ANALYSIS OF CONSUMERS’ VISUAL PERCEPTION OF GARMENT FIT: AN EYE TRACKING STUDY by Lijuan Pi A thesis submitted to the Faculty of the University of Delaware in partial fulfillment of the requirements for the degree of Master of Science in Fashion Studies Spring 2011 Copyright 2011 Lijuan Pi All Rights Reserved
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
ANALYSIS OF CONSUMERS’ VISUAL PERCEPTION OF GARMENT FIT:
AN EYE TRACKING STUDY
by
Lijuan Pi
A thesis submitted to the Faculty of the University of Delaware in partial
fulfillment of the requirements for the degree of Master of Science in Fashion Studies
Spring 2011
Copyright 2011 Lijuan Pi
All Rights Reserved
ANALYSIS OF CONSUMERS’ VISUAL PERCEPTION OF GARMENT FIT:
AN EYE TRACKING STUDY
by
Lijuan Pi
Approved: __________________________________________________________
Sharron J. Lennon, Ph.D.
Professor in charge of thesis on behalf of the Advisory Committee
Approved: __________________________________________________________
Marsha A. Dickson, Ph.D.
Chair of the Department of Fashion and Apparel Studies
Approved: __________________________________________________________
George H. Watson, Ph.D.
Dean of the College of Arts and Sciences
Approved: __________________________________________________________
Charles G. Riordan, Ph.D.
Vice Provost for Graduate and Professional Education
iii
ACKNOWLEDGMENTS
I would like to thank everyone who offered help in my master study and
research. First, I like to express my sincere appreciation to my advisor, Professor
Sharron J. Lennon for her generous support, continuous direction, and endless
encouragement in my master study. I would like to thank Professor Jaehee Jung for her
valuable comments in my proposal and research. I also would like to thank Professor
Kelly Cobb for her encouragement and kindly help from the origination of the research
topic. I would like to thank my friend Rich Burns and his advisor Professor Sandra
Carberry at University of Delaware for offering me this valuable access to eye-tracker.
I also would like to thank all the other professors and staffs in Dept. of Fashion and
Apparel Studies for giving me all kinds of help in the past two years.
Finally, I would like to thank my family and my friends who give me the
mental power to complete my study.
iv
TABLE OF CONTENTS
LIST OF TABLES ......................................................................................................... vi
LIST OF FIGURES ...................................................................................................... vii
ABSTRACT ................................................................................................................ viii
Chapter
1. INTRODUCTION ...................................................................................................... 1
Statement of Problem ......................................................................................... 2
Purpose of the Study ........................................................................................... 6
Definition of Terms .......................................................................................... 10
2. REVIEW OF LITERATURE ................................................................................... 12
Online Shopping for Apparel ........................................................................... 12
Online Visual Product Presentation Techniques .............................................. 13
Self-Discrepancy Theory .................................................................................. 15
Body Cathexis ................................................................................................... 18
Body Image Discrepancy .................................................................................. 19
Visual Attention, Eye Movement and Eye Tracker .......................................... 21
Negative Psychological State on Visual Attention ........................................... 24
Consumers’ Concern with Fit ........................................................................... 25
Confidence in Fit Judgment .............................................................................. 27
Purchase Intention ............................................................................................ 29
Body Cathexis and Body Image Discrepancy ................................................... 31
Body Cathexis, Body Image Discrepancy and Visual Perception .................... 32
Visual Perception of Garment Fit and Concern with Fit .................................. 36
Concern with Fit and Confidence in Fit Judgment ........................................... 38
Concerns with Fit and Purchase Intent ............................................................. 39
Confidence in Judgment and Purchase Intention.............................................. 39
Effect of Body Sites on Visual Perception ....................................................... 40
v
3. METHOD ................................................................................................................. 42
Sample .............................................................................................................. 42
Stimuli .............................................................................................................. 42
Experimental Design ........................................................................................ 44
Instruments ....................................................................................................... 45
Procedure .......................................................................................................... 51
4. ANALYSIS OF RESULTS ...................................................................................... 56
Demographics ................................................................................................... 58
Descriptive Statistics for Variables .................................................................. 58
•Body Cathexis ........................................................................................ 58
•Body Discrepancy .................................................................................. 59
•Visual Attention ..................................................................................... 61
•Concern with Fit..................................................................................... 62
•Confidence in Judgment......................................................................... 63
•Purchase Intention .................................................................................. 64
Hypotheses Testing ........................................................................................... 65
5. DISCUSSION AND CONCLUSIONS .................................................................... 84
Discussion ......................................................................................................... 85
Conclusions and Implications ........................................................................... 96
Limitations ...................................................................................................... 101
Appendix A- Garment Stimuli ................................................................................... 102
Appendix B- Body cathexis scale and Body image discrepancy scales ..................... 106
Appendix C- Concern with Fit, Confidence in Judgment, and Purchase Intent ......... 109
Appendix D- Permission for using photos of human model ...................................... 110
Appendix E- University of Delaware IRB Approval .................................................. 111
References .................................................................................................................. 112
vi
LIST OF TABLES
Table 4.1 Statistic analysis ......................................................................................... 57
Table 4.2 Demographics: Age, Weight, and Height ................................................... 58
Table 4.3 Descriptive Statistics for Body cathexis and sub-items .............................. 59
Table 4.4 Descriptive Statistics for Body self-discrepancy ........................................ 61
Table 4.5 Descriptive Statistics for duration and numbers of fixations ..................... 62
Table 4.6 Descriptive Statistics for concern with fit .................................................. 63
Table 4.7 Descriptive Statistics for Confidence of judgment ..................................... 64
Table 4.8 Descriptive Statistics for Purchase intent ................................................... 64
Table 4.9 Linear Regression Results for Hypotheses 2 .............................................. 66
Table 4.10 Linear Regression Results for Hypotheses 3 .............................................. 68
Table 4.11 Linear Regression Result for Hypothesis 4 ................................................ 70
Table 4.12 Linear Regression Result for Hypothesis 5 ................................................ 71
Table 4.13 Linear Regression Result for Hypotheses 6 ................................................ 72
Table 4.14 Linear Regression Result for Hypotheses 7 ................................................ 72
Table 4.15 One-way repeat measures ANOVA Result for Hypothesis 8 ..................... 75
Table 4.16 Means of duration of fixations and number of fixations ............................ 76
Table 4.17 Summery of Hypotheses Testing Results ................................................... 81
vii
LIST OF FIGURES
Figure 2.1 Theoretical framework of current study ..................................................... 41
Figure 3.1 An example of stimulus divided into AOIs ................................................ 48
Figure 3.2 An example of “hot-spot” graph ................................................................ 49
Figure 3.3 An example of “scan-path” graph .............................................................. 50
Figure 3.4 Tobii eye tracker (www.tobii.com) ............................................................ 54
Figure 3.5 An example of a questionnaire item divided into AOIs ............................. 55
Figure 4.1 Hypotheses tested in this study .................................................................. 56
Figure 4.2a Effect of body sites on duration of fixations in Front view ....................... 78
Figure 4.2b Effect of body sites on duration of fixations in back view ........................ 78
Figure 4.3a Effect of body sites on number of fixations in Front View ....................... 79
Figure 4.3b Effect of body sites on number of fixations in back view ......................... 80
Figure 5.1 Construction details and draping effects might draw attention ................. 96
Figure A.1 Front-view and back view of blazer of size 2 ......................................... 102
Figure A.2 Front-view and back view of blazer of size 4 ......................................... 103
Figure A.3 Front-view and back view of blazer of size 6 ......................................... 104
Figure A.4 Front-view and back view of blazer of size 8 ......................................... 105
viii
ABSTRACT
The current online apparel retailers are not able to provide enough visual
information about garment fit to online apparel shoppers, even though various visual
product presentation techniques have been developed. The present study was designed
to analyze consumers’ visual perception of garment fit in order to provide managerial
suggestions for online apparel retailers. This study used self-discrepancy theory
(Higgins, 1987) as a theoretical base. The purposes of this study were to examine the
effect of subjective factors (body image discrepancy, body satisfaction) and objective
factors (body sites) on consumers’ visual perceptions (duration of fixations and
number of fixations) of garment fit, and to examine the effect of visual perception on
consumers’ concern with fit judgment, confidence in fit judgment, and purchase
intention for the garment. Forty-five college women participated in this study for extra
credit and incentives. Eight photos (front view and back view of blazers of four sizes
worn by a human model) were created as visual stimuli. Participants’ duration time
and number of fixations on the visual images of blazers when making judgments of fit
of blazers were measured by an eye tracker. Participant’s body image discrepancy,
body satisfaction, concern with fit, confidence in fit judgment, and purchase intention
were also measured. Using descriptive statistics, one-way univariate analyses of
variance and simple regression analyses, the results of present study indicated: 1) there
was a significant relationship between female college consumers’ body image
discrepancy and their body satisfaction-Hypothesis 1 was supported; 2) female
consumers’ body image discrepancy and body satisfaction significantly predicted their
ix
visual attention over the garments (how long and how often they looked at the
garments) when making garment fit judgments-Hypotheses 2 and 3 were supported; 3)
no relationships was found between female consumers’ visual attention and their
concern with garment fit-Hypothesis 4 was not supported; 4) there was no significant
relationship between female consumers’ concern with garment fit and their confidence
in fit judgments-Hypothesis 5 was not supported; 5) female consumers’ concern with
fit was a significant predictor of their purchase intent for the garment- Hypothesis 6
was supported; 6) there was no significant relationship between female consumers’
confidence in fit judgments and purchase intention- Hypothesis 7 was not supported;
7) Consumers attended to some human body sites more than others -Hypothesis 8 was
supported.
