires significant financial expenditure (e. g., computers), one might expect that trust in
W sellers would exhibit a strong role in the price search phenomena, whereas it now
shows little effect.

This study also would have benefited from improved construct operationalizations
to reduce those instances where the correlations between constructs were exceptionally

high, particularly with regard to the mediating variables in the model. The construct

126

values obtained in the current study, while not precluding model estimation, generated a
less than optimal measurement model. As was suggested previously, the strong linear
dependency between the various aspects of price search (e.g., ability and motivation) may
also suggest that the differences in these facets have relationships of extreme subtle
nuance. In addition to more distinct operationalizations, one might consider, therefore, a
more sensitive measurement technique, since a method requiring individuals to
consciously differentiate between these aspects (i.e., survey) may not be precise enough.
This question remains unexplained in this study.
Future Research Directions

The model of online price search behavior appears to contain moderate
explanatory power on the whole and richness in meaning with regard to the structural
relationships between determinant variables. A considerable amount of the variance in
online price search remained unexplained by the model (62%). There are, therefore,
factors not specified in the model with influence on consumer online price search
behavior. Further speculation on the origins of these determinants and empirical
verification of these factors is necessary to build a fuller understanding of this critical
phenomenon. A broader perspective on factors influencing price search unique to a
computer mediated shopping environment may lend direction to this effort. This research
may need to be exploratory in nature and also may be conducted best from a qualitative
orientation. This qualitative approach might include focus groups and/or observations of
individuals who can be identified as possessing a propensity toward online comparison-
price shopping. Additionally, future research should consider the use of experimental

methods to study the model’s structure and the online price search phenomenon. The use

127

of an experimental online shopping environment would be optimal for assessment of
certain constructs, for example, ability to online price search.

Having been tested against a convenience sample of students, the model’s
external validity remains unknown. The model, therefore, now should be examined
against a sample of actual online purchasers.

In addition to determining the model’s external validity, the use of an actual
online consumer sample would be an opportunity to improve the reliability of a number
of measures, if the time lag between the shopping incident under question and study
participation could be minimized. Ideally, participation should be sought and obtained
immediately following an individual’s WWW purchase. In such circumstances, the
recall reliability of participants’ prepurchase online price search behavior could be
assessed as very high.

Measurement of time pressure, ease of access to online information, and WW
shopping enthusiasm also would be improved dramatically. In the current model,
perceived time pressure was conceptualized as a measure of individuals’ general sense of
time scarcity when shopping online, arid it was found to make significant contributions to
individuals’ sense of the costs associated with online price search. Using an actual
consumer sample shortly following a WWW purchase, time pressure could be tested as it
relates to the specific purchasing incident under study.

There also would be improvements in the accuracy of measures assessing the
online shopping location (e.g., home vs. work) and intemet access speed. Though ease of
access to online information (i.e., access speed) was not found to have a significant effect

on perceived costs of online price search, there remains a strong belief that access speed,

128

by way of its relationship to time, is related to costs. In exploring intemet access speed,
there also is an opportunity to begin answering burgeoning questions regarding the
effects of broadband connections on the behavior of online consumers. Dial-up intemet
connection currently dominates US households, but broadband (e.g., DSL and cable)
continues a steady pace of diffusion. The effect of high-speed access to electronic
markets is an issue researchers of online consumer behavior are already contemplating.
Objective measures of WW access ease, for example actual speed of home WWW
connection and processing speed of home computer at the time of purchase, are an
alternative operationalization that might be considered for this construct in subsequent
related research.

Questioning shoppers shortly following WWW purchase also would allow for
assessments of the extent to which the shopping incident was directed toward a specific
product, brand, and/or model or was part of a general tendency toward enthusiastic
browsing behavior. Any prepurchase online price search behavior then might be
understood in terms of it being intentional vs. incidental. Subsequent related research
should adopt measures that tap into price search as a secondary or tertiary goal of hedonic
online shoppers. In this manner, there could be better representation and verification of
the relationships suggested in the findings of Wolfinbarger and Gilly (2001), in which
price search was found to be a factor involved in the browsing behavior of WW
shopping enthusiasts.

The use of interactive technology makes it possible to obtain study participants
shortly following or immediately after their WWW purchase. Timing of this type,

however, clearly requires, the partnering with and approval of an online seller(s), and

129

therefore, the likely focus of the model on a specific type of product class. The model of
consumer online price search was conceptualized as a general model. Given its general
orientation, it is expected that the model possess broad applicability to specific product
classes. The model’s replication across multiple product classes would increase its
validity. An area of interest left uncharted in this research is the affect of online purchase
price and frequency on the structure of the model. For example, how varied are the
factors influencing online price search between less expensive more regularly purchased
items (e.g., books) vs. more expensive less frequently purchased items (e.g., computers)?
Future research also easily might compare the model’s structure when tested against
products that vary in purchase involvement (i.e., high vs. low) or goods that differ in
orientation (e. g., search vs. experience or commodity vs. differentiated).

In specifying a product class against which to test the model, future research also
might consider the influence of the online marketplace structure of different types of
merchandises. That is to say, that depending on the type of product class under study,
one likely will find variance in the number of viable vendors. An electronic marketplace
in which there are many sellers may have effects on actual price dispersion and/or
consumer perception of price variability. One also might expect effects on the time
online consumers spend in price search, and also thus the costs of their search, depending
on the differing size of the online market for a given product class. Still deeper analysis
of specific product class’ online market structure and their effects on online price search
might consider the role of the varying number of well known and branded sites. Future
research applying the model to specific product classes also would be keen to keep in

mind the association between time and price with particular products (e.g., airline

130

tickets). This is yet another type of market structure with likely effects on online price
search behavior.

While this model brought empirical verification to a theoretically integrated
perspective of the online price search phenomenon, unifying the economic and marketing
approachs to consumer search, it remain unknown which perspective possesses the
strongest explanatory power. The model’s structure is based principally on research
investigating information search that was adapted to study price search specifically. As
was revealed in the extremely close association between consumers’ perceptions of
benefits and motivations to price search, when considering price search specifically, as
opposed to information search in general, this behavior has a distinct economic
orientation. This appears fact, despite motivation to online price search demonstrating
the strongest influence on search behavior. Future research, therefore, should consider
splitting the model into its economics and psychological components. The two resulting
models then could be compared for their ability to explain online price search, and
definitive conclusions could be drawn.

There are, in fact, many enhancement and extensions to the current model that
would further knowledge of consumer external price search behavior in an electronic
marketplace. It is a topic that deserves the serious attention of researchers seeking to
fillly understand this new and unique consumer distribution channel, currently made
available through the mass diffusion of the WWW. While online purchasing only
accounts for a small portion of total US retail sales, this figure is growing as consumer
use and facility with online shopping increases. One wonders, in closing, how might

price search in non-electronic markets change as consumers, increasing keen to the

131

advantages of shopping in an electronic marketplace, become aware and then accustomed
to shopping with vast information? Indeed, there remain many questions with regard to
the development of the W as a vast electronic marketplace and the consequences of
such technology on our consumer oriented society. This study has attempted to

contribute to our understanding by answering only but a portion of these questions.

132

APPENDIXES

133

APPENDIX A

134

APPENDIX A

PILOT RESEARCH INSTRUMENT

CONSENT FORM

Thank you for your voluntary participation in completing this questionnaire.

To participate in this study you must:

1. Be an undergraduate or graduate student at Michigan State University
2. Be over 18 years old
3. Have purchased at least one item on the world-wide-web in the past three months

The researcher of this study, which is titled A Model of Consumer External Price Search Behavior in an
Electroni c Marketplace (World-Wide-Web ), is conducting this survey in order to understand what factors

influence a consumer’s search for lower prices when shopping on the world-wide-web.

Voluntary completion of this questionnaire will take approximately a half an hour. If at any time you feel
unable or unwilling to complete the questionnaire, please feel free to stop.

A“ quellified participants are automatically entered into a drawing for a $100 cash prize. The winner will
be notlfied by phone. Please remember to include your name, phone number, and email address at the end
of the questionnaire for notification purposes. The prize will be drawn upon full completion of the data.

In certain cases, for your participation, you may also receive extra credit points as determined by

your instructor. In such cases, upon completion your name will be submitted to the instructor of
your CO“ rse to receive extra credit points.

Other than in those cases where your name will be submitted to the instructor of your course to receive
extra credit points, your name, phone number, and e-mail address will not be distributed. All information
regarding participants will remain confidential. Any information that you provide and the data collected

from the questionnaire will be kept in the strictest confidentiality. Your privacy will be protected to the
max1mum extent allowable by law.

In addition, we will not identify any individual who participates in this survey by name, phone number or
email In any report or publication. You are also given the option to decline to answer any question at any

time. .If you have any questions or concerns about the study you may contact James Ramos at
W or (517) 372-8095, or Dr. Charles Salmon, Senior Associate Dean, College of

communication at salmon@msu.edu or (517) 355-3410.

Ifyou baVe any questions or concerns regarding your rights as a study participant, or are dissatisfied at any

"gniwlth _ any aspect of this study, you may contact — anonymously, if you wish - Dr. Ashir Kumar, Chair

(fa): 235(1):;Versity Committee on Research Involving Human Subjects (UCRIHS) by phone: (517) 355-2180,
' 3

53-2976, e-mail: ucrihs@msu.edu, or regular mail: 202 Olds Hall, East Lansing, MI 48824.

By Slgning and dating this page, you indicate your voluntary agreement to participate in this study.

Signature Date

 

 

Pnnt Name

 

135

You will be asked to indicate the time you start and finish this questionnaire. It is of
great interest to us to assess the time it takes to complete this entire questionnaire.
Therefore we kindly request that once you begin the questionnaire you complete it
without taking any breaks.

Please do not forget to provide us with your name, telephone number, and email address
at the end of the survey so that we may notify you if you win a prize.

Thank you.

Start time:

 

The follow questions are related to shopping and the world-wide-web and the prices you
paid for items. When asked about price, this means the total cost of an item, including
for example, any tax or shipping and handling fees. There are no right or wrong answers
to the following questions and people’s answers vary widely. Kindly indicate y_our_
personal Opinion by circling any one number. In all the questions below, do not include
any purchases from auction websites like eBay.

1) Have you made at least one purchase on the world-wide-web in the past three months,
not including any purchases from auction websites like eBay?