1
Chapter 1
INTRODUCTION
The online retailing market is fast growing and profitable. According to
the US Department of Commerce, total online retail sales were $134.9 billion for
2009, and online retail sales for the fourth quarter of 2009 increased 14.6% from that
of the fourth quarter of 2008 (US Census Bureau News, 2010). Online retail in both
the US and Western Europe is expected to maintain double-digit growth over the next
five years, according to the forecasts by Forrester Research Inc (Forrester Research,
2010).. Apparel is one of the three product categories that dominate the online retail
market; the others are consumer electronics and consumer computer peripherals. In
2009, online apparel retail sales were $12 billion, accounting for 4.9% of total apparel
retail sales.
Although online apparel shopping offers some benefits, such as
convenience and diverse product categories (Bhatnagar, Misra, & Rao, 2000), there are
a number of obstacles preventing consumers from buying apparel products online. One
of the obstacles is the inability to characterize the product accurately (Hammond &
Kohler, 2000). Many of the important characteristics of a garment, such as color, feel,
and fit, are difficult to communicate “virtually” online, which impedes consumers’
decision making process when they purchase apparel from online stores.
2
Currently, consumers make their purchase and size-selection decisions
based on two dimensional (2D) photos of garments, videos, and sizing charts; they are
not able to try clothes on when they shop online (Citrin, Stern, Spangenberg, & Clark,
2003). Concerns with fit and size of garments have become serious issues and are the
biggest sources of perceived risk of online apparel shopping (Beck, 2003). The
difficulty in presenting fit information adequately and the inability to try on garments
contributes to a high return rate in online apparel retailing (Horrigan, 2008).
According to Beck (2003), the return rate for apparel bought online was 14%, about
twice as high as return rates for other products bought on the Web. High return rates
increase the cost of selling merchandise for online retailers due to restocking and
reselling costs (Mollenkopf, Rabinovich, Laseter, & Boyer, 2007).
Statement of Problem
To improve the ability to present fit and size information and thereby
reduce the return rates of apparel products, computer aided design (CAD) companies,
such as “Optitex.com” and “My virtual model.com,” have developed virtual-try-on
techniques for the apparel industry, enabling visualization of garments on three
dimensional (3D) avatars. Virtual try-on is defined as a computer simulation that
enables customers to select their garments, and try them on 3D virtual models (Volino,
Cordier, & Magnenat-Thalmann, 2005). The online virtual model is a visual tool that
can improve the ability to represent garment information (color, design, texture, and
fit) and simulate the garment’s look on a consumer’s body (Istook, 2008). Fit
information is the most important type of information that online retailers want to
3
present to consumers through the virtual model (Ashdown & DeLong, 2007). Lands’
End adopted the virtual try-on service from My Virtual Model (www.mvm.com) in
1998, in an attempt to provide more fit information to online shoppers.
In addition, with the development of the 3D body scanner (TC2, 2010), 3D
body measurements of consumers can be easily obtained. Based on digital 3D body
measurements, more accurate virtual models and garment draping effects can be
created and presented on a computer screen through the use of some computer
algorithms (Istook & Hwang, 2001).
However, most of these virtual models are not as effective as people
imagined. Now, online retailers, such as Lands’ End, have removed the virtual try-on
service from their websites. Usually, these virtual models were developed from a
technical perspective since the developers were computer scientists. However, a
consumer point of view needs to be considered with respect to buying apparel online
(Kim, 2009), because clothing is the "skin" one chooses to wear to project one's self-
image to the public and hence is intimately tied to one's sense of self (Hammond &
Kohler, 2000).
Little research has analyzed consumers’ responses toward virtual models.
In Kim’s (2009) study, questionnaires and an interview were used to access
consumers’ fit evaluation of virtual pants on a virtual model and real pants on a real
body. The results showed that the visual information from the virtual model provided
participants’ an overall idea and perception of the pants fit, but some aspects of the
virtual model made the image illustrating the pants fit inaccurate, such as locations of
tightness and looseness. Lim (2009) did similar research comparing people’s
4
evaluation of images of virtual garments worn by an avatar to the evaluation of real
garments worn by human models. Her results indicated most respondents preferred
real garments on the real body. Although Kim (2009) indicated that the locations of
tightness and looseness are important for consumers’ evaluation of fit, she did not
obtain information about the specific body locations and how important they are in fit
judgments and in confidence of fit judgments.
Apparel fit judgments are closely linked to peoples' feelings about
themselves and the image they wish to project (Hammond & Kohler, 2000; Kim &
Damhorst, 2010). Body image is defined as a mental image of one’s body and how
individuals perceive their own bodies (Garner & Garfinkel, 1981). Thus, it is
reasonable to expect that apparel fit judgments may be associated with body image and
one way to assess body image is through ratings of body satisfaction. Feather, Ford,
and Herr (1997) investigated 503 female collegiate basketball players’ perceptions of
body and garment fit satisfaction; the result showed that these female athletes’
satisfaction with the fit of their uniform parallels their satisfaction with their body.
Researchers have found that clothing has the potential to improve an individual’s body
image because women evaluate their bodies higher when clothed rather than unclothed
(Markee, Carey & Pedersen, 1990).
Ashdown and Delong (2007) mentioned that the current evaluation and
analysis of garment fit is mainly based on expert judgment and live models’
anthropometric data and also examined the issue of fit of women’s ready-to-wear
(RTW) from the perspective of garment makers. Only a few studies have analyzed
garment fit from the consumers’ perspective. For example, LaBat and DeLong (1990)
5
analyzed the body cathexis (body satisfaction) and the satisfaction with fit of clothing
of 107 female consumers using survey methodology. The results indicated that the
body satisfaction scores were slightly lower for consumers’ lower body, and also
confirmed a significant relationship between the consumers’ satisfaction with fit and
their satisfaction with their body. Ashdown and DeLong (1995) studied consumers’
tactile perceptions of apparel ease by conducting experiments with a set of computer-
generated pants, the patterns of these pants were created and graded using software.
The results revealed that the people’s perceptible fit variation could be quite small, and
individuals vary in their tolerance for fit variations at different locations on the body.
These studies investigated consumers’ general attitudes about fit and their tactile
perception of fit when they try on or think about clothes.
When a female consumer tries on a garment, her judgment of fit is based
on visual feedback (Delong, Kim, & Larntz, 1993) and tactile perception (Ashdown &
Delong, 1995). The tactile perception could result from the nerves in the surface of the
skin and the deep pressure sensations felt by the nerves in muscles and joints
(McBurney & Collins, 1977). The pressure sensation may result from an overall
feeling of pressure on the whole body and the separate feedback from localized body
sites. In order to learn about fit, these various tactile and visual responses need to be
considered separately. How important is the visual feedback in the fit judgment
process? To answer this question, consider what consumers do when trying on clothes
in the fitting room. One important and inevitable step is to observe their figures in a
mirror. Without the visual feedback obtained from mirror, we would not be able to
make a confident judgment about the garment fit. “Our clothing … is ordinarily felt as
6
an integral part of ourselves” (Knoblich, Thornton, Grosjean, & Shiffar, 2006, p.34).
We could infer the important role of visual feedback in the perception of clothing fit
from its role in human body perception. Kinsbourne and Lempert (1989) found that
congenitally blind children may mistakenly represent some parts of their bodies when
they are asked to reproduce models of their own bodies with plasticine, which means
that subjective perceptions of one’s own body may be distorted in an absence of
vision.
Purpose of the Study
In the online shopping context, consumers cannot obtain any tactile
sensation from the virtual model, although consumers have the need to touch apparel
products (Workman, 2009). They can only infer a tactile perception from the verbal
descriptions and visual presentations of the textile on the website (Kim & Lennon,
2008).
We expect virtual models to reflect the characteristics of real garments and
real bodies as accurate and realistically as possible. Based on the realism concepts
proposed in Hagen (1986), Ferwerda (2003) identified three varieties of realism in
computer graphics to evaluate how accurately a natural scene was represented by
computer generated images: Physical realism, photo realism, and functional realism.
Physical realism means the computer generated image could provide the same visual
stimulation as the natural scene, which is almost impossible to achieve based on
current computer techniques; photo realism means that the virtual image is able to
produce the same visual response as the natural scene; functional realism means that
7
the image provides as much visual information as the natural scene (Ferwerda, 2003).
In addition, Ferwerda also proposed two criteria to measure the functional realism of
computer graphics: (1) accuracy, which means that natural object’s physically
measurable properties are correctly represented in the image; (2) fidelity, which means
that the image could provide enough information so that the viewers are able to
perform tasks with the image as they do in the real world. Kim (2009) used these two
criteria to evaluate the effectiveness of virtual models in representing the fit of pants.
She compared 37 participants’ evaluations of fit of pants on a virtual model (generated
by their body-scanning data) and the fit of pants on their bodies. The result showed
that although virtual models have the ability to accurately present the overall visual
information of how pants look on a human body, some aspects of the virtual
simulation made the pants fit image inaccurate. To increase the accuracy and provide
more functional information from the virtual image, computer scientists developed
algorithms for rendering the virtual image (Ferwerda, Westin, Smith & Pawlicki,
2004). However, these algorithms are very expensive and need computational time,
hence it is impossible to use these algorithms with online virtual models. To save
computational time, a visual attention-based rendering algorithm has been developed
to improve the realism of virtual images. This means that only those important cues in
people’s visual attention are rendered with high accuracy and fidelity (Sundstedt,
Debattista, Longhurst, Chalmers & Troscianko, 2005). Researchers have found that a
human’s response to visual stimuli involves two stages; the first is an essentially
parallel stage which occurs within the first glimpse of the stimuli and provides a global
impression; the second stage is to serially examine the detailed information using eye
8
movements (Todd & Kramer, 1993). According to previous studies in human visual
attention (Torralba, Oliva, Castelhano, & Henderson, 2006; Treisman, 1982), when a
visual stimulus is presented to people, some important visual cues are serially selected
and used to construct the visual perception, while others are ignored. Based on the
knowledge of selective human visual attention and the assistance from some high-tech
equipment (i.e., an eye tracker), researchers can identify important cues in women’s
visual perception of garment fit. These important cues will help scientists develop a
perception-based rendering algorithm to improve the accuracy and fidelity of virtual
models, and finally makes the virtual model more realistic and more effective in
conveying the information of fit and size. Hence, the current study might have future
commercial applications in online retailing.