(1) (2)
Yes No
(Got to Question 85)

2) How often do you purchase items on the world-wide-web (not including any purchases
from auction websites like eBay)?

(1) (2) (3) (4) (5) (6) (7)
Once a Once Once Once Once a Once Once a
year or every six every every two month every two week or

less months three months to three more

months weeks

3) How many individual items have you purchased on the world-wide-web in the past
three months (not including any purchases from auction websites like eBay)?

(1) (2) (3) (4) (5) (6) (7)
1-2 items 3-5 items 6-10 items 1 1-15 15-19 20-25 25+ items
items items items

4) How much have you spent buying items on the world-wide-web in the past three
months (not including any purchases from auction websites like eBay)?

(1) (2) (3) (4) (5) (6) (7)
$1 - $49 $50 - $99 $100 - $200 - $500 - $1,000 - $2,500+
$199 $499 $999 $2,500

136

We would like you to take a minute and think back to the last time you bought something
on the world-wide-web (not including any purchases from auction websites like eBay).
Thinking back on that particular world-wide-web shopping experience, please respond to
the following questions. Again, there are no right or wrong answers. Kindly indicate
you; personal Opinion by circling any one answer. Thank you.

5) When was the last time you purchased an item on the world-wide-web (not including
any purchases from auction websites like eBay)?
(1) (2) (3) (4) (5)
Within the Within the Within the Within the Within the
past two past four past six past eight past three
weeks weeks weeks weeks months

6) What exactly did you purchase? (Please check the appropriate box)

Cl Item, please specify:
Cl Prefer not to specify
CI Cannot remember the exact item (Go to Question 14)

7) Under what category would you classify your last purchase? (Please check the
appropriate box)

Travel (Airline Tickets)

Travel (Not including airline tickets)

Books

Movies (DVD’s, VHS, etc.)

Music (CD’S, Tapes, etc.)

Car accessories

Clothing and accessories

Computer hardware

Computer software

Consumer electronics

Electronic games

Event tickets (for music concerts, sporting events, or cultural events)
Flowers

Health and beauty

Home and garden

Office related products

Sporting goods

Other, prefer not to specify

Other, gleam specify

DDDDDUUDDDDUDUUDDDU

 

137

Plow much do you know about the category you indicated in uestion 7?

 

 

 

 

 

 

 

Strongly Strongly
Disagree Agree

8) I know more about the category indicated in Question 7
than most people. 1 2 3 4 5 6 7

9) Among my circle of friends, I’m one of the “experts” on
the category indicated in Question 7. 1 2 3 4 5 6 7

10) I feel very knowledgeable about the category indicated

flgyestion 7. l 2 3 4 5 6 7

 

Please consider the category indicated in Question 7. To you this category of products is:
11) Unimportant l 2 3 4 5 6 7 Important

12) Means Means a lot to
nothing to 1 2 3 4 5 6 7 me
me

13) Irrelevant 1 2 3 4 5 6 7 Relevant

14) Did you purchase this item for yourself or for another person?

(1) (2)
Myself Another person or group

15) Please think about the web retailer from which you made your last web purchase.
Before making your last purchase from this web retailer, had you ever bought something
from this website?

(1) (2) (3)
Yes No Do not recall
(Go to Question 17) (Go to Question 17)

16) What influenced your decision to buy from this web retailer again?
(Please circle afl that apply)

(1) (2) (3) (4)
Satisfaction Felt it was Did not want to Knew they sold
with previous a fair price shop around what you
transaction(s) wanted to buy

138

17) Approximately how much did you pay for your last purchase on the world-wide-
web?

(1) (2) (3) (4) (5) (6) (7)
Less than $25 - $49 $50 - $99 $100 — $200 - $500 - $1,000+
$25 $199 $499 $999

18) Before making your last purchase on the web, did you compare different prices on the
web? (Comparing prices could involve visiting more than one web retailer or using a
shop-bot website, like mysimon.com)

(1) (2) (3)
Yes — I compared prices No — I did not compared Do not recall
before buying prices before buying (Go to Question 20)
(Go to Question 20)

19) How many prices did you compare on the web before making your last purchase on

the world-wide-web? (Please check the appropriate box)
[:1 2 El 3 Cl 4 D 5 D 6 El 7 El 8+

20) Do you own a computer with access to the world-wide-web?

(1) (2)
Yes No

21) How often do you access the world-wide-web?
Never 1 2 3 4 5 6 7 Everyday

22) How often do you access the world-wide-web to search for information?
Never 1 2 3 4 5 6 7 Everyday

139

23) From what location do you access the world-wide-web the most?

(1) (2) (3) (4) (5) (6)
Your home Your Friend’s University Public Other
workplace home computer library
lab or
library

24) From what location do you do most of your shopping on the world-wide-web?

(1) (2) (3) (4) (5) (6)
Your home Your Friend’s University Public Other
workplace home computer library
lab or
library

Please think about the location from where you do most of your shopping on the world-
wide-web when answering the following question.

The speed at which web pages load from this location is:

25) Intolerable 1 2 3 4 5 6 7 Tolerable

26) Unsatisfactory 1 2 3 4 5 6 7 Satisfactory

27) Slow 1 2 3 4 5 6 7 Fast

Please describe your ability to search for lower prices when shopping on the world-wide-
web:

28) Low 1 2 3 4 5 6 7 High

29) Inadequate 1 2 3 4 5 6 7 Superior

30) Weak l 2 3 4 5 6 7 Strong

3 1) Incapable 1 2 3 4 5 6 7 Capable

32) Not confident Confident in
in my ability 1 2 3 4 5 6 7 my ability

33) Uncomfortable Comfortable
with my 1 2 3 4 5 6 7 with my
ability ability

To me the world-wide-web is:
34) Unimportant 1 2 3 4 5 6 7 Important

35) Means Means a lot to
nothing to l 2 3 4 5 6 7 me
me

36) Irrelevant 1 2 3 4 5 6 7 Relevant

37) Worthless 1 2 3 4 5 6 7 Valuable

38) Not needed 1 2 3 4 5 6 7 Needed

140

The statements that follow concern activities, interests, and opinions related to shopping
on the world-wide-web. There are no right or wrong answers to the following statements
and people agree and disagree. Kindly indicate my personal opinion by circling any one
number for each statement from 1 (Strongly Disagree) to 7 (Strongly Agree).

Strongly Strongly
Disagree Agree

39) When shopping on the world-wide-web, “Never buy the
first one you look at” is a good motto. 1 2 3 4 5 6 7

 

40) When shopping on the world-wide-web, I am willing to
go to extra effort to find lower prices. 1 2 3 4 5 6 7

 

41) I make it a rule to visit more than one web retailer or
use a shop-bot to compare prices before I buy on the
world-wide-web. 1 2 3 4 5 6 7

 

42) I would never shop at more than one web site just to
find low prices before I buy on the world-wide-web. 1 2 3 4 5 6 7

 

43) When shopping on the world-wide-web, I often
compare the prices at more than one web retailer before
deciding where to buy from. 1 2 3 4 5 6 7

 

44) I feel very knowledgeable about searching for
information on the world-wide-web, 1 2 3 4 5 6 7

 

45) Among my circle of fiiends, I’m one of the “experts” on
searching for information on the world—wide-web. 1 2 3 4 5 6 7

 

I know how to search for information on the world-
46) wide-web. 1 2 3 4 5 6 7

 

47) I know more about searching for information on the
world-wide-web than mostlleople, 1 2 3 4 5 6 7

 

48) It is important for me to get the best price when
shopping on the world-wide-web. 1 2 3 4 5 6 7

 

49) It really doesn’t occur to me to search for lower prices
when shopping on the world-wide-web. 1 2 3 4 5 6 7

 

 

50) I am motivated to search for lower prices when
shopping on the world-wide-web. _ _1.____2_ 3 4 5 6 7 _

 

 

51) I am enthusiastic about searching for lower prices when
shopping on the world-wide-web. 1 2 _ 3 4 5 6 7

 

52) I feel driven to find the best price, when shopping on the
world-wide-web. 1 2 3 4 5 6 7 _

 

 

141

53)

54)

55)

56)

57)

58)

59)
60)

61)

62)

63)

64)

65)

66)

67)
68)

Strongly Strongly
Disagree Agree

 

I really want to find the best prices, when shopping on
the world-wide-web.

__The notion of thinking abstractly is appealing to me.

 

Thinking is not my idea of fun.

 

I prefer my life to be filled with puzzles that I must
solve.

I only think as hard as I have to.

Learning new ways to think doesn’t excite me too
much.

 

Shopping on the world-wide-web feels like an escape.

 

Compared to other things I could do, the time spent
shopping on the world-wide-web is truly enjoyable.

 

I enjoy shopping on the world-wide-web for its own
sake, not just for the items I may purchase.

 

By searching around on the world-wide—web for price
information, I am certain of making the best buy.

 

It doesn’t pay to shop around before buying on the
world-wide-web.

 

Shopping around at more than one web site helps me to
find the lowest price.

 

When shopping on the world-wide-web, there is too
much to lose by being ignorant about prices.

 

By rushing into a purchase when shopping on the
world-wide-web, one is bound to miss a good deal.

Some web sites have prices that are much lower than
others.

Prices of individual items may vary between web sites,
but overall, there isn’t much difference in the prices
between web sites.

 

142

_L ,2__ L5. __L _-.6_7_.

Strongly Strongly
Disagree Agree

 

 

69) The price of individual items often varies a lot between
_web sites. ____ 1 2 3 4 5 6 7

 

 

70) _l_fr_e_q11_entlyhave problems staying within my__l_)u_dget. 1 2 3 4 5 6 7

 

 

71) My lludgetisnalwaystight. 1 2 3 4 5 6 7

 

72) _I_o_ftenh haye {0.51.3999 morgmoney than I have available. 1 2 3 4 5 6 7

 

73) I would buy from a web site I never heard of before. 1 2 3 4 5 6 7

 

74) I am not concerned about a potential financial loss when
buying from a web retailer I never heard of before,. 1 2 3 4 5 6 7

 

75) I don’t consider it risky to buy from a web site I have
never heard of before. 1 2 3 4 5 6 7

 

 

76) When shopping on the world-wide-web, it feels like a
chore to search for lower prices. 1 2 3 4 5 6 7

 

77) When shopping on the world-wide-web, I feel it is a
large sacrifice of @1910 search for lower prices. 1 2 3 4 5 6 7

 

 

78) When shopping on the world-wide-web, I usually don’t
lave _thgpatiencg to, search for lower prices. 1 2 3 4 5 __6_7__

 

79) When shopping on the world-wide-web, it takes too

EMEQEBQ .IQVYC_Y_PQQ?S- 1__-_2_ _3.-- A- _5__.____6__ 7

 

80) When shopping on the world-wide-web, I hate spending

time to gatherflforrnation on products. 1 _2_____3____4__5_§ 6 7__

 

81) When I shop on the world-wide-web, I find myself
pressed for time. 1 2 3 4 5 6 7

 

 

32) When I shop on the world-wide-web, I have more than
enough01119939109199_Lnx§hopping- 1 2 7__. 4 .5. ___§._._7

 

 

83) When I shop on the world-wide-web, I finish my
_shopgingfagbecause I have other things to do. ___-________-1__- 2 _3 _ 4_ 5_6 _7__

 

84) I know exactly where on the web to begin any search
for information. . _ 772“} "if.-- “5_6“ _ _7

 

 

143

Finally, we would like to ask you some information about you and your family members
for classification purposes only.

Please answer each question by circling the appropriate response.

 

85) Your sex:
' Female
0 Male
86) Your age
87) Number of years of formal education you have completed
l3:14:15:16:17:18:219
Undergraduate Graduate

88) Are you currently employed?
' No
0 Yes

89) What is your current individual annual income?
(1) (2) (3) (4) (5) (6) (7)
$0 (Not Less than $10,000 - $20,000 - $30,000 - $40,000 - $50,000+
employed) $10,000 $19,999 $29,999 $39,999 $49,999

90) What is the current income of your household (or family income)?
(1) (2) (3) (4) (5) (6)
Less than $10,000 - $20,000 - $30,000 - $40,000 - $50,000 -
$10,000 $19,999 $29,999 $3 9,999 $49,999 $59,999

(7) (8) (9) (10) (11) (12)

$60,000 - $70,000 - $80,000 - $90,000 - $100,000+ Not sure
$69,999 $79,999 $89,999 $99,999

Finish time:

 

Please do not forget to provide us with your name, telephone number, and email
address on the following page so that we may notify you if you win a prize.

144

THANK YOU VERY MUCH FOR YOUR TIME

PLEASE CHECK TO MAKE SURE YOU HAVE ANSWERED ALL OF THE QUESTIONS
COMPLETELY

Please provide us with your name, telephone number, and email so that we may notify
you if you win a prize.

Name

Telephone number
Email

 

 

 

If you are taking this survey for class credit, please provide us with the class name,
number, and section and the name of the professor so that we might notify them of your
participation

Class name

Class number

Class section

Professor’s name

 

 

 

 

Please feel free to use the space below to comment on any part or this questionnaire. Any and all
comments you may make are greatly appreciated. They will help future versions of this survey.
Thank you again.

 

 

 

 

 

 

 

 

 

145

Construct Question Key

R = Reversed
R = Question un-reversed

 

0
4t

Construct/Control

Rvsd

 

Participation qualifying question

 

Frequency of web purchases

 

Web shopping in past three months

 

Web shopping in past three months

 

When last purchase

 

Item last purchased

 

Product category of last purchase

 

Subjective knowledge of product class (last purchase)

 

Subjective knowledge of product class

 

fl
oomQONKII-§UJNs—I

Subjective knowledge of product class

 

p—a
_a

Enduring involvement with product class (last purchase)

 

y.—
N

Enduring involvement witlglroduct class

 

p—
U)

Enduring involvement with product class

 

yin—n
A

Personal purchase (last purchase)

 

_n
kit

Latest buy is a repeat purchase

 

fl
0’)

Possible reasons for repeat purchase

 

p—a
\)

Price of last purchase

 

u—s
00

Price comparison before last purchase (Y/N)

 

u—t
\0

Number of prices compared (specific search behavior)

 

N
O

Objective knowledge of intemet search

 

N
.—o

Objective knowledge of intemet search

 

N
N

Objective knowledge of intemet search

 

N
U.)

Ease of access to online information

 

N
A

Ease of access to online information

 

N
kit

Ease of access to online information

 

N
0\

Ease of access to online information

 

N
\1

Ease of access to online information

 

N
00

Ability to online price search

 

N
\0

Ability to online price search

 

t»
O

Ability to online price search

 

W
—0

Ability to online price search

 

b)
N

Ability to online price search

 

DJ
LN

Ability to online price search

 

DJ
.b

lntemet involvement

 

Lo)
LII

lntemet involvement

 

0.)
0\

Internet involvement

 

b)
\1

Internet involvement

 

DJ
00

lntemet involvement

 

 

b)
\O

 

Online price search (general tendency)

 

 

146

 

 

 

 

 

 

 

 

 

 

40 Online price search

41 Online price search

42 Online price search

43 Online price search

44 Subjective knowledge of intemet search

45 Subjective knowledge of intemet search
46 Sugective knowledge of intemet search
47 Subjective knowledge of intemet search
48 Motivation to online price search

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

49 Motivation to online price search
50 Motivation to online price search
51 Motivation to online price search
52 Motivation to online price search
53 Motivation to online price search
54 Need for cognition

55 Need for cognition

56 Need for cognition

57 Need for cognition

58 Need for cognition

59 WW shopping enthusiasm

60 WW shopping enthusiasm

61 WW shopping enthusiasm

62 Benefits of online price search

63 Benefits of online price search
64 Benefits of online price search
65 Benefits of online price search
66 Benefits of online price search
67 Perceived price dispersion

68 Perceived price dispersion

69 Perceived price disgrsion

70 Budget constraints

71 Budget constraints

72 Budget constraints

73 Trust in WWW sellers

74 Trust in WWW sellers

75 Trust in WWW sellers

76 Costs of online price search

77 Costs of online price search

78 Costs of online price search

79 Costs of online price search

80 Costs of online mice search

81 Time pressure

82 Time pressure

83 Time pressure

84 Objective knowledge of intemet search

 

147

 

 

 

85

Sex

 

86

Age

 

87

Education

 

88

Currently employed

 

89

Individual income

 

 

90

 

Household income

 

148

 

 

APPENDIX B

149

APPENDIX B

MODIFICATIONS TO PILOT RESEARCH INSTRUMENT

Ability to Online Price Search

 

Pilot Statement
Please describe your ability to search for
lower prices when shopping on the world-

wide-web:

1. Low vs. High

2. Inadequate vs. Superior

3. Weak vs. Strong

4. Incapable vs. Capable
5. Not confident in my ability vs. Confident in
my ability

6. Uncomfortable with my ability vs.
Comfortable with my ability

Finalized Statement

When it comes to searching for lower prices on
the world-wide-web, I am an expert.

Searching for lower prices on the world-wide-
web is something I can do with great case.

When it comes to searching for lower prices on
the world-wide-web, I am better at it than most
people.

When I shop on the world-wide-web, I have
superior techniques for locating lower prices.

1 am extremely effective at locating lower prices
when I search for them on the world-wide-web.

1 am very confident in my ability to search for
lower prices on the world-wide-web.

 

150

Benefits of Online Price Search

 

Pilot Statement

Finalized Statement

 

1. By searching around on the world-wide-web
for price information, 1 am certain of making
the best buy.

2. It doesn’t pay to shop around before buying
on the world-wide-web. ( R )

3. Shopping around at more than one web site
helps me to find the lowest price.

4. When shopping on the world-wide-web, there
is too much to lose by being ignorant about
prices.

5. By rushing into a purchase when shopping on
the world-wide-web, one is bound to miss a
good deal.

When shopping on the world-wide-web, I am
certain of making the best buy by searching
around for price information.

It pays to shop around before buying on the
world-wide-web.

Not modified

Not modified

When shopping on the world-wide-web, one is
bound to miss a good deal by rushing into a
purchase.

Before buying on the world-wide-web, a person
can save a lot of money if they compare prices
from different websites.

Before buying on the world-wide-web, there are
definite benefits to comparing the prices at
different websites.

 

151

Costs of Online Price Search

 

Pilot Statement

F inalized Statement

 

1. When shopping on the world-wide-web, it
feels like a chore to search for lower prices.

2. When shopping on the world-wide-web, I feel
it is a large sacrifice of time to search for
lower prices.

3. When shopping on the world-wide-web, 1
usually don’t have the patience to search for
lower prices.

4. When shopping on the world-wide-web, it
takes too much effort to find lower prices.

5. When shopping on the world-wide-web, I hate
spending time to gather information on
products.

Not modified

Not modified

Not modified

Not modified

Not modified

 

Ease of Access to Online Information

 

Pilot Statement

Please think about the location from where
you do most of your shopping on the world-
wide-web when answering the following
question.

The speed at which web pages load from this
location is:

l. Intolerable vs. Tolerable

2. Unsatisfactory vs. Satisfactory

3. Slow vs. Fast

Finalized Statement

 

The speed at which web pages load on the
computer where I do most of my web shopping
is tolerable.

I am completely satisfied with the speed at
which web pages load on the computer where 1
do most of my web shopping.

Web pages load quickly on the computer where 1
do most of my web shopping.

 

152

 

lntemet Involvement

 

Pilot Statement
To me the world-wide-web is:

l. Unimportant vs. Important

2. Mean nothing to me vs. Means a lot to me

3. Irrelevant vs. Relevant

4. Worthless vs. Valuable

5. Not needed vs. Needed

Finalized Statement

The world-wide-web plays an extremely
important role ill my life.

It means a lot to me to have daily access to the
world-wide-web.

The world-wide-web is a very valuable tool in
my life.

The world-wide-web is a relevant part of my
daily life.

I need the world-wide-web.

The day isn’t complete unless I have used the
world-wide-web.

Daily access to the world-wide-web is
something I could not live without.

I feel like I use the world-wide-web for
everything.