The purposes of this study were to: 1) examine the effect of body sites on
female consumers’ visual perceptions of a garment; 2) examine the relationship among
female consumers’ body image discrepancy, body satisfaction and visual perception of
garment fit (focus on how long and how often they look at the garments); 3) examine
the relationship among female consumers’ visual perception, concern with garment fit,
confidence in fit judgment, and purchase intent. Generally, the study analyzed the
effect of the subjective factors (body image discrepancy, body satisfaction) and
objective factors (body sites) on consumers’ visual perception (duration of fixations
and number of fixations) of garment fit and how the visual perception affects
consumers’ purchase decision making process.
Eight hypotheses were investigated in this study:
9
Hypothesis 1- Participants’ total body image discrepancy will be
negatively related to their total body satisfaction;
Hypothesis 2a -Participants’ total body image discrepancy will be
positively related to the total duration of fixations over the garment when making
judgments of fit;
Hypothesis 2b -Participants’ total body image discrepancy will be
positively related to the total number of fixations over the garment when making
judgments of fit;
Hypothesis 3a - Participants’ total body satisfaction will be negatively
related to the total duration of fixations over the garment when making judgments of
fit;
Hypothesis 3b - Participants’ total body satisfaction will be negatively
related to the total number of fixations over the garment when making judgments of
fit;
Hypothesis 4a - Duration of fixations on human model will be negatively
related to concern with fit;
Hypothesis 4b - Number of fixations on human model will be negatively
related to concern with fit;
Hypothesis 5 - Consumers’ concern with garment fit will be negatively
related to their confidence in fit judgments;
Hypothesis 6 - Consumers’ concern with fit will be negatively related to
their purchase intent;
10
Hypothesis 7 - Consumers’ confidence with fit judgments will be
positively related to their purchase intent;
Hypothesis 8a - There is a significant difference among the body sites of
the human model on duration of fixations when judging the garment fit;
Hypothesis 8b - There is a significant difference among the body sites of
the human model on number of fixations when judging the garment fit.
Definition of Terms
Terms used in this study are defined as follows:
Body cathexis: Body cathexis is people’s satisfaction or dissatisfaction for
various parts of their body (Rosen & Ross, 1973).
Body image discrepancy: Derived from self-discrepancy theory (Higgins,
1987), it is the differences between aspects of an individual’s real self (i.e. perceived
body shape or size) and idealized self (Thompson, Heinberg, Altabe & Tantleff-Dunn,
1999, p. 134).
Concern with fit: “the subjectively determined expectations and amount
of risk perceived by a shopper in relation to the fit and size of the garment in
contemplating a particular purchase decision” (Kim & Damhorst 2010, p. 242).
Confidence of judgment: The strength of belief about the quality of a
judgment or choice (Sniezek, 1992).
Duration of fixation: In this study, how long a participant’s eyes fixate on
an area.
11
Eye Tracker: It is a measurement device used for measuring eye
movement (Duchowski, 2007).
Fixation: It is the eye movement that stabilizes the retina over a stationary
object of interest and it corresponds to the desire to maintain one’s gaze on an object
of interest (Duchowski, 2007).
Number of fixations: In this study, how many times a participant’s eyes
fixate on an area.
Purchase intention: It reflects “what we think we will buy” (Blackwell
Miniard, and Engel, 2001, p. 283).
Visual attention: Visual attention means people focus their mental
capacities on the selection of sensory input so that the mind can successfully process
the visual stimulus of interest (Duchowski, 2007).
12
Chapter 2
REVIEW OF LITERATURE
The present study analyzed consumers’ visual perceptions of garment fit in
order to provide managerial suggestions for online apparel retailers. At the beginning
of this chapter, online apparel shopping and visual product presentation techniques
will be discussed as contexts of this study. Self-discrepancy theory will be used as the
theoretical framework of this study. The important concepts, body cathexis, body
image discrepancy, concern with fit, visual perception, confidence of judgment, and
purchase intent, will also be discussed. Eye movement as an observable and
measureable factor which reflects people’s visual perception process will be reviewed.
Online Shopping for Apparel
The growth in the online retailing sector has become the major driving
force for a lot of multichannel retailers. In 2009, the online retail sales have reached
$134.9 billion, and have increased 14.6% in the fourth quarter (US Census Bureau
News, 2010). A Forrester Research report expected that online retail in both the US
and Western Europe would continue a double-digit growth (Forrester Research, 2010).
Apparel is one of the three popular product categories for online shopping; the others
are consumer electronics and consumer computer peripherals. In 2009, 4.9% of total
apparel retail sales came from online retailing and the amount of online sales reached
$12 billion. Consumers use the internet as an important source of information for
13
buying apparel, because they enjoy some characteristics of online shopping, such as
convenience, unlimited product variety, time-saving nature, and easy accessibility
(Kim, Kim & Lennon, 2007).
In the online apparel shopping context, consumers are not able to try
clothes on before they receive the apparel products. Lack of information about size
and fit of apparel was a major reason why early online apparel sales lagged behind
books and music (Elliot & Fowell, 2000). The concerns with fit and size of garments
have become serious issues and remain important sources of perceived risk of online
apparel shopping (Beck, 2003). In apparel retail websites, attractive visual presentation
of apparel products is considered to be high task-relevant information for online
shoppers (Kim, Kim, & Lennon, 2007), because they depict the characteristics of
apparel products, such as color, texture, and fit, and help consumers make confident
purchase judgments (Eroglu, Machleit, & Davis, 2001). In research by Then and
Delong (1999), a large portion of respondents mentioned that they would like to see a
garment displayed on a human model to see how the garment fit the model’s body.
Online Visual Product Presentation Techniques
Therefore, online apparel retailers have tried innovative technologies to
enhance visual product presentation and provide size and fit information. According to
Retail Forward’s report in 2001, 3D images, virtual models, digital images, and
zooming technology were introduced to display apparel products and provide fit
information. Although Fan, Yu and Hunter (2004) mentioned 3D rotational images
and virtual models as promising trends and believed these new technologies could help
14
consumers make confident judgments about garment fit, these is still much room for
these new technologies to be improved. Landsend.com removed their MVM (My
Virtual Model) technology from their retail website after using it only for a few years
because the online virtual model was barely used by their customers. Kim (2009) and
Lim (2009) did similar lab experiments, which tested consumers’ attitude and
cognitive responses to computer generated virtual models and found similar results
that consumers preferred the real garments on real human models more than 3D virtual
garments worn by virtual models when they need to make the judgment about garment
fit.
Although some online retailers use online virtual models to provide ways
to virtually present 3D garment information to consumers (Ha, Kwon, & Lennon,
2007), they face the same problem of how to increase the realism of the virtual
presentation or how to display a virtual object in an accurate way. Consumers expect
virtual models to reflect the characteristics of real garments and real bodies as
accurately and realistically as possible. Ferwerda (2003) proposed two criteria to
measure the functional realism of computer graphics: (1) accuracy, which means that
the natural object’s physically measurable properties are correctly represented in the
image; (2) fidelity, which means that the image provides enough information so that
the viewers are able to perform tasks with the image as they do in the real world. Kim
(2009) used these two criteria to evaluate the effectiveness of virtual models in
representing the fit of pants. She compared 37 participants’ evaluations of the fit of
pants on a virtual model (generated by their body-scanning data) and the fit of pants on
their bodies. The result showed that although virtual models have the ability to
15
accurately present the overall visual information of how pants look on a human body,
some aspects of the virtual simulation made the pants fit image inaccurate. To increase
the accuracy and provide more functional information from the virtual image,
computer scientists have developed algorithms for rendering the virtual image
(Ferwerda et al., 2004).
However, these algorithms need extensive computational time, hence it is
inconvenient to use these algorithms with online virtual models. To save
computational time, a visual attention-based rendering algorithm has been developed
to improve the realism of virtual images. A visual attention-based rendering algorithm
is one for which only the areas in a visual image, on which people tend to place a large
amount of visual attention, are rendered with high accuracy and fidelity. The areas that
receive intense visual attention can be identified by using some attention capture
devices, such as an eye tracker (Sundstedt, Debattista, Longhurst, Chalmers &
Troscianko, 2005). Therefore, one of the goals of the current study is to identify and
analyze the areas in the visual image of garments, on which women are inclined to
place a large amount of visual attention when they are evaluating fit of a garment on
their bodies. The result of this study may help computer scientists to develop better
perception-based rendering algorithms and create a virtual model with better accuracy
and fidelity.
Self-Discrepancy Theory
The major theoretical framework of this study is self-discrepancy theory.
Higgins (1987) proposed self-discrepancy theory that “different types of self-
16
discrepancies represent different types of negative psychological situations that are
associated with different kinds of discomfort”. Self-discrepancy is defined as the gap
between those characteristics possessed by the actual self and those characteristics
people would like to possess (ideal self). The characteristics could be physical
appearance, academic performance, social popularity, and so on (Jung, Lennon &
Rudd, 2001). Self-discrepancy theory assumed that individuals had a tendency to
compare their perceived selves with an idealized self and this comparison process
could lead to negative psychological states and corresponding behavior (Thompson,
1990; Heinberg & Thompson, 1995). Researchers found that when the discrepancy
between one’s ideal body and actual perceived body becomes larger it may be
associated with psychological depression and may lead to eating disorders (Connell,
Ulrich, Brannon, Alexander, & Presley, 2006; Jacobi & Cash, 1994). Cash and
Szymanski (1995) mentioned that individuals were inclined to attain a match between
their actual self-concept and internal ideal. Rudd and Lennon (1994) also suggest that
people have the tendency to use coping strategies to reduce the self-discrepancy.
The self-discrepancy has been measured by a number of researchers.
Strauman and Higgins (1987) developed the Self-Discrepancy Questionnaire.