 

153

Motivation to Online Search

 

Pilot Statement

F inalized Statement

 

1. It is important for me to get the best price
when shopping on the world-wide-web.

2. It really doesn’t occur to me to search for
lower prices when shopping on the world-
wide-web. ( R )

3. I am motivated to search for lower prices
when shopping on the world-wide-web.

4. I am enthusiastic about searching for lower
prices when shopping on the world-wide-web.

5. I feel driven to find the best price, when
shopping on the world-wide-web.

6. I really want to find the best prices, when
shopping on the world-wide-web.

When I shop on the world-wide-web, it is
important for me to get the best price.

When I shop on the world-wide-web, it really
doesn’t occur to me to search for lower prices.

(R)

When I shop on the world-wide-web, I am
motivated to search for lower prices.

When I shop on the world-wide—web, I am
enthusiastic about searching for lower prices.

When I shop on the world-wide-web, I feel
driven to find the best price.

When I shop on the world-wide-web, I really
want to find the best prices.

 

Need for Cognition

 

Pilot Statement

Finalized Statement

 

l. The notion of thinking abstractly is appealing
to me.

2. Thinking is not my idea of fun. ( R )

3. I prefer my life to be filled with puzzles that I
must solve.

4. 1 only think as hard as I have to. ( R )

5. Learning new ways to think doesn’t excite me
too much. ( R )

Not modified

Not modified

Not modified

Not modified

Not modified

1 would rather do something that requires little
thought than something that is sure to challenge
my thinking. ( R )

It’s enough for me that something gets the job
done: I don’t care how or why it works. ( R )

 

154

Online Price Search

 

Pilot Statement

1. How many prices did you compare on the web
before making your last purchase on the
world-wide-web?

2. When shopping on the world-wide—web,
“Never buy the first one you look at” is a good
motto.

3. When shopping on the world-wide-web, 1 am
willing to go to extra effort to find lower
prices.

4. I make it a rule to visit more than one web
retailer or use a shop-bot to compare prices
before I buy on the world-wide-web.

5. I would never shop at more than one web site
just to find low prices before I buy on the
world-wide-web. ( R )

6. When shopping on the world-wide-web, I
often compare the prices at more than one web
retailer before deciding where to buy from.

F inalized Statement

Not modified

Not modified

Not modified

Before I buy on the world-wide-web, I make it a
rule to compare the prices of different web sites.

I would not buy anything on the world-wide-web
without first comparing the price at another web
site.

When I shop on the world-wide-web, I often
compare the prices between web site before
deciding whom to buy from.

 

Perceived Budget Constraints

 

Pilot Statement

 

l. I frequently have problems staying within my
budget.

2. My budget is always tight.

3. I often have to spend more money than I have
available.

F inalized Statement

 

I frequently have problems only spending what 1
am suppose to for the week.

My spending budget is usually tight.

I often feel I have to spend more money than I
have available.

I don’t usually feel like 1 have enough money to
spend during the week.

 

155

Perceived Price Dispersion

 

Pilot Statement

Some web sites have prices that are much
lower than others.

2. Prices of individual items may vary between
web sites, but overall, there isn’t much
difference in the prices between web sites.

(R)

The price of individual items ofien varies a lot
between web sites.

Finalized Statement

No matter what 1 am shopping for on the world-
wide-web, I believe that some web site is selling
it at a much lower price than others.

When shopping on the world-wide-web, one
should expect the price of an item to really differ
between web sites.

For almost any item one can purchase on the
world-wide-web, the price often varies a lot
between web sites.

 

Perceived Time Pressure

 

Pilot Statement

1. When I shop on the world-wide-web, I find
myself pressed for time.

2. When I shop on the world-wide-web, l have
more than enough time to complete my
shopping. ( R )

3. When I shop on the world-wide-web, I finish
my shopping fast because I have other things
to do.

Finalized Statement

Not modified

Not modified

Not modified

When I shop on the world-wide-web, I am in a
hurry.

When I shop on the world-wide-web, I only have
a limited amount of time to finish my shopping.

 

156

 

Subiective Knowledge of lntemet Search

 

Pilot Statement

F inalized Statement

 

l. I feel very knowledgeable about searching for
information on the world-wide-web,

2. Among my circle of friends, I’m one of the
“experts” on searching for information on the
world-wide-web.

3. I know how to search for information on the
world-wide—web.

4. 1 know more about searching for information
on the world-wide-web than most people,

Not modified

Not modified

Search for information on the world-wide-web is
something I know how to do really well.

Not modified

 

Trust in WWW Sellers

 

Pilot Statement

F inalized Statement

 

l. I would buy from a web site I never heard of
before.

2. I am not concerned about a potential financial
loss when buying from a web retailer I never
heard of before,.

3. I don’t consider it risky to buy from a web site

I have never heard of before.

If I have never heard of a web site, 1 will not buy
from it. ( R)

When buying from a web site I have never heard
of before, I am concerned about a potential
financial loss. ( R)

I consider it risky to buy from a web site I have
never heard of before. ( R)

When buying on the world-wise-web, 1 need to
have heard of the web site before 1 purchase
anything. ( R )

 

157

"5’
n

 

WWW Shopping Enthusiasm

 

Pilot Statement F inalized Statement

 

 

1. Shopping on the world-wide-web feels like an
escape.

2. Compared to other things I could do, the time
spent shopping on the world-wide-web is truly
enjoyable.

3. I enjoy shopping on the world-wide-wcb for
its own sake, not just for the items 1 may
purchase.

4. While I shop on the world-wide-web, I am able
to forget my problems.

 

 

 

158

APPENDIX C

159

i-""‘_-"’I

 

 

APPENDIX C

RESEARCH INSTRUMENT

CONSENT FORM
Thank you for your voluntary participation to complete this questionnaire.

To participate in this study you must:
1. Be an undergraduate or graduate student at Michigan State University
2. Be over 18 years old
3. Have purchased at least one item on the world-wide-web in the past three months

The researcher of this study, which is titled A Model of Consumer Extemflrice Sgrch Behavior in an
Electronic Marketplace (World-Wide-Web). is conducting this survey in order to understand what factors
influence a consumer’s search for lower prices when shopping on the world-wide-web.

Voluntary completion of this questionnaire will take approximately a half an hour. If at any time you feel
unable or unwilling to complete the questionnaire, please feel free to stop.

All qualified participants are automatically entered into a drawing for a $100 cash prize. The winner will
be notified by phone. Please remember to include your name, phone number, and email address at the end
of the questionnaire for notification purposes. The prize will be drawn upon full completion of the data.

In certain cases, for your participation, you may also receive extra credit points as determined by
your instructor. In such cases, upon completion your name will be submitted to the instructor of
your course to receive extra credit points.

Other than in those cases where your name will be submitted to the instructor of your course to receive
extra credit points, your name, phone number, and e-mail address will not be distributed. All information
regarding participants will remain confidential. Any information that you provide and the data collected
from the questionnaire will be kept in the strictest confidentiality. Your privacy will be protected to the
maximum extent allowable by law.

In addition, we will not identify any individual who participates in this survey by name, phone number or
email in any report or publication. You are also given the option to decline to answer any question at any
time. If you have any questions or concerns about the study you may contact James Ramos at
ramosiam@msu.edu or (517) 372-8095, or Dr. Charles Salmon, Senior Associate Dean, College of
Communication at salmonfizmsuedu or (517) 355-3410.

If you have any questions or concerns regarding your rights as a study participant, or are dissatisfied at any
time with any aspect of this study, you may contact - anonymously, if you wish - Dr. Ashir Kumar, Chair
of the University Committee on Research Involving Human Subjects (UCRIHS) by phone: (517) 355-2180,
fax: (517) 353-2976, e-mail: ucrihs@msu.edu, or regular mail: 202 Olds Hall, East Lansing, MI 48824.

By signing and dating this page, you indicate your voluntary agreement to participate in this experiment.

Signiture Date

 

 

Print Name

 

160

 

 

,1...

The follow questions are related to shop on the world-wide-web and the prices you paid
for items you purchased. When asked about price, this means the total cost of an item,
including for example, any tax or shipping and handling fees. There are no right or
wrong answers to the following questions and people answers’ vary widely. Kindly
indicate m personal opinion by circling any one number.

Thank you.

Start time:

 

1) Have you made at least one purchase on the world-wide-web in the past three months,
not including any purchases from auction websites like eBay?

(1) (2)
Yes No

2) How often do you purchase items on the world-wide-web (not including any purchases
fiom auction websites like eBay)?

(1) (2) (3) (4) (5) (6) (7)
Once a Once Once Once Once a Once Once a
year or every six every every two month every two week or

less months three months to three more

months weeks

3) How many individual items have you purchased on the world-wide-web in the past
three months (not including any purchases from auction websites like eBay)?

(1) (2) (3) (4) (5) (6) (7)
1-2 items 3-5 items 6-10 items 11-15 15-19 20-25 25+ items
items items items

4) How much have you spent buying items on the world-wide-web in the past three
months (not including any purchases from auction websites like eBay)?

(1) (2) (3) (4) (5) (6) (7)
$1 - $49 $50 - $99 $100 - $200 - $500 - $1,000 - $2,500+
$199 $499 $999 $2,500

161

We would like you to take a minute and think back to the last time you bought something
on the world-wide-web (not including any purchases from auction websites like eBay).
Thinking back on that particular world-wide-web shopping experience, please respond to
the following questions. Again, there are no right or wrong answers. Kindly indicate
your personal opinion by circling any one answer. Thank you.

5) When was the last time you purchased an item on the world-wide-web (not including
any purchases from auction websites like eBay)?

( 1) (2) (3) (4) (5)
Within the Within the Within the Within the Within the
past two past four past six past eight past three
weeks weeks weeks weeks months

6) What exactly did you purchase?
' Item:
' Prefer not to specify
° Cannot remember the exact item (Go to Question 14)

 

7) Under what category would you classify your last purchase?

Travel (Airline Tickets)

Travel (Not including airline tickets)

Books

Movies (DVD’S, VHS, etc.)

Music (CD’S, Tapes, etc.)

Car accessories

Clothing and accessories

Computer hardware

Computer software

Consumer electronics

Electronic games

Event tickets (for music concerts, sporting events, or cultural events)
Flowers

Health and beauty

Home and garden

Office related products

Sporting goods

Other, prefer not to specify

Other, please Specify

 

162

 

How much do you know about the category you indicated in Question 7?