Individuals were asked to list the traits related to their actual self and to rate the extent
to which they and their ideal self possesses the trait. The ideal rating was compared
with the actual rating to generate a discrepancy score. Fitzgibbon, Blackman and
Avellone (2000) used a figure rating scale which contained nine schematic figures of
women to measure 389 participants’ actual perceived size and their ideal size.
Participants were asked to select their actual size and their ideal size from the
17
schematic figures. Cash and Szymanski (1995) developed the widely used Body Image
Ideal Questionnaire (BIQ) for accessing self-ideal discrepancies. The BIQ asked
participants to rate the discrepancy between ideal selves and actual selves on eleven
attributes, such as facial features, skin complexion, height, weight, chest size, hair
thickness, muscle tone and definition, body proportions, physical strength, and
physical coordination.
In the present study, self-discrepancy theory is used to explain consumers’
fit evaluation process in an online apparel shopping context. Previous studies have
found that apparel fit judgments are closely linked to peoples' feelings about
themselves and the image they wish to project (Hammond & Kohler, 2000; Kim &
Damhorst, 2010). When a female consumer tries to buy a garment, she needs to
evaluate how the garment will fit her body, such as “Does the garment fit me or look
good on me”, and also think about if the garment presents a proper image to others,
such as “Do I look positive with this garment”. Obviously, this female consumer not
only needs to perceive herself in terms of her body but also needs to compare her
perceived self appearance with an imagined ideal self. In an online shopping context,
the consumer’s self-ideal comparison process is “visualized” mainly in the consumer’s
imagination with some external assistance (i.e. visual image of garment and verbal
instructions), while in an offline shopping context, this comparison could be visualized
in a mirror, or be perceived by feeling the tightness or looseness of the garment. Kim
and Damhorst (2010) used self-discrepancy to examine the relationships among online
consumers’ body image discrepancy, body dissatisfaction, apparel involvement,
concerns with fit and size of garments, and purchase intentions. The results confirmed
18
that perceived body image discrepancies were related to body dissatisfaction and
influenced online consumers’ concerns with fit and size of garments. In the current
study, participants’ visual attention to pictures of garments worn by a human model for
an online shopping scenario was analyzed; the result might provide some clues about
how online consumers “visualize” this self-ideal comparison process.
Body Cathexis
Body cathexis is defined as people’s satisfaction or dissatisfaction with
various parts of their body. (Mahoney & Finch, 1976; Rosen & Ross, 1973; Secord &
Jourard, 1953). The original body cathexis scale in Secord and Jourard (1953)
contained 46 items to measure the satisfaction or dissatisfaction toward physical
characteristics on a seven-point Likert scale. Rosen and Ross (1973) developed the
most frequently used body cathexis scale to measure an individual’s body cathexis on
a nine-point Likert scale. The other frequently used instruments to measure people’s
satisfaction toward their body is the Body Areas Satisfaction Scale (BASS), a subscale
of the Multidimensional Body-Self Relations Questionnaire (MBSRQ: Cash &
Strachan, 1999; Cash, 1995). LaBat and DeLong (1990) used the scale developed by
Rosen and Ross (1973) to study relationships between consumers’ satisfaction with fit
and body cathexis. Their results showed that body cathexis was an important factor
related to consumers’ satisfaction with fit. The researchers also found that women
were dissatisfied with the fit of RTW (ready to wear clothing) for the lower body (pant
length, crotch, thigh, hip, and buttocks) and relatively satisfied with upper body fit.
19
Likewise, they were more satisfied with their upper bodies and less satisfied with their
lower bodies.
Markee, Carey, and Pedersen’s (1990) study showed that clothing
functions to improve an individual’s body cathexis because women were more
satisfied with their bodies when they wore clothes. Apparel fit could also have an
effect on consumers’ body cathexis. Apparel companies use an idealized body type to
develop and present their products. When women do not see themselves fitting that
particular standard, they blame their bodies rather than the clothes for improper fit
(LaBat & DeLong, 1990). It is reasonable to expect that body cathexis could affect the
consumers” evaluation of fit and even have an effect on consumers’ visual perception
of garments.
Body Image Discrepancy
Cash (1990) defined body image as the perception of one’s body and the
feelings about the perception. A more recent view is that body image is a
psychological construct that includes perceptual, affective, and cognitive and
behavioral components (Thompson, Heinberg, Altabe, & Tantleff-Dunn, 1999, p. 9).
The perceptual component measures tap how people perceive themselves in terms of
their bodies, such as weight and body shape (Brown, Cash, & Mikulka, 1990; Cash,
1994; Cash & Szymanski, 1995; Jung & Lennon, 2003). The affective component taps
the feelings (i.e. upset, distressed, or anxious) about one’s own appearance; the
cognitive component reflects the interpretation of certain appearance features; a
behavioral aspect might include the avoidance of situations that related to body image
20
(i.e. swimming, wearing revealing clothes) or performance (e.g., exercising, dieting) of
certain behaviors related to body image (Thompson, Heinberg, Altabe & Tantleff-
Dunn, 1999, p9).
Body image discrepancy was derived from self-discrepancy theory
(Higgins, 1987), which assumed that individuals have the tendency to compare their
real selves (i.e. perceived body shape or size) with an idealized self (i.e. the ideal body
shape or size they aspire to). The result of this comparison process is thought to lead to
a discrepancy between the perceived self and the ideal self (Thompson, Heinberg,
Altabe & Tantleff-Dunn, 1999, p134). Researchers have confirmed the existence of
body image discrepancy and also found that body image discrepancy could be an
important predictor of body image dissatisfaction and eating disorders (Altabe &
Thompson, 1992; Cash & Deagle, 1997; Chen & Swalm, 1998; Jacobi & Cash, 1994).
To tap body image discrepancy, Cash and Szymanski (1995) developed
the Body-Image Ideals Questionnaire (BIQ). The BIQ scale measures body image
discrepancy and includes two aspects: (1) the extent to which an individual believes
that his/her physical characteristics match his/her physical ideals, and (2) the
importance associated with having or attaining those ideals. Cash (1994) indicated that
body image discrepancy could influence people’s affective, cognitive and appearance
related behaviors. Kozar and Damhorst (2009) sampled 281 women between the ages
of 30 and 80 to examine the relationship between women’s age identity, body image
discrepancy, and tendency to compare themselves with fashion models. The result
indicated that respondents who reported a larger difference between their cognitive and
actual ages tended to have larger perceived appearance discrepancy and lower body
21
satisfaction, and were more willing to compare themselves with fashion models. Kim
and Damhorst (2010) found that the discrepancy between the perceived body of a
consumer and the body of an ideal apparel model depicted on a retail website
increased the consumer’s body dissatisfaction and hence influenced the concern with
fit and loyalty intention.
Visual Attention, Eye Movement and Eye Tracker
When people attend to visual stimuli they focus their mental capacities on
the selection of sensory input so that the mind can successfully process the visual
stimulus of interest (Rayner, 1998; Duchowski, 2007). There are many models and
terms developed for describing visual attention. The basic words “where” and “what”
roughly describe the selective nature of visual attention (Duchowski, 2007). The
“where” represents the visual selection of specific regions of interest from the entire
visual stimulus for detailed inspection (Von Helmholtz, 1925). The “what” of visual
attention corresponds to the detailed inspection of the spatial region through a
perceptual channel (James, 1981). For example, when considering a visual image,
people are attracted to certain regions in the image over others. The located image
features may attract people’s attention to the places “where” they need to look, and
then they may identify “what” detail is present at those locations. The dual “what” and
“where” feature-driven view of vision is a simplistic but useful way to analyze visual
attention. To gain insight into viewers’ visual attention, analysis of viewers’ eye
movement is one of the commonly used methods. The measurement device most often
22
used for analyzing human’s visual attention is an eye-tracker, which can measure
viewers’ eye movement when they see a visual stimulus.
Fixations are the most frequently analyzed eye movement in various eye
tracking applications (Ju & Johnson, 2010; Nguyen, Isaacowitz, & Rubin, 2009) and
visual attention studies (Pieters & Wedel, 2004). Fixation is an eye movement that
stabilizes the retina over a stationary object of interest; it corresponds to the desire to
maintain one’s gaze on an object of interest (Duchowski, 2007). Fixation is an eye
movement that will be analyzed in this study. According to Jacob and Karn (2003), the
most commonly recorded eye tracking metrics are: number of fixations over the whole
image, number of fixations over a specific visual area, fixation duration time of overall
image, fixation duration time of a specific visual area. Both the number of fixations
and fixation duration time are indicators of visual attention load (longer duration and
larger numbers of fixations indicate an increase in visual attention) (Duchowski,
2007). In the present research, the numbers and duration time of fixations over a static
photo of human models will be examined to identify the body sites that receive the
most attention when consumers make fit judgments.
An eye tracker is a measurement device used for measuring eye movement
There are four categories of eye movement measurement methods: Electro-Oculo-
Graphy (EOG), sclera contact lens coil, Photo-Oculo-Graphy (POG), and video-based
pupil and corneal reflection (Duchowski, 2007). The first three methodologies have
been developed and widely used for decades and are invasive measurements. The
video-based eye tracker utilizes relatively inexpensive cameras to measure the visible
features of the eye, such as iris-sclera boundary, and it is the most non-invasive
23
method of assessing eye tracking. By using an eye tracker, eye movements of people
can be generally recorded to ascertain their attention patterns over a visual stimulus.
The eye tracking technology provides objective and quantitative evidence of a person’s
visual and attention processes (Duchowski, 2007). According to Duchowski, the eye
tracker has been widely used in diverse domains related to human visual attention,
such as psychology, linguistics, advertising, computer science, and even
ophthalmology.