 

 

 

 

 

Strongly Strongly
Disagree Agree
8) I know more about the category indicated in Question 7
than most people. 1 2 3 4 5 6 7
9) Among my circle of friends, I’m one of the “experts” on
“the categogindicated in Question 7. 1 2 3 4 5 6 7
10) I feel very knowledgeable about the categonj indicated
in Question7. 1 2 3 4 5 6 7

 

Please consider the category indicated in Question 7. To you this category of products is:
1 l) Unimportant 1 2 3 4 5 6 7 Important
12) Means 1 2 3 4 5 6 7 Means a lot to
nothing me
13) Irrelevant 1 2 3 4 5 6 7 Relevant

14) Did you purchase this item for yourself or for another person?
(1) (2) (3)
Yourself Another person or group You and
another person or group

15) Please think about the web retailer from which you made your last web purchase.
Before making your last purchase from this web retailer, had you ever bought something
from this website?

(1) (2) (3)
Yes No Do not recall
(Go to Question 17) (Go to Question 17)

16) What reasons do you believe influenced your decision to buy from this web retailer
again?
(Please circle a_ll that apply)

(1) (2) (3) (4)
Satisfaction Felt it was Did not want to Knew they sold
with previous a fair price shop around what you
transaction(s) wanted to buy

163

17) Approximately how much did you pay for your last purchase on the world-wide-
web?

(1) (2) (3) (4) (5) (6) (7)
Less than $25 - $49 $50 - $99 $100 - $200 - $500 - $1,000+
$25 $199 $499 $999

18) Before making your last purchase on the web, did you compare the price of that item
at different web Sites? (Comparing prices could involve visiting more than one web
retailer or using a shop-bot website, like mysimon.com)

(1) (2) (3)
Yes — I compared prices No — I did not compared Do not recall
before buying prices before buying (Go to Question 20)
(Go to Question 20)

19) How many prices did you compare on the web before making your last purchase on

the world-wide-web?
2:3:4:5:6:7:8+

20) Do you own a computer with access to the world-wide—web?

(1) (2)
Yes No
21) How often do you access the world-wide-web?
Never 1 2 3 4 5 6 7 Everyday
22) How often do you access the world-wide-web to search for information?
Never 1 2 3 4 5 6 7 Everyday
23) From what location do you access the world-wide-web the most?
(1) (2) (3) (4) (5) (6)
Your home Your Friend’s University Public Other
workplace home computer library
lab or
library
24) From what location do you do most of your shopping on the world-wide-web?
(1) (2) (3) (4) (5) (6)
Your home Your Friend’s University Public Other
workplace home computer library
lab or
library

164

The statements that follow concern activities, interests, and opinions related to shopping
on the world-wide-web. There are no right or wrong answers to the following statements
and large numbers of people agree and disagree. Kindly indicate yo_ur personal opinion
by circling any one number for each statement from 1 (Strongly Disagree) to 7 (Strongly

Agree).
Strongly Strongly
Disagree Agree
2 5) When I shop on the world-wide-web, I hate spending
time to gather information on products. 1 2 3 4 5 6 7
26) The world-wide-web plays an extremely important role
inmylife. 1 2 3 4 5 6 7
27) When I Shop on the world-wide-web, it is important for
me to get the best price. 1 2 3 4 5 6 7
28)
The notion of thinking abstractly is appealing to me. 1 2 3 4 5 6 7
29) When it comes to searching for lower prices on the
world-wide-web, I am an expert. I 2 3 4 5 6 7
30)
_Shopping on the world-wide-web feels like an escape. 1 2 3 4 5 6 7
31) When Shopping on the world-wide-web, “Never buy the
_first one you look at” is a good motto to follow. 1 2 3 4 5 6 7
No matter what I am shopping for on the world-wide-
32) web, I believe that some web site is selling it at a much
_lower price than others. 1 2 3 4 5 6 7
When Shopping on the world-wide-web, I am certain of
33) making the best buy by searching around for price
information. _1_ #2 3 4 5 6 7
34) I frequently have problems only spending what I am
_suppose to for the week. _1_- 2 3 4 5 6 7_
35) I feel very knowledgeable about searching for
”information on the world-wide-web. l 2 3 4 5 6 7_
36) If I have never heard of a web site, I will not buy from
it. 1 __ 2 3 4 5 6 7
37) When Shopping on the world-wide-web, it feels like a
__chore to search for lower prices. _____ _ _ ___ “122 3 4 ___5__ 6 7_
38)
-..Th1111€1118._1_31191_£nl15193.9111111:- 1. 2 _ 3 4 5.. 6 7

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

165

Strongly Strongly

 

 

Disagee Agree
39) Shopping around at more than one web site helps me to
find the lowest price. 1 2 3 4 5 6 7
40) Compared to other things I could do, the time spent
Shopping on the world-wide-web is truly enjoyable. I 2 3 4 5 6 7

 

41) Among my circle of friends, I’m one of the “experts” on
searching for information on the world-wide-web. 1 2 3 4 5 6 7
When shopping on the world-wide-web, one should

42) expect the price of an item to really differ between web

 

 

sites. 1 2 3 4 5 6 7
43) When I shop on the world-wide-web, I am willing to go

to extra effort to find lower prices. 1 2 3 4 5 6 7
44)

My spending budget is usually tight. 1 2 3 4 5 6 7

 

45) When I shop on the world-wide-web, it really doesn’t
occur to me to search for lower prices. 1 2 3 4 5 6 7

 

46) When buying from a web site I have never heard of
before, I am concerned about a potential financial loss. 1 2 3 4 5 6 7

 

47) Searching for lower prices on the world-wide-web is
something I can do with great ease. I 2 3 4 5 6 7

48) When I shop on the world-wide-web, I have more than
enough time to complete my shopping. 1 2_ 2 4 5 6 7

 

 

49) When I Shop on the world-wide-web, I usually don’t
have the patience to search for lower prices. 1 2 3 4 5 6 7

 

50) I enjoy shopping on the world-wide-web for it’s own
sake, not just for the items I may purchase. 1 2 3 4 5 6 7

51) When it comes to searching for lower prices on the
world-wide-web, I am better at it than most people. 1 2 3 4 5 6 7

5 2) For almost any item one can purchase on the world-
wide-web, the price often varies a lot between web sites. 1 2 3 4 5 6 7
Before buying on the world-wide-web, a person can

53) save a lot of money if they compare prices from
different websites.

 

 

54) I often feel I have to spend more money than I have
available. “1 2 3__ 4___5______6 7

 

 

166

55)

56)

57)

58)

59)

60)

61)

62)

63)

64)

65)

66)

67)

68)

69)

70)

Strongly Strongly
Disagree Agree

 

Before I buy on the world-wide-web, I make it a rule to
compare the prices of different web sites.

I consider it risky to buy from a web site I have never
heard of before.

Search for information on the world-wide-web is
something I know how to do really well.

When I shop on the world-wide-web, I finish my
shopping fast because I have other things to do.

When I shop on the world-wide-web, I am motivated to
search for lower prices.

Web pages load quickly on the computer where I do
most of my web shopping.

When shopping on the world-wide-web, it’s takes too
much effort to find lower prices.

I don’t usually feel like I have enough money to spend
during the week.

I know more about searching for information on the
world-wide-web than most people.

When buying on the world-wide-web, I need to have
heard of the web site before I purchase anything.

When I shop on the world-wide-web, I am enthusiastic
about searching for lower prices.

When I shop on the world-wide-web, I am in a hurry.
When I shop on the world-wide-web, I often compare
the prices between web Site before deciding whom to
buy from.

The world-wide-web is a relevant part of my daily life.

When I shop on the world-wide-web, I have superior
techniques for locating lower mices.

 

1 only think_as_ha:d_ a§___1_,h_9._v_9_t92_

 

167

 

71)

72)

73)

74)

75)

76)

77)

78)

79)

80)

81)

82)

83)

84)

85)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Strongly Strongly
Disagree Agree
When Shopping on the world-wide-web, there is too
much to lose by being ignorant about prices. 1 2 4 5 6 7
While I Shop on the world-wide-web, I am able to forget
my problems. 1 2 4 5 6 7
When I Shop on the world—wide-web, I feel it is a large
sacrifice of time to search for lower prices. 1 2 4 5 6 7
When I shop on the world-wide-web, I only have a
limited amount of time to finish my Shopping. 1 2 4 5 6 7
I am extremely effective at locating lower prices when I
search for them on the world-wide-web. 1 2 4 5 6 7
I need the world-wide-web. l 2 4 5 6 7
It pays to shop around before buying on the world-wide-
web. 1 2 4 5 6 7
Learning new ways to think doesn’t excite me too
much. 1 2 4 5 6 7
I would not buy anything on the world-wide-web
without first comparing the price at anothe_r_v_vepsite. 1 2 4 5 6 7
The speed at which web pages load on the computer
where I do most of my web shopping is tolerable. l 2 4 5 6 7
Before buying on the world-wide-web, there are definite
benefits 39 hcyoprparing the prices at different websites. 1 2 4 5____ 6 7
It means a lot to me to have daily access to the world-
wide-web. l 2 4 __5 6 7
When I Shop on the world-wide-web, I feel driven to
find the best price. 1 2 4 __5___ 6 7
I prefer my life to be filled with puzzles that I must
_solve. -__-_ _____ l 2 4 5 6 7
___I..f9¢_1_1_i_159 1.11.89 the wgtl.d:wjésr.wehf9r.everything--4 1 2 4 5 __ 6 7

 

 

168

86)

87)

88)

89)

90)

91)

92)

93)

94)

95)

 

 

 

 

 

 

 

 

 

 

 

Strongly Strongly
_ fl _ Disagree Agree
The world-wide-web is a very valuable tool in my life. 1 2 5 6 7
I am completely satisfied with the speed at which web
pages load on the computer where I do most of my web
shopping. 1 2 5 6 7
It’s enough for me that something gets the job done: I
don’t care how or why it works. 1 2 5 6 7
I am very confident in my ability to search for lower
prices on the world-wide-web. 1 2 5 6 7
Daily access to the world-wide-web is something I
could not live without. 1 2 5 6 7
When shopping on the world-wide-web, one is bound to
miss a good deal by rushing into a purchase. 1 2 5 6 7
I would rather do something that requires little thought
than somethinglhat is sure to challenge my thinking. 1 2 5 6 7
The day is not complete unless I have used the world-
wide-web. l 2 5 6 7
When I shop on the world-wide-web, I find myself
pressed for time. 1 2 5 6 7
When I shop on the world-wide-web, I really want to
find the best prices. 1 2 5 6 7

 

Please do not forget to provide us with your name, telephone
number, and email address on the following page so that we
may notify you if you win a prize.