For example, in the advertising area, eye tracking can provide insight into
how the consumer disperses visual attention over different forms of advertising. In a
recent study of eye movements over advertisements, Wedel and Pieters (2000)
reported that fixations to the picture and the brand systematically promoted accurate
brand memory but that text fixations did not. Results showed that the more
information that was extracted from an ad during fixations, the shorter the time for
recollection of the brand. An interesting example in ophthalmology is that people’s eye
movements are recorded by a video-based eye tracker to find the facial cues used by
people when they make judgments about how old or tired a face appears (Nguyen,
Isaacowitz, & Rubin, 2009). The results demonstrated that participants spent the most
time looking at the eye region, which means people tend to use the eye region as first
criteria for their judgments of oldness and tiredness. A recent application in fashion
research is in the study of Ju and Johnson (2010), who used use eye tracking
technology to measure to what extent young women, focused their attention on models
in fashion advertisements. The results suggested that young women did pay most
24
attention to the model because they tend to compare themselves to the models depicted
in the advertisements.
The motivation for utilizing an eye tracker in the present study is to record
and analyze consumers’ visual fixations on the visual image of a garment on a human
model when they make judgments of garment fit. The result of this study may provide
some implication for understanding how consumers perceive visual information for
making fit judgments when they buy clothes online.
Negative Psychological State on Visual Attention
Schwarz, Bless and Bohner (1991) proposed that people in a negative
psychological state tended to use more cognitive effort to process relevant information
than people in a positive psychological state, because they were more motivated than
their counterparts. In the present study, this theory was applied to explain the effect of
body image discrepancy and body dissatisfaction on consumers’ visual attention for
judging garment fit. According to self-discrepancy theory, females who had large body
image discrepancy and were less dissatisfied with their body might experience a
negative psychological state. In addition, Kuo, Hsu and Day (2009) used the duration
of fixations and numbers of fixation obtained from eye-tracking technology to indicate
people’s cognitive effort in processing information for making judgments or decisions.
The results indicated that respondents under negative mood tended to fixate longer and
more frequently at the visual image in making a judgment than their counterparts
under positive mood. Hence, it is reasonable to expect that those females who had
large body image discrepancy and less body satisfaction would have the longer
25
fixation duration and a larger number of fixations on the human model when they
made judgments of fit of a garment.
Consumers’ Concern with Fit
Fit refers to how well the garment conforms to the 3D human body. Good
fit is crucial to consumer satisfaction. Fit may be evaluated at many different points in
the product life cycle for apparel manufacturers, retailers, and consumers (Brown &
Rice, 2000; Goldsberry, Shim, & Reich; 1996). Five classic elements for evaluating fit
are: (1) grain, (2) set, (3) line, (4) balance, and (5) ease (Brown & Rice, 2000; Erwin,
Kinchen, & Peters, 1979). For apparel manufacturers, all areas of the garment must be
examined when evaluating the fit of a garment. Some body sites are more important,
such as shoulder, bust, neckline, collar, lapel, armscyes, sleeves, waistline, hips,
crotch, and garment length (Brown & Rice, 2000). For a female consumer, her
judgment of fit is based on visual feedback (Delong, Kim, & Larntz, 1993) and tactile
perceptions (Ashdown & Delong, 1995). The tactile perception could come from the
pressure sensations felt by the nerves in skin, muscles and joints (McBurney &
Collins, 1977).
In an online shopping context, it is impossible to obtain tactile sensory
information based on current technology. Hence, the visual image of the garment has
become the major source for consumers to visualize the proper fit of a garment during
online shopping. Hence, when consumers purchase garments online, they have high
levels of concern with size and fit, because they lack some necessary sensory
information (e.g., pressure sensation) to evaluate garment fit. They make purchase
26
judgments based on the limited visual and verbal information about the garments
presented on the website. The lack of necessary information about size and fit has
become a major perceived risk for purchasing apparel online (Forsythe, Liu, Shannon,
& Gardner, 2006).
Researchers defined online apparel shoppers’ concerns with fit and size
“as the subjectively determined expectations and amount of risk perceived by a
shopper in relation to the fit and size of the garment in contemplating a particular
purchase decision” (Kim & Damhorst 2010, p. 242). They also developed an
instrument for measuring concern with fit and size, which included 5 factors: concerns
with overall appearance, concerns with inability to try on in online shopping, concerns
with projecting a correct impression, concerns with unavailability of size, concerns
with imagining fit/size in online shopping. The concern with fit and size scale was
tested among 348 female college students using a web-based survey, the results
showed that two factors (concern with overall fit and concern with imagining fit/size
in online shopping) had negative relationships with consumers’ online purchase
intention. The concern with overall fit included the item, “The garment may not look
good on me”; the concern with imagining fit/size in online shopping included item
“Shopping in the website, I may have a hard time picturing myself wearing the
garment”. These items suggest that the lack of necessary visual information displaying
how garments look on consumers’ bodies was the major source of perceived risk for
buying clothes online. Hence, there is a need for online retailers to develop new visual
presentation techniques to visualize how well garment fit conforms to a consumer’s
body and how garments of various sizes drape differently on consumers’ bodies. This
27
new technology may decrease consumers’ concern with fit and size, increase their
confidence shopping online, and finally increase their willingness to purchase online.
Confidence in Fit Judgment
Sniezek (1992) stated that confidence in judgment reflected an
individual’s strength of belief about the quality of a judgment or choice. People’s
confidence or uncertainty about a judgment determines whether the people would act
upon the judgment (Sniezek & Henry, 1989). Laroche, Kim and Zhou (1996) revealed
that consumers’ confidence about their selection of a brand had a positive impact on
their intention to purchase a product from that brand. Peterson and Pitz (1988)
demonstrated that confidence and uncertainty of a judgment or selection were affected
by the available information. When the amount of information was manipulated within
subjects, the confidence in judgment or selection increased as more information
become available. When the amount of information was manipulated between
subjects, the effect of amount of information on confidence in judgment was not
observable. According to Sicilia and Ruiz (2010), the amount of information has
effects on consumers’ selection and attitudes toward the product presented on the
website. In terms of online apparel shopping, Park, Stoel and Lennon (2008) indicated
that consumers relied on the visual and verbal information presented on the website to
make their purchase decision in online shopping.
Urbany, Dickson, and Wilkie (1989) divided consumer’s confidence into
two categories, knowledge confidence and choice confidence. Knowledge confidence
means how certain a consumer is regarding what is known about the attributes of a
28
product. Choice confidence represents how confident a consumer is to make a choice
among alternatives. Based on this conceptualization, the confidence in judgment of
garment fit would include two aspects, how certain an individual is regarding her
knowledge of fit and how confident an individual is that the selected garment would fit
the body. For apparel manufacturers and apparel researchers, panels of expert judges
and trained paraprofessional judges have frequently been used to judge garment fit
(Ashdown & O'Connell, 2006). The standard knowledge framework accepted by
apparel manufacturers and researchers for judging fit was usually developed based on
the five components of fit: ease, line, grain, balance, and set (Erwin, Kinchen, &
Peters, 1979). However, there is little research about what kind of judging rule is
generally used by consumers. Since apparel consumers usually learn information about
products from apparel manufacturers and sales people (Häubl & Trifts, 2000), they
may have adopted part of theses rules from daily interaction with apparel salespersons.
According to Urbany et al. (1989), confidence in choice is related to the perceived
information for making a judgment. Ashdown and O'Connell (2006) mentioned that
experts made judgments of fit by visually assessing a garment on a body; the visual
perceptual assessment was a common strategy used to judge the fit of apparel products
in apparel product development and research. Hence, the confidence in choice in
garment fit judgments should be highly related to the visual information perceived
from the garment. In an online shopping context, the visual presentation of the
garment is the major source of visual information for judgment of garment fit. Even
though some websites have size charts which can help consumers select the right size
of a garment, consumers still need to picture the size of the garment in their minds
29
based on the text information of the size charts. Some consumers might experience
difficulty in picturing the size of the garment in their minds (Kim & Damhorst, 2010).
Obviously, the perceived visual information used in fit judgments in online apparel
shopping is very limited. Hence, online shoppers feel uncertain about their garment fit
judgments and have a high level of concern with garment fit. Finally, consumers’
willingness to buy garments online could be affected (Rosa, Garbarino & Malter,
2006).
Purchase Intention
Purchase intention is the last stage in a consumer’s decision-making
process for an apparel product (Chen-Yu & Kincade, 2001). Purchase intention
reflects “what we think we will buy” (Blackwell, Miniard, & Engel, 2001, p. 283). In
the online apparel shopping context, the brand familiarity, and apparel information
presented on a website could affect consumers’ purchase intention for apparel (Kim,
Fiore, & Lee, 2007;Park & Stoel, 2005; Szymanski & Hise, 2000). Researchers also
found that consumers’ perceived risk toward the product and online retailer influenced
consumers’ intention to shop online (Jarvenpaa & Todd, 1997). Further, apparel
products were considered to be more risky than books and software by consumers in
online shopping (Bhatnagar et al., 2000). The previous studies suggest that inability to
try on apparel in an online shopping context was one of the major reasons for
perceived risk which in turn affected consumers’ purchase intention.
Previous research has found that consumers’ willingness to purchase
garments online could be affected by the methods used to present product information,
30
such as visual image, verbal message, and rotating garment display. Chen-Yu and
Kincade (2001) found that online consumers made a purchase decision by gathering
information about garment quality and fit through what was displayed on the screen.
Eroglu, Machleit, and Davis (2001) proposed an online shopping model that suggested
that verbal information about size, price, customer service, and visual image affected
consumers’ purchase decisions. Furthermore, Chau and Tam (2000) argued that visual
images were more effective and efficient in communicating information compared to
verbal messages in the online shopping context. Researchers found that providing a
better visual display could reduce perceived risk and increase consumers’ intention to
purchase (Park, Lennon, & Stoel, 2005). Then and DeLong (1999) found that a large
percent of their respondents preferred that apparel be displayed on a realistic human
model rather than displayed on a mannequin or a flat surface so they could visualize
the garment silhouette. Park, Stoel and Lennon (2008) examined the effect of a
rotational display on consumers’ response; the results indicated that the rotational
product display decreased perceived risk and increased purchase intent.
Khakimdjanova and Park (2005) suggested that online retailers could increase
consumer’s interest in purchasing by adding visual features to their website, such as a
whole view of the model, alternative body shapes for a mannequin and alternative
human model poses. Therefore, product information, such as verbal information, a
visual image, displaying garments on human model, and rotating images could be
important for reducing consumers’ perceived risk and facilitating online shopping.
Previous studies also found that consumer’s beliefs and attitudes toward
online shopping and their previous online shopping history had significant effects on
31
their current intention to buy products from an online retailer (Lee & Park, 2009). The
more positive people’s previous experience, the higher the possibility they would like
to engage in online shopping activities (Szymanski & Hise, 2000; Park, Lennon, &
Stoel, 2005).
In addition, the Internet has also been widely used by consumers as a
source for product information (McQuitty & Peterson, 2000). Based on survey data
collected from 706 respondents who had who resided in 15 U.S. metropolitan areas
and had access to a computer at home or work, researchers found that consumers’
intention to search for product information online was not only a strong predictor of
online purchase intention but also a mediator between purchase intention, attitude
toward online shopping and previous Internet purchase history (Shim, Eastlick, Lotz &
Warrington; 2001).
Body Cathexis and Body Image Discrepancy
Body cathexis reflects people’s satisfaction or dissatisfaction for the body.
Body image discrepancy means the differences between individual’s real perceived
body and an idealized one. According to self-discrepancy theory, the discrepancy
between a real self and an idealized self could lead to negative psychological state and
corresponding behavior (Thompson, Heinberg, Altabe & Tantleff-Dunn, 1999). Since
people’s body cathexis is a kind of psychological state toward their body, it is
reasonable to infer that people’s body image discrepancy could predict their body
cathexis. The larger body image discrepancy might lead to higher level of body
dissatisfaction. Previous research confirmed that body image discrepancy could be an
32
important predictor for body dissatisfaction and eating disorders (Jacobi & Cash,
1994; Altabe & Thompson, 1992). Williamson, Gleaves, Watkins and Schlundt (1993)
even conceptualized body dissatisfaction as the discrepancy between self and ideal
body size estimates. A sample of 110 women were instructed to estimate their current
body size and their ideal body size by making a selection from nine silhouettes of
different body sizes (from very thin to very obese) on separate cards. Individuals’ body
dissatisfaction was measured by a Body Dissatisfaction Scale (Garner, Olmstead &
Polivy, 1983). The results showed that measures of self-ideal body size discrepancy
were positively correlated with measures of body dissatisfaction. Kim and Damhorst
(2010) used the Body Image Ideal Questionnaire (Cash & Szymanski, 1995) to
measure 348 college women’s body image discrepancy and tapped their body
dissatisfaction in items borrowed from Heinberg and Thompson (1995). Their result
indicated that consumers’ perceived body image discrepancy was positively related to
body dissatisfaction. This rationale led to the following hypothesis.
Hypothesis 1: Participants’ total body image discrepancy will be
negatively related to their total body satisfaction.
Body Cathexis, Body Image Discrepancy and Visual Perception
Garment fit may be closely related to body cathexis, because garments
have the potential to improve the people’s body shape and enhance their perception
and feelings about the body (Markee, Carey & Pedersen, 1990; Petrie, Tripp, &
Harvey, 2000; Wendel & Lester,1988). Consumers’ fit satisfaction might be related to
their body satisfaction. LaBat and DeLong (1990) indicated that body cathexis is an
33
important predictor for consumers’ satisfaction toward garment fit. The researchers
found that female consumers were more satisfied with their upper bodies than with
their lower bodies; correspondingly, their satisfaction with the fit of RTW for the
upper body was higher than their satisfaction with fit for lower body.
In addition, previous studies found that consumers’ perceptions of garment
fit are closely related to people’s feelings about themselves and the image they wish to
project (Hammond & Kohler, 2000; Kim & Damhorst, 2010). Consumers not only
need to perceive the visual information about how the garment conforms to the body,
they also think about if the garment would present a proper image to others. Self-
discrepancy could be used to explain this internal visualization process. In a real-store
environment, a female consumer usually perceives the visual information about how
the garment is displayed on her body from a mirror and internally compares this visual
image with an imagined ideal image about how this garment should look on her body.
In an online apparel shopping context, the consumer perceives the visual information
of the garment itself, uses the visual information to picture an image in her mind about
how the garment would display on her body, and finally compares this image with the
ideal look she would like to project. Based on self-discrepancy theory, Kim and
Damhorst (2010) found that individuals who had a large discrepancy between their
real body and ideal body were more dissatisfied with their bodies and finally felt a
higher degree of concern with fit and size of garments. Their instrument included the
items, such as “The garment may not look good on me” and “Shopping in the website,
I may have a hard time picturing myself wearing the garment”. Thus, this research
34
demonstrates that consumers’ visual perception of garment of fit is related to their
body image discrepancy.
Therefore, it is reasonable to expect that individuals who have a large
body image discrepancy and higher level of body dissatisfaction might need to look
longer to obtain enough information to assess themselves in a garment and hence their
visual perception process for a garment might take more time than those with less
discrepancy and less dissatisfaction. With the assistance of an eye-tracker, researchers
can study people’s visual perception of visual stimuli by accurately measuring the
visual attention over them (Keith, 2009). The eye tracker can capture and measure the
fixations, which is an eye movement that stabilizes the retina over a stationary object
of interest. The number of fixations and the duration of fixations reveal how much
visual attention is placed on visual stimuli. Longer duration and more fixations reflect
an increase in visual attention.
Some researchers use the eye tracker to study consumer behavior. For
example, Pieters and Warlop (1999) used eye tracking data to analyze consumers’
perceptual process of brand choice. The research found that consumers accelerated
visual information acquisition under high time pressure, because they decreased the
duration of eye fixations on the stimuli. Ju and Johnson (2010), who used use an eye
tracker to measure to what extent young women, focused their attention on models in
fashion advertisements. They found that young women who tended to compare
themselves to the models depicted in the advertisements paid most attention to the
model.
35
Kuo, Hsu and Day (2009) used the duration of fixations and numbers of
fixations obtained from eye-tracking technology to indicate people’s cognitive effort in
processing information for making judgments or decisions. They found that
respondents under negative mood were prone to take more cognitive effort (longer
duration of fixations and greater number of fixations) in making a judgment than their
counterparts under positive mood. According to self-discrepancy theory, consumers
who have a large body image discrepancy and a high level of body dissatisfaction may
experience a negative psychological state. Hence, it is reasonable to infer that
consumers’ negative emotion induced by a large body image discrepancy and a high
level of body dissatisfaction would lead to longer durations and more fixations on the
visual display of a human model wearing a garment when consumers evaluate the fit
of the garment. Therefore, the following hypotheses were developed.
Hypothesis 2a: Participants’ total body image discrepancy will be
positively related to the total duration of fixations over the human models when
making judgments of fit. (i.e., Consumers will look longer at the human model if
they havea large body image discrepancy.)
Hypothesis 2b: Participants’ total body image discrepancy will be
positively related to the total number of fixations over the human model when
making judgments of fit. (i.e., Consumers will look more often at the human model
if they have a large body image discrepancy.)
Hypothesis 3a: Participants’ total body satisfaction will be negatively
related to the total duration of fixations over the human model when making
36
judgments of fit. (i.e., Consumers will look longer at the human model if they are
less satisfied with their bodies.)
Hypothesis 3b: Participants’ total body satisfaction will be negatively
related to the total number of fixations over the human model when making
judgments of fit. (i.e., Consumers will look more often at the human model if they
are less satisfied with their bodies.)
Visual Perception of Garment Fit and Concern with Fit
Online shoppers perceive visual information about garment fit from the
visual display of garments presented on websites. Park, Stoel and Lennon (2008)
indicated that visual display of garments was a crucial factor for consumer’s selection
of garments in the online shopping context. Then and Delong (1999) found that
consumers preferred a visual image of a garment displayed on a human model over a
visual image of a garment displayed on a mannequin or a flat surface. They also found
that consumers felt less uncertain about purchasing apparel online when the garment
was displayed on a human model. The results imply that consumers may perceive
more information about fit from the picture of a garment wore by a human model,
which could reduce consumers’ perceived risk for buying garment online.
Kim and Damhorst (2010) defined concern with fit as the perceived risk
related to fit of a garment when making a purchase decision. The researchers
constructed this concept with five aspects: concerns with overall appearance, concerns
with product performance, concerns with unavailability of size, concerns with
projecting a correct impression, and concerns with imaging fit online. They found that
37
only two aspects, concern with overall appearance and concern with imagining fit
online was negatively related to consumer’s apparel purchasing decision. The results
may imply that the consumers who have difficulty imagining their overall appearance
based on the visual information perceived from the visual display of a garment might
need more information to feel confident about choosing a well fitted garment online.
Park and Stoel (2002) suggested that sensory (i.e., visual and tactile) information such
as garment draping effect and fabric construction was helpful to reduce perceived risk
for online apparel shoppers. Consumers who obtained more visual information from a
visual display of a garment on the website would have less concern with fit when
buying clothes online. Hence, it is reasonable to propose that consumers who look
longer and more often at the visual display of a garment might obtain more
information and reduce their concern with fit.
Based on the previous discussion, an eye tracker could help researchers
analyze the process of perceiving visual information from visual stimuli. The number
of fixations and the duration of fixations indicate how much visual attention is loaded
on visual stimuli. Longer duration and more fixations reveal people place more
attention on a visual image and might obtain more information from the visual image.
Therefore, the following hypotheses were developed.
Hypothesis 4-a: Duration of fixations on a human model will be
negatively related to concern with fit (i.e., Consumers who look longer at the photos
of a human model wearing blazers have less concern with fit.)
38
Hypothesis 4-b: Number of fixations on a human model will be
negatively related to concern with fit (i.e., Consumers who look more often at the
photos of a human model wearing blazers have less concern with fit.)