169

Finally, we would like to ask you some information about you and your family members
for classification purposes only.

Please answer each question by circling the appropriate response.

 

96) Your sex:
0 Female
' Male
97) Your age
98) Number of years of formal education you have completed
13:14:15216 : 17:18:219
Undergraduate Graduate

99) Are you currently employed?
' No
° Yes

100) What is your current individual annual income?
(1) (2) (3) (4) (5) (6) (7)
$0 (Not Less than $10,000 - $20,000 - $30,000 - $40,000 - $50,000+
employed) $10,000 $19,999 $29,999 $39,999 $49,999

101) What is the current income of your household (or family income)?
(1) (2) (3) (4) (5) (6)
Less than $10,000 - $20,000 - $30,000 - $40,000 - $50,000 -
$10,000 $ 19,999 $29,999 $39,999 $49,999 $59,999

(7) (8) (9) (10) (11) (12)

$60,000 - $70,000 - $80,000 - $90,000 - $100,000+ Not sure
$69,999 $79,999 $89,999 $99,999

Finish time:

 

Please do not forget to provide us with your name, telephone
number, and email address on the following page so that we
may notify you if you win a prize.

170

THANK YOU VERY MUCH FOR YOUR TIME

PLEASE CHECK TO MAKE SURE YOU HAVE ANSWERED ALL OF THE QUESTIONS
COMPLETELY

Please provide us with your name, telephone number, and email so that we may notify
you if you win a prize.

Name

Telephone number
Email

 

 

 

If you are taking this survey for class credit, please provide us with the class name,
number, and section and the name of the professor so that we might notify them of your
participation

Class name

Class number

Class section

Professor’s name

 

 

 

 

Please feel free to use the space below to comment on any part or this questionnaire. Any and all
comments you may make are greatly appreciated. They will help future versions of this survey.
Thank you again.

 

 

 

 

 

 

 

 

 

171

Construcmuestion Key

R = Reversed

 

Q#

Construct/Control (Item Number)

Rvsd

 

Participation qualifying question

 

Frequency of web purchases

 

Web shopping in past three months

 

Web shopping in past three months

 

When last purchase

 

Item last purchased

 

Product category of last purchase

 

Subjective knowledge of product class (last purchase)

 

cmflam-‘KWNu—t

Subjective knowledge of product class

 

p—l
C

Subjective knowledge of product class

 

p—n
p—a

Enduring involvement with product class (last purchase)

 

_a
N

Enduring involvement with product class

 

p—A
U.)

Enduring involvement with product class

 

p—a
&

Personal purchase (last purchase)

 

p—s
kl!

Latest buy is a repeat purchase

 

p—A
0\

Possible reasons for repeat purchase

 

p—n
\1

Price of last purchase

 

u—s
00

Price comparison before last purchase (Y/N)

 

p—n
\0

Online price search (search behavior) (OPSl)

 

N
O

Objective knowledge of intemet search

 

N
—

Objective knowledge of intemet search

 

N
N

Objective knowledge of intemet search

 

N
U)

Ease of access to online information (priming question)

 

24

Ease of access to online information (priming
questiony

 

25

Online price search (general tendency) (OPSZ)

 

26

lntemet involvement (IIVl)

 

27

Ability to online price search (AOPSI)

 

28

Need for cognition (NFC 1)

 

29

Subjective knowledge of intemet search (SKISl)

 

30

WW Shopping enthusiasm (WSEIL

 

31

Motivation to online price search (MOPS l)

 

32

Perceived price dispersion (PPDl)

 

33

Benefits of online price search (BOPSl)

 

34

Perceived budget constraints (PBCl)

 

35

Costs of online price search (COPSl)

 

36

Trust in WWW sellers (TWS l)

 

37

Online mce search (general tendency) (OPS3)

 

38

Need for cognition (N FC2)

 

 

39

 

Subjective knowledge of intemet search (SKISZ)

 

 

172

 

 

40

WW shopping enthusiasm (WSEZ)

 

41

Ability to online price search (AOPS2)

 

42

Perceived price dispersion (PPD2)

 

43

Benefits of online price search (BOPS2)

 

Perceived budget constraints (PBC2)

 

45

Motivation to online price search (MOPS2)

 

46

Trust in w sellers (Twsz)

 

47

Costs of online price search (COPSZ)

 

48

Perceived time pressure (PTP2)

 

49

Online price search (general tendency) (OPS4)

 

50

WW shopping enthusiasm (WSE3)

 

51

Benefits of online price search (BOPS3)

 

52

Perceived price dispersion (PPD3)

 

53

Motivation to online price search (MOPS3)

 

54

Perceived budget constraints (PBC3)

 

55

Ability to online price search (AOPS3)

 

56

Trust in W sellers (TWS3)

 

57

Subjective knowledge of intemet search (SKIS3)

 

58

Perceived time pressure (PTP3)

 

59

Costs of online price search (COPS3)

 

60

Ease of access to online information (EAOI3)

 

61

Online price search (general tendency) (OPSS)

 

62

Perceived budget constraints (PBC4)

 

63

Motivation to online price search (MOPS4)

 

64

Trust in WWW sellers (TWSID

 

65

Benefits of online price search (BOPS4)

 

66

Perceived time pressure (PTP4)

 

67

Subjective knowledge of intemet search (SKIS4)

 

68

lntemet involvement (IIV4)

 

69

Ability to online price search (AOPS4)

 

70

Need for cognition (N F C4)

 

71

Costs of online price search (COPS4)

 

72

WW shopping enthusiasm (WSE4)

 

73

Online price search (general tendency) (OPS6)

 

74

Perceived time pressure (PTPS)

 

75

Benefits of online price search (BOPSS)

 

76

lntemet involvement (IIVS)

 

77

Motivation to online price search (MOPSS)

 

78

Need for cggnition (N FCS)

 

79

Ability to online price search (AOPSS)

 

80

Ease of access to online information (EAOI 1)

 

81

Benefits of online price search (BOPS7)

 

82

lntemet involvement (IIV2)

 

83

Costs of online price search LCOPSS)

 

 

84

 

Need for cognition (NFC3)

 

173

 

 

 

85

Benefits of online price search (BOPS6)

 

86

lntemet involvement (IIV3)

 

87

Ability to online price search (AOPS6)

 

88

Need for cognition (N FC7)

 

89

Motivation to online price search (MOPS6)

 

90

lntemet involvement (IIV7)

 

91

Ease of access to online information (EAOIZ)

 

92

Need for cognition N C6)

 

93

lntemet involvement (HV 6)

 

94

Perceived time pressure (PTPl)

 

95

lntemet involvement (IIV8)

 

96

Sex

 

97

Age

 

98

Education

 

99

Currently employed

 

100

Individual income

 

 

101

 

Household income

 

174

 

 

REFERENCES

Alba, J., Lynch, J ., Weitz, B., Janiszewski, C., Lutz, R., Sawyer, A., et al. (1997).
Interactive home shopping: Consumer, retailer, and manufacturer incentives to
participate in electronic marketplaces. Journal of Marketing, 61(3), 38-53.

Alba, J. W., & Hutchinson, J. W. (1987). Dimensions of consumer expertise. Journal of
Consumer Research, 13(4), 41 1-454.

Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A
review and recommended two-step approach. Psychological Bulletin, 103(3),
411-423.

Andreasen, A. (1965). Attitudes and customer behavior: A decision model. In L. Preston
(Ed.), New Research in Marketing. Berkley, CA: California Institute of Business
and Economics Research, University of California.

Babin, B. J ., Darden, W. R., & Griffin, M. (1994). Work and/or fun: Measuring hedonic
and utilitarian shopping value. Journal of Consumer Research, 20(4), 644-656.

Bakos, J. Y. (1991). A strategic analysis of electronic marketplaces. MIS Quarterly,
15(3), 295-310.

Bakos, J. Y. (1997). Reducing buyer search costs: Implications for electronic
marketplaces. Management Science, 43(12), 1676-1692.

Barki, H., & Hartwick, J. (1989). Rethinking the concept of user involvement. MIS
Quarterly, 13(1), 53-62.

Bauer, R. A. (1967). Consumer behavior as risk taking. In D. F. Cox (Ed.), Risk taking
and information handling in consumer behavior (pp. 23-33). Boston: Division of
Research, Graduate School of Business Administration, Harvard University.

Beatty, S. E., Homer, P., & Kahle, L. R. (1988). The involvement commitment model:
Theory and implications. Journal of Business Research, 16(2), 149-167.

Beatty, S. E., & Smith, S. M. (1987). External search effort: An investigation across
several product categories. Journal of Consumer Research, 14(1), 83-95.

Beme, C., Mugica, J. M., Pedraja, M., & Rivera, P. (2001). Factors involved in price
information-seeking behavior. Journal of Retailing and Consumer Services, 8(2),
71-84.

Bettman, J. R. (1979). An information processing theory of consumer choice. Reading,
MA: Addison-Wesley.

175

Bettman, J. R., & Park, C. W. (1980). Effects of prior knowledge and experience and
phase of the choice process on consumer decision-processes: A protocol analysis.
Journal of Consumer Research, 7(3), 234-248.

Blodgett, J. G., Hill, D. J ., & Stone, G. (1995). A model of the determinants of retail
search. In F. R Kardes & M. Sujan (Eds.), Advances in consumer research (V 01.
22, pp. 518-523). Provo, UT: Association for Consumer Research.

Brucks, M. (1985). The effects of product class knowledge on information search
behavior. Journal of Consumer Research, 12(1), 1-16.

Bucklin, L. P. (1966). Testing propensities to shop. Journal of Marketing, 30, 22-27.

Butters, G. R. (1977). Equilibrium distributions of sales and advertising prices. Review of
Economic Studies, 44(3), 465-491.

Cacioppo, J. T., & Petty, R. E. (1982). The need for cognition. Journal of Personality and
Social Psychology, 42(1), 116-131.

Cacioppo, J. T., Petty, R. E., & Kao, C. F. (1984). The efficient assessment of need for
cognition. Journal of Personality Assessment, 48(3), 306-307.