Concern with Fit and Confidence in Fit Judgment
Concern with fit was defined as the perceived risk related to fit of a
garment when making a purchase decision (Kim & Damhorst, 2010). Based on the
definition of knowledge confidence and choice confidence in Urbany et al. (1989),
consumers’ confidence in fit judgment refers to how certain an individual is regarding
her knowledge of fit and how confident she believes the selected garment would fit the
body. Researchers found that consumers’ confidence for selecting a product was
related to consumers’ previous shopping experience (Laroche et al., 1996), the amount
of information available about the product (Peterson & Pitz, 1988; Urbany et al.,
1989), and the perceived risk relevant to purchasing the product (Cox & Stuart, 1964;
Bennett & Harrell, 1975). Perceived risk researchers (Cox & Stuart, 1964) mentioned
that the perceived risk for buyers was closely related to the degree of confidence the
consumer felt about his/her ability to judge the outcome of making a purchase. In the
case of buying garments online, consumers need to judge the fit of the garment before
they buy it. Hence the perceived risk for buying a garment would be related to the level
of confidence a consumer has in his/her ability to make a correct fit evaluation.
Therefore, the following hypothesis was developed.
Hypothesis 5: Consumers’ concern with garment fit will be negatively
related to their confidence in fit judgments.
39
Concerns with Fit and Purchase Intent
Concern with fit is an important issue for online apparel shoppers, because
they are not able to try the garments on before they purchase. Thus, they are not able to
obtain enough information about how the garment will look on their body and whether
the garment would present the proper image they desire (Rosa, Garbarino & Malter,
2006). The lack of enough information about garment fit increases the perceived risk
for online apparel shopping. Hence, concern with fit was conceptualized as
subjectively perceived risk related to fit of a garment when making a purchase decision
(Kim & Damhorst, 2010). These researchers found that the online apparel shoppers’
concern with fit was negatively related to consumers’ intention to make a purchase
online. Other researchers found similar results that the concern with fit of garments
due to the inability to examine the garment online had a negative impact on
consumers’ online shopping behavior (Kim & Lennon, 2010; Rosa et al., 2006). The
concern with garment fit inhibits the consumer from making a purchase decision.
Therefore, the following hypothesis was developed.
Hypothesis 6: Consumers’ concern with fit will be negatively related to
their purchase intent.
Confidence in Judgment and Purchase Intention
A certain number of researchers have indicated that consumers’
confidence in judgment about attributes of products or brands affect their purchase
intent (Cox & Stuart, 1964; Howard & Sheth, 1969; Peterson & Pitz, 1988; Urbany et
al. 1989) Laroche, Kim and Zhou (1996) stated that intention to buy a specific product,
such as cough medications, will be positively affected by a consumer's knowledge
40
confidence and choice confidence toward a product. Bennett and Harrell (1975) used
survey methodology to examine the role of consumers’ confidence to judge attributes
of a product in the formation of their purchase intention; the results demonstrated a
positive relationship between people’s confidence in their ability to judge attributes of
a product and intention to purchase the product. Fit is an important attribute of a
garment which is necessarily considered by consumers before making a purchase
(Ashdown & O'Connell, 2006). Consumers judge garment fit based on tactile
feedback, how the garment feels on the body, and visual feedback, how the garment
looks on the body (Ashdown & Delong, 1995). When consumers buy clothes online,
they judge garment fit based on the visual and verbal information presented on the
website (Park et al., 2008). Being afraid to buy a badly fitted garment prevents people
from buying clothes online (Beck, 2003). Hence, consumers’ confidence of fit
judgment would be an important predictor for their intention to purchase clothes.
Therefore, the following hypothesis was developed.
Hypothesis 7: Consumers’ confidence with fit judgments will be
positively related to their online purchase intent.
Effect of Body Sites on Visual Perception
Human visual attention researchers found that people are more easily
attracted by salient features in a visual image, such as vivid color, contour, shape, and
texture (Kanwisher & Driver, 1992). A garment hanging on a human body may display
various winkles; wrinkles at some specific body sites, such as bust and chest, are
important features for apparel experts to judge the fit of the garment (Brown & Rice,
2001). This implies that some body sites may offer salient features which could draw
41
more attention in people’s visual perception of fit. In addition, researchers also find
that people had different body cathexis scores over the body sites, which means people
are more satisfied with some specific body sites than the others. Labat and Delong
(1990) investigated 107 female college students and found that these consumers were
more satisfied with their upper body sites, such as neck and shoulder, and less satisfied
with their lower bodies, such as hip and buttock. Furthermore, these female
consumers’ satisfaction with the fit of RTW garments was correspondingly varied.
Since consumers’ perception of garment fit is based on tactile response and visual
information (Ashdown & Delong, 1995), people’s satisfaction with specific body sites
may affect this perception process. The following hypotheses were developed.
Hypothesis 8-a: There is a significant difference among the body sites
of the human model on duration of fixations.
Hypothesis 8-b: There is a significant difference among body sites of the
human model on number of fixations.
Figure 2.1 Theoretical framework of current study
Body image
Discrepancy
Body
Satisfaction
Visual Attention
Fixation
Duration
Fixation
Number
Concern
with Fit
Confidence
of Judgment
Purchase
Intention H1
H2
H3
H4 H5
H6
H7
Body
Sites
H8
42
Chapter 3
METHOD
Sample
Data were collected from a convenience sample of female college students
from a medium-sized mid-Atlantic university. The students were recruited by making
announcements on classes. The recruiting announcements were made in three classes
of fashion majors. The course instructor offered some extra credit for students who
participated in this research. The researcher made individual appointments for each
volunteer participant to complete the research tasks. College students were chosen as
participants, because they use the internet frequently and they are likely customers for
online apparel retailing (Lee & Lin, 2005). Most college students tend to use the
Internet for gaining product information before purchasing a product and also use the
Internet to purchase products (Kim et al., 2007; Yoo, Lee, & Park, 2010).
Stimuli
A business blazer was chosen as the garment stimulus, because it covers
most of the body, is more fitted than other types of garments worn on the upper body,
and fits closer to the body than some other apparel items, such as sweaters or t-shirts.
Some researchers have used the business blazer as a typical garment silhouette for
understanding size (Ashdown, & Dunne 2006), to assess perception of garment
43
proportion (Delong, Kim, & Larntz, 1993), and to assess fit preference (Alexander,
Connell, & Presley, 2005). For example, Delong et al. (1993) tested 172 undergraduate
women’s responses to twelve sets of blazer silhouettes varying proportionally in
components of lapel, yoke, and pocket details. The research results indicated that
participants detected differences in proportions of garment details within the context of
blazer silhouettes and preferred proportions adjusted to smaller sizes.
For the current study, four business blazers in the same color (dark grey),
fabric (made from 80% wool, 19% cotton, and 1% spandex), and style (notched lapel)
but different sizes (2, 4, 6 and 8) were purchased from a major fashion brand that
targets young female consumers. This set of blazers was selected so that the only
variation among the blazers was the size. The purpose of using four different sizes of
blazers in this study is to simulate the situation in a real fitting room where consumers
try on garments of different sizes and observe how the garments look from the mirror.
A size 4 female college student was recruited to model the four blazers in photographs.
She wore the same white blouse and black high heel pumps with each size of blazer.
No jewelry and other accessories were worn by the model to avoid introducing other
factors that could affect participant responses. One front-view photo and one back-
view photo were taken for each size of the blazer which was worn by the model. The
distance between the model and the camera was fixed, so that the four blazers could be
displayed in the same area of the photos. Totally, eight photos were taken. In these
photos, the model’s face was removed, in order to reduce the impact of facial
appearance of the model on people’s perceptions of garments. For example,
researchers have found that consumers who had more positive beliefs about a model's
44
facial appearance indicated higher intention to purchase the model's clothing, and
perceived the model's clothing to be more fashionable (Kozar & Damhorst, 2008;
Kozar, 2010). In addition, the size of the blazers was labeled in the four photos. To
remove the impact of size number on participants’ fit judgments, the numbers (2, 4, 6,
and 8) were replaced with the symbols ( ).The eight photos as visual stimuli
were uploaded to the eye tracker (See Appendix A).
Experimental Design
The whole experiment consisted of three parts. The first part was to test
whether and how each participant’s visual attention to the stimulus garment would
vary across the body sites (e.g., neck, shoulder, bust, waist, arm, abdomen).
Participants’ visual attention (in terms of duration of fixations and number of
fixations) on all the body sites of the human model was recorded by an eye tracker,
when they were asked to make fit judgments of the blazers. Four different sizes of
blazers were used to simulate the scenario that consumers try on garments of different
sizes in a real fitting room. Hence, the body-sites were a within-subjects factor. In this
part of experiment, the participants’ visual attention was the dependent variable, the
body sites of human model was used as the independent variable. Participants were
also requested to complete the measures of concern with fit, confidence of fit
judgment, and purchase intention, after they viewed the photoes of each blazer. Scores
on these measures were used to test whether participant’s overall visual attention was
related to their concern with fit, and to examine the relationship among these variables.
45
In the second part of the experiment, participants completed measures of
body cathexis and body discrepancy. Scores on these measures were used to examine
the relationship between the two variables and to examine whether these variables
could be related to participants’ visual attention on blazers.