Carlson, J. A., & Gieseke, R. J. (1983). Price search in a product market. Journal of
Consumer Research, 9(4), 357-365.

Chandler, D. (2001). Technological or media determinism. Retrieved February 6, 2001,
from http://www.aber.ac.uk/media/Documents/tecdet/tecdet.html

Claxton, J. D., Fry, J. N., & Portis, B. A. (1974). A taxonomy of prepurchase information
gathering patterns. Journal of Consumer Research, [(3), 35-42.

Culnan, M. J. (1983). Environmental scanning: The effects of task complexity and source
accessibility on information gathering behavior. Decision Sciences, 14(2), 194-
206.

Darke, P. R., & Freedman, J. L. (1995). Nonfinancial motives and bargain hunting.
Journal of Applied Social Psychologi, 25(18), 1597-1610.

Deighton, J. (1997). Exploring the implications of the lntemet for consumer marketing:
Commentary. Journal of the Academy of Marketing Science, 25(4), 347-351.

Dommermuth, W. P. (1965). The shopping matrix and marketing strategy. Journal of
Marketing Research, 2, 128-132.

176

Dommermuth, W. P., & Cundiff, E. W. (1967). Shopping goods, shopping centers and
selling strategies. Journal of Marketing, 31, 32-36.

Doney, P. M., & Cannon, J. P. (1997). An examination of the nature of trust in buyer-
seller relationships. Journal of Marketing, 61(2), 35-51.

Donthu, N., & Gilliland, D. (1996). The infomercial shopper. Journal of Advertising
Research, 36(2), 69-76.

Duncan, C. P., & Olshavsky, R. W. (1982). External search: The role of consumer
beliefs. Journal of Marketing Research, 19(1), 32-43.

Engel, J. F ., Blackwell, R. D., & Kollat, D. T. (1978). Consumer behavior (3d ed.).
Hinsdale, IL: Dryden Press.

Engel, J. F ., Kollat, D. T., & Blackwell, R. D. (1968). Consumer behavior. New York:
Holt Rinehart and Winston.

Engel, J. F ., Kollat, D. T., & Blackwell, R. D. (1973). Consumer behavior (2d ed.). New
York: Holt Rinehart and Winston.

Fiske, C. A., Luebbehusen, L. A., Miyazaki, A. D., & Urbany, J. E. (1994). The
Relationship between knowledge and search: It depends. In K. P. Corfman & J. G.
Lynch (Eds), Advances in consumer research (V 01. 21, pp. 43-50). Provo, UT:

Association for Consumer Reseach.

Flynn, L. R., & Goldsmith, R. E. (1999). A short, reliable measure of subjective
knowledge. Journal of Business Research, 46(1), 57-66.

Geyskens, I., Steenkamp, E. M., & Kumar, N. (1996). The effects of trust and
interdependence on relationship commitment: A trans-atlantic study. International
Journal of Research in Marketing, 13(4), 303-317.

Goldman, A. (1977). Consumer knowledge of food prices as an indicator of shopping
effectiveness. Journal of Marketing, 41(4), 67-75.

Goldman, A., & Johansson, J. K. (1978). Determinants of search for lower prices:
Empirical assessment of economics of information theory. Journal of Consumer

Research, 5(3), 176-186.

Guo, C. Q. (2001). A review on consumer external search: Amount and determinants.
Journal of Business and Psychology, 15(3), 505-519.

GVU. (1998). WWW User Survey (10th wave), 1998]. from www.gvu.gatech.edu/user—
survey

177

Hawes, J. M., & Lumpkin, J. R. (1984). Understanding the outshopper. Journal of the
Academy of Marketing Science, 12(4), 200-218.

Hodkinson, C., Kiel, G., & McColl-Kennedy, J. R. (2000). Consumer web search
behaviour: Diagrammatic illustration of wayfinding on the web. International
Journal of Human—Computer Studies, 52(5), 805-830.

Hoffman, D. L., & Novak, T. P. (1996). Marketing in hypermedia computer-mediated
environments: Conceptual foundations. Journal of Marketing, 60(3), 50-68.

Holbrook, M. B. (1986). Aims, concepts, and methods for the representation of individual
differences in esthetic responses to design features. Journal of Consumer
Research, 13(3), 337-347.

Hoque, A. Y., & Lohse, G. L. (1999). An information search cost perspective for
designing interfaces for electronic commerce. Journal of Marketing Research,
36(3), 387-394.

Howard, J. A. (1977). Consumer behavior: Application of theory. New York: McGraw-
Hill.

Howard, J. A., & Sheth, J. N. (1969). The theory of buyer behavior. New York: Wiley.

Inman, J. J ., McAlister, L., & Hoyer, W. D. (1990). Promotion signal: Proxy for a price
cut. Journal of Consumer Research, 17(1), 74-81.

James, L. R., Mulaik, S. A., & Brett, J. M. (1982). Causal analysis: Assumptions, models,
and data. Beverly Hills, CA: Lawrence Erlbaum Associates.

Johnson, B. T., & Eagly, A. H. (1990). Involvement and persuasion: Types, traditions,
and the evidence. Psychological Bulletin, 107(3), 375-384.

Johnson, E. J ., & Russo, J. E. (1984). Product familiarity and learning new information.
Journal of Consumer Research, 11(1), 542-550.

Jones, J. M., & Vijayasarathy, L. R. (1998). lntemet consumer catalog shopping:
Findings from an exploratory study and directions for future research. Internet
Research—Electronic Networking Applications and Policy, 8(4), 322-330.

Joreskog, K. G. (1973). A general method for estimating a linear structural equation
system. In A. S. Goldberger & O. D. Duncan (Eds), Structural equation models
in social science (pp. 85-112). New York: Seminar Press.

Joreskog, K. G., & Sorbom, D. (1981). LISREL: Analysis of linear structural

relationships by the method of maximum liklihood. Chicago: International
Educational Services.

178

Joreskog, K. G., & Sorbom, D. (1993). LISREL8: Structral equation modeling with the
SIMPLIS command language. Hillsdale, NJ: Erlbaum Associates.

Katona, G., & Mueller, E. (1955). A study of purchase decisions. In L. H. Clark (Ed.),
Consumer behavior, the dynamics of consumer reaction (pp. 30-87). New York:
New York University Press.

Keeney, R. L. (1999). The value of lntemet commerce to the customer. Management
Science, 45(4), 533-542.

Kiel, G. C., & Layton, R. A. (1981). Dimensions of consumer information seeking
behavior. Journal of Marketing Research, 18(2), 233-239.

Kohn, M. G., & Shavell, S. (1974). The theory of search. Journal of Economic Theory,
9(2), 93-123.

Kujala, J. T., & Johnson, M. D. (1993). Price knowledge and search behavior for
habitual, low involvement food purchases. Journal of Economic Psychology,
14(2), 249-265.

Lee, H., Herr, P. M., Kardes, F. R., & Kim, C. (1999). Motivated search: Effects of
choice accountability, issue involvement, and prior knowledge on information
acquisition and use. Journal of Business Research, 45(1), 75-88.

Lichtenstein, D. R., Ridgway, N. M., & Netemeyer, R. G. (1993). Price perceptions and

consumer shopping behavior: A field study. Journal of Marketing Research,
30(2), 234-245.

Lohse, G. L., Bellman, S., & Johnson, E. J. (2000). Consumer buying behavior on the
intemet: Findings from panel data. Journal of Interactive Marketing, 14(1), 15-29.

Lukosius, V., Hyman, M. R., & Stratemeyer, A. W. (2001). A model of online
information search. Paper presented at the American Marketing Association 2001
AMA Winter Educators' Conference, Scottsdale, Arizona.

Lynch, J. G., & Ariely, D. (2000). Wine online: Search costs affect competition on price,
quality, and distribution. Marketing Science, 19(1), 83-103.

MacCallum, R. C., Browne, M. W., & Sugawara, H. M. (1996). Power analysis and
determination of sample size for covariance structure modeling. Psychological
Methods, [(2), 130-149.

Maclnnis, D. J., Moorman, C., & Jaworski, B. J. (1991). Enhancing and measuring

consumers motivation, opportunity, and ability to process brand information from
ads. Journal of Marketing, 55(4), 32-53.

179

 

 

MacKenzie, S. B. (2001). Opportunities for improving consumer research through latent
variable structural equation modeling. Journal of Consumer Research, 28(1), 159-
166.

Maute, M. F., & Forrester, W. R. (1991). The effect of attribute qualities on consumer
decision-making: A causal model of external information search. Journal of
Economic Psychology, 12(4), 643-666.

McQuarrie, E. F., & Munson, J. M. (1992). A revised product involvement inventory:
Improved usability and validity. In J. F. Sherry & B. Stemthal (Eds), Advances in
consumer research (V 01. 19, pp. 108-115). Provo, UT: Association for Consumer

Research.

Michael, R. T. (1972). The eflect of education on efficiency in consumption. New York:
National Bureau of Economic Research.

Miller, H. J. (1993). Consumer search and retail analysis. Journal of Retailing, 69(2),
160-192.

Mittal, B. (1989). Must consumer involvement always imply more information search. In
T. K. Srull (Ed.), Advances in consumer research (Vol. 16, pp. 167-172). Provo,
UT: Association for Consumer Research.

Mittal, B., & Lee, M. S. (1989). A causal model of consumer involvement. Journal of
Economic Psychology, 10(3), 363-389.

Moore, W. L., & Lehmann, D. R. (1980). Individual-differences in search behavior for a
nondurable. Journal of Consumer Research, 7(3), 296-307.

Moorthy, S., Ratchford, B. T., & Talukdar, D. (1997). Consumer information search
revisited: Theory and empirical analysis. Journal of Consumer Research, 23(4),
263-277.

Newman, J ., & Staelin, R. (1972). Prepurchase information seeking for new cars and
major household appliances. Journal of Marketing Research, 9(3), 249-257.

Newman, J. W. (1977). Consumer external search: Amount and determinants. In A. G.
Woodside, J. N. Sheth & P. D. Bennett (Eds), Consumer and industrial buying
behavior (pp. 79-94). New York: North-Holland.

Newman, J. W., & Lockeman, B. D. (1975). Measuring prepurchase information search.
Journal of Consumer Research, 2, 216-222.

Newman, J. W., & Staelin, R. (1971). Multivariate analysis of differences in buyer
decision time. Journal of Marketing Research, 8, 192-198.

180

Nicosia, F. (1966). Consumer Decision Processes: Marketing and Advertising
Implications. New York: Prentice Hall.

Nystrom, H., Tamsons, H., & Thams, R. (1975). An experiment in price generalization
and discrimination. Journal of Marketing Research, 12(2), 177-181.

O'Reilly, C. A. (1982). Variations in decision makers' use of information scources: The

impact of quality and accessibility of information. Academy of Management
Journal, 25(4), 756-771.

Olshavsky, R. W. (1985). Towards a more comprehensive theory of choice. In E. C.
Hirschman & M. B. Holbrook (Eds), Advances in consumer research (Vol. 12,
pp. 465-470). Provo, UT: Association for Consumer Research.

Park, C. W., & Mittal, B. (1985). A theory of involvement in consumer behavior:

Problems and issues. In J. N. Sheth (Ed.), Research in consumer behavior (V 01. 1,
pp. 201-231). Greenwich, CN: JAI Press.

Park, C. W., Mothersbaugh, D. L., & Feick, L. (1994). Consumer knowledge assessment.
Journal of Consumer Research, 21(1), 71-82.

Parsons, A. G. (2002). Non-functional motives for online shoppers: Why we click.
Journal of Consumer Marketing, 19(5), 380-392.

Peterson, R. A., Balasubramanian, S., & Bronnenberg, B. J. (1997). Exploring the
implications of the lntemet for consumer marketing. Journal of the Academy of
Marketing Science, 25(4), 329-346.

Petty, R. E., & Cacioppo, J. T. (1981). Attitudes and persuasion: Classic and
contemporary approaches. Dubuque, IA: W.C. Brown Co.

Petty, R. E., & Cacioppo, J. T. (1986). Communication and persuasion: Central and
peripheral routes to attitude change. New York: Springer-Verlag.

Punj, G., & Srinivasan, N. (1989). Influence of expertise and purchase experience on the
formation of evoked sets. In T. K. Srull (Ed.), Advances in consumer research

(V01. 16, pp. 507-514). Provo, UT: Association for Consumer Research.

Punj, G. N., & Staelin, R. (1983). A model of consumer information-search behavior for
new automobiles. Journal of Consumer Research, 9(4), 366-380.

Putrevu, S., & Ratchford, B. T. (1997). A model of search behavior with an application to
grocery shopping. Journal of Retailing, 73(4), 463-486.

181

Rao, A. R., & Sieben, W. A. (1992). The effect of prior knowledge on price acceptability
and the type of information examined. Journal of Consumer Research, 19(2),
256-270.

Ratchford, B. T. (1982). Cost-benefit models for explaining consumer choice and
information seeking behavior. Management Science, 28(2), 197-212.

Ratchford, B. T., & Srinivasan, N. (1993). An empirical investigation of returns to
search. Marketing Science, 12(1), 73-87.

Ratchford, B. T., Talukdar, D., & Lee, M. S. (2001). A model of consumer choice of the
lntemet as an information source. International Journal of Electronic Commerce,
5(3), 7-21.

Rose, G. M., Lees, J ., & Meuter, M. L. (2001). A refined view of download time impacts
on e-consumer attitudes and patronage intentions toward e-retailers. Retrieved
September, 2001, from
www.mediajourn_al.org/netacademy/publications.nsf/all pk/1946

Rothschild, M. ( 1974). Searching for the lowest price when the distribution of prices is
unknown. Journal of Political Economy, 82(4), 689-711.

Salam, A. F. (1998). Three essays on electronic commerce: WWW information content,
interenet involvement, and perceived risk. Unpublished Doctorial Dissertation,
State University of New York at Buffalo, Buffalo.

Salmon, C. T. (1986). Perspectives on involvement in consumer and communication
research. In D. B. & V. M. (Eds), Progress in Communication Sciences (V 01. 7,
pp. 243-269). Norwood, NJ: Ablex Publishing Company.

Salop, 3., 8c Stiglitz, J. (1977). Bargains and ripoffs: Model of monopolistically
competitive price dispersion. Review of Economic Studies, 44(3), 493-510.

Schaninger, C. M., & Sciglimpaglia, D. (1981). The influence of cognitive personality-
traits and demographics on consumer information acquisition. Journal of
Consumer Research, 8(2), 208-216.

Schmidt, J. B., & Spreng, R. A. (1996). A proposed model of external consumer
information search. Journal of the Academy of Marketing Science, 24(3), 246-
256.

Smith, J. B., & Bristor, J. M. (1994). Uncertainty orientation: Explaining differences in

purchase involvement and external search. Psychology & Marketing, 11(6), 587-
607.

182

Smith, M. D., Bailey, J ., & Brynjolfsson, E. (2000). Understanding digital markets:
Review and assessment. In E. Brynjolfsson & B. Kahin (Eds), Understanding the
digital economy: Data, tools, and research (pp. 99-136). Cambridge, MA: MIT
Press.

Smith, M. D., & Brynjolfsson, E. (2001). Consumer decision-making at an Internet
shopbot: Brand still matters. Journal of Industrial Economics, 49(4), 541-558.

Spreng, R. A., & Olshavsky, R. W. (1989). Exploring the headwaters of the prior
knowledge-search relationship. Paper presented at the American Marketing
Association Summer Educators' Conference, Chicago.

Srinivasan, N. (1986). A path analytic model of external search for information for new
automobiles. In M. E. Wallendorf & P. Anderson (Eds), Advances in consumer
research (V o]. 14, pp. 319-322). Provo, UT: Association for Consumer Research.

Srinivasan, N. (1987). A causal model of external serach for information for durables: A
particular investigation in the case of new automobiles. Unpublished Doctorial
Dissertation, State University of New York at Buffalo, Buffalo, NY.

Srinivasan, N. (1990). Pre-purchase external search for information. In V. A. Zeithaml
(Ed.), Review of marketing (pp. 153-189). Chicago: American Marketing
Association.

Srinivasan, N., & Ratchford, B. T. (1991). An empirical-test of a model of external
search for automobiles. Journal of Consumer Research, 18(2), 233-242.

Steinberg, J. (2000, February 20, 2000). DealTime.com dares to compare. Jerusalem
Post, p. 8.

Stigler, G. J. (1961). The economics of information. Journal of Political Economy, 69(3),
213-225.

Sti glitz, J. E. (1979). Equilibrium in product markets with imperfect information.
American Economic Review, 69(2), 339-345.

Sundaram, D. S., & Taylor, R. D. (1998). An investigation of external information search
effort: Replication in in-home shopping situations. In J. W. Alba & J. W.
Hutchinson (Eds), Advances in consumer research (Vol. 25, pp. 440-445). Provo,
UT: Association for Consumer Research.

Telser, L. G. (1973). Searching for the lowest price. American Economic Review, 63(2),
40-49.

Udell, J. G. (1966). Prepurchase behavior of buyers of small electrical appliances.
Journal of Marketing, 30(4), 50-52.

183

Urbany, J. E. (1986). An experimental examination of the economics of information.
Journal of Consumer Research, 13(2), 257-271.

Urbany, J. E., Bearden, W. 0., Kaicker, A., & Smith-deBorrero, M. (1997). Transaction
utility effects when quality is uncertain. Journal of the Academy of Marketing
Science, 25(???), 45-55.

Urbany, J. E., Dickson, P. R., & Kalapurakal, R. (1996). Price search in the retail grocery
market. Journal of Marketing, 60(2), 91-104.

Urbany, J. E., Dickson, P. R., & Sawyer, A. G. (2000). Insights into cross and within
store price search: Retailer estimates vs. consumer self-reports. Journal of
Retailing, 76(2), 243-258.

Urbany, J. E., Dickson, P. R., & Wilkie, W. L. (1989). Buyer uncertainty and information
search. Journal of Consumer Research, 16(2), 208-215.

Van den Poel, D., & Leunis, J. (1999). Consumer acceptance of the intemet as a channel
of distribution. Journal of Business Research, 45(3), 249-256.

Verplanken, B., Hazenberg, P. T., & Palenewen, G. R. (1992). Need for cognition and
external information search effort. Journal of Research in Personality, 26(2), 128-
136.

Wakefield, K. L., & Inman, J. J. (1993). Who are the price vigilantes: An investigation of
differentiating characteristics influencing price information processing. Journal of

Retailing, 69(2), 216-233.

Ward, M. R., & Lee, M. J. (2000). Internet shopping, consumer search and product
branding. Journal of Product and Brand Management, 9(1), 6-20.

Ward, M. R., & Morganosky, M. (2000). Online consumer search and purchase in a
multiple channel environment, 2000]. from www.uiuc.edu/ph/www/wa;d_l

Weitzman, M. L. (1979). Optimal Search for the Best Alternative. Econometrica, 47(3),
641-654.

Westbrook, R. A., & Black, W. C. (1985). A motivation-based shopping typology.
Journal of Retailing, 61(1), 78-103.

Whelan, D. (2001, September 2001). A tale of two consumers. American Demographics,
23, 54-57.

Wilde, L. L., & Schwartz, A. (1979). Equilibrium comparison shopping. Review of
Economic Studies, 46(3), 543-553.

184

 

Wilder, C. (1998, December 7, 1998). Call your agent for online shopping.
Informationweek, 126-128.

Wilkie, W. L., & Dickson, P. R. (1991). Shopping for appliances: Consumers’ strategies
and patterns of information search. In H. H. Kassaijian & T. S. Robertson (Eds),
Perspectives in consumer behavior (4th ed., pp. 1-26). Englewood Cliffs, NJ:
Prentice Hall.

Wolfmbarger, M., & Gilly, M. C. (2001). Shopping online for freedom, control, and fun.
California Management Review, 43(2), 34-55.

Zaichkowsky, J. (1985). Measuring the involvement construct. Journal of Consumer
Research, 12(3), 341-353.

Zaichkowsky, J. L. (1994). The personal involvement inventory: Reduction, revision, and
appllication to advertising. Journal of Advertising, 23(4), 59-69.

Zimmermann, L. K., & Geistfeld, L. V. (1984). Economic factors which influence

consumer search for price information. Journal of Consumer Affairs, 18(1), 119-
130.

185

 

 

  

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

1111111111111111111111

1230