Instruments
The Body cathexis scale (See Appendix B) was adapted from the body
cathexis scale used in the study of Labat and Delong (1990). The body cathexis scale
used in Labat and Delong’s (1990) research used a nine-point Likert-type scale (1=
very dissatisfied, 9=very satisfied) measuring feelings of satisfaction with 22 body
sites for upper body, lower body, and total body. In Labat and Delong’s (1990)
research the researchers examined the correlation between female consumers’ body
cathexis and their satisfaction with fit of ready-to-wear garments. The body cathexis at
eight body sites (bust, arm, back, shoulder, hip, thigh, buttock, and abdomen) was
found to have a significant correlation with consumer’s fit satisfaction at
corresponding body sites. This scale was widely used to assess participants’ feeling
about the self and their satisfaction/dissatisfaction with their bodies (Markee, Carey &
Pedersen, 1990; Pisut & Connell, 2007) The reliability of this scale was reported to be
0.83 (Shim, Kotsiopulos, & Knoll, 1991). In the present study, the body cathexis scale
(See. Table 3.1) measured individual’s satisfaction or dissatisfaction with nine body
sites (neck, shoulder, bust, back, waist, abdomen, hip, buttock, and arm). The nine
upper body sites were chosen, because the business blazer is worn on the upper body
and was the garment stimuli used in the current research, and individuals’ body
46
satisfaction with the upper body was the research focus. In addition, seven upper body
sites (bust, arm, back, shoulder, hip, buttock, and abdomen) have been confirmed to be
related to fit satisfaction (Labat & Delong, 1990). Another two body sites (neck, waist)
were important in the judgment of garment (i.e., blazer) fit were on the upper body
(Pisut & Connell, 2007). The overall body cathexis score was calculated by summing
together the cathexis scores for the nine body sites. Higher cathexis scores indicated a
higher level of body satisfaction.
Body Image Discrepancy Scale (See Appendix B). The discrepancy
between participants’ perceived bodies and ideal bodies was measured by a scale
adapted from Body-image Ideals Questionnaire items (Cash & Szymanski, 1995). The
original BIQ assessed internalized ideals and included 10 appearance characteristics
(height, skin complexion, hair texture and thickness, facial features, muscle tone and
definition, body proportions, weight, chest size, physical strength, and physical
coordination). For each characteristic, participants were asked to think of their ideal
for each characteristic (how they wish they were) and then to evaluate how well they
actually match that ideal.
For the present research, the BIQ scale was modified to tap respondent’s
self-ideal discrepancy on nine body sites (neck, shoulder, bust, arm, back, waist, hip,
buttock, and abdomen). The nine relevant body sites were chosen because female
consumers’ perceptions of these body sites either was confirmed to be significant
predictors of their satisfaction with fit of garments (Labat & Delong, 1990), or played
important roles in the process of judging fit of a garment (Ashdown & O'Connell,
2006). Discrepancy between the ideal status and actual status of a body site was
47
measured using a 4-point scale: 0 (exactly as I am), +1 (almost as I am), +2 (fairly
unlike me), and +3 (very unlike me). Importance of each body site was also measured
using a 4-point scale: 0 (not important), 1(somewhat important), 2 (moderately
important), and 3 (very important). The final body image discrepancy was calculated
by multiplying the discrepancy score for each body site by the importance scores of the
same body site and then summing the weighted scores for all nine body sites. Higher
scores indicate higher levels of body image discrepancy. The internal consistency of
original BIQ was reported to be 0.77 in Cash and Szymanski (1995)’s study.
Measurement of visual attention. Participant’s visual attention was
captured by the Tobii ET-17 eye tracker, which is a computer with a camera built into
the monitor. Two major parameters of visual attention, the number of visual fixations
and duration of fixations were measured. Duration time, measured in milliseconds,
refers to the amount of time that a person’s eye stays on a specific area. Number of
fixations refers to the number of times the eye located or hit on the same area.
Duration time and number of fixations are used to indicate how much visual attention
is placed on specific body sites. In order to output the duration time and number of
fixations on each body site, the software of the eye tracker allowed the researcher to
divide the visual stimulus (photo of human model wearing blazers) into areas of
interest (AOI) based on the relevant body sites, including neck, shoulder, bust, waist,
arm, back, abdomen, hip, and buttocks (See Figure 3.1). The specific areas of concern
are the same as assessed using the BIQ and body cathexis. The output of the eye
tracker software included the “hot-spot” graph (See Figure 3.2), “scan-path” graph
(See Figure 3.3), and AOI data files. The number of fixations and duration times were
48
respectively calculated for each AOI and saved in AOI data files by the eye tracker
software. In addition, participant’s overall number of fixations and duration time of the
entire body of human model are also part of the AOI data of the eye tracker.
Figure 3.1 an example of stimulus divided into AOIs
49
Figure 3.2 an example of “hot-spot” graph
50
Figure 3.3 an example of “scan-path” graph
Concern with fit and size scale (See Appendix C). Kim and Damhorst
(2010) developed a scale for tapping consumers’ concerns with fit and size, which
included five latent factors: Concerns with body image and overall appearance,
concerns with product performance, concerns with unavailability of size, concerns
with projecting a correct impression, and concerns with uncertainty about the sizing
system. The Cronbach’s alpha coefficient of the latent factors ranged from .79 to .93.
The study found consumers’ concerns with fit and size of garments was affected by
their body satisfaction, body image self-discrepancy, and influenced their loyalty
intention. In the present research, the three items used in the study of Kim and
Damhorst (2010) were adapted to measure consumers’ concern with fit. The items
include “The blazer of this size may not fit me”, “The blazer of this size may not look
good on me”, and “I may have a hard time picturing myself wearing the garment”. The
51
three items used a five-point Likert format with endpoints of 1= strongly disagree and
5= strongly agree.
Confidence in Judgment scale (See Appendix C). The overall confidence
in fit judgment was measured by five-point scale items (1=not confident/certain at all,
5=very confident/ certain) modified from the two-item confidence in judgment scale
used in (Laroche, Kim & Zhou, 1996). The reliabilities of the two-item measures were
reported from 0.81 to 0.89 for different brands. In the present study, the two items
were “how confident you are about your evaluations of fit of garments” and “to which
extent you are certain about this size of blazer is fit for you”
Purchase intention scale (See Appendix C). Previous studies have shown
that consumers’ body image (Alexander et al., 2005; Rosa et al., 2006; Shim et al.,
1991) and concern for fit (Kim & Damhorst, 2010) contributes to variation in purchase
intention for apparel products. One purchase intent item was adopted from the scale
used in Kim and Lennon (2007) to measure individual’s intention to purchase a blazer
after viewing the visual image of the garment wore by a human model. The item “I
would like to buy a blazer of this size from this retailer” used a five-point Likert
format with endpoints ranging from 1 (strongly disagree) to 5 (strongly agree).
Procedure
The eight photos of four blazers (See Appendix A) were uploaded to the
eye tracker (Tobii ET-17) as the visual stimuli. The Tobii ET-17 eye tracker is a table-
mounted video-based tracker, which includes a computer monitor equipped with a
camera illuminated by an infrared light source and image process hardware and
52
software (See Figure 3.4). The camera and the infrared light source were located inside
the monitor, which was mounted on a table surface about two feet in front of the
participant. The eye tracker is capable of measuring a viewer’s eye movement by
capturing and calculating the relative position between the corneal reflection and pupil
center (Duchowski, 2007).
Participants completed the experimental task individually. The whole
experiment consisted of two parts. In the first part the eye movement of each
participant was recorded when she viewed the visual stimuli and perceived
information to judge the fit of blazers of four sizes. In addition, the items that
measured participant’s concern with fit, confidence in fit judgment, and purchase
intent were also presented on the computer screen immediately following presentation
of the visual stimuli. Participants rated their concern with fit, confidence in fit
judgment and purchase intention for all the four blazers. In the second part of the
experiment, participants completed measures of body cathexis and body image
discrepancy using a paper-based survey. Participant’s body cathexis and body image
discrepancy were only measured once.
Before the experiment started, each participant chose a 5-digit random
number as her identification number in this research. This identification number was
used to combine the data collected from the two parts of the experiment and also
protect participant’s privacy.
First, participants entered identification numbers to the eye-tracker, and
then were instructed to do a calibration to make sure the measurement of the eye
movements was accurate. In the calibration process, the participant was asked to stare
53
at a blue round spot and visually follow the track of the blue spot. At the same time,
the participant’s eye movement was automatically recorded and analyzed by the eye
tracker. The calibration succeeded only when participant’s eye movement was very
close to the track of the blue spot. After completing the calibration, the real experiment
started. Participants were given the following instructions on the computer monitor:
“Please imagine this situation. You need to buy a blazer for a job
interview. However, you are too busy to try some blazers in local stores. One of your
friends has the similar body size as you, and she went to the store to try the clothes for
you. She sent you some photos that were taken when she tried on four blazers of
different sizes. Now, you are going to view these photos and think about if the blazers
will properly fit you. The four sizes of the blazers are size , size , size , and
size . Each size has a front view and a back view. After viewing the photos, you
need to finish a short questionnaire about this blazer. The questionnaire will be
presented on the computer screen; you need to complete the questionnaire using your
eyes. Stare at the option which you believe to be correct about two seconds long, so
that the computer can automatically recognize your choice. Please press the “space”
bar to start the experiment.”
Participants read the instructions and pressed the space bar to browse the
photos of the blazers. After participants finished the ratings of the photos of the
blazers, a short questionnaire appeared on the screen. The questionnaire included three
items for rating each participant’s concern with the fit of the blazer on her body; two
items tapped confidence in her fit judgments regarding the blazers; and one item
assessed purchase intention. Participants completed the items by staring at the selected
54
choice for about two seconds, so that the computer was able to recognize a
participant’s choice. Specifically, each option was divided into an AOI (See Figure
3.5) in the software of eye tracker. The software was able to calculate the duration time
and fixation number on each AOI and finally recognize participant’s choice by
comparing the duration time and fixation numbers of AOIs. For each blazer,
participants completed the same items about her concerns with fit, confidence in fit
judgments, and purchase intention.
In the second part of this study, participants were asked to complete a
paper-based questionnaire (see Appendix A), which included the scales for measuring
participants’ body cathexis and body image discrepancy. Participants also wrote down
the 5-digit identification number on the questionnaire, so that the answers from the
questionnaire could be linked with the eye tracking recording obtained in the first part
of the study.
Figure 3.4 Tobii eye tracker (www.tobii.com)
55
Figure 3.5 an example of a questionnaire item divided into AOIs
56
Chapter 4
ANALYSIS OF RESULTS
SPSS computer programs were used for statistical data analysis.
Descriptive statistical analysis was conducted for all variables. One-way repeated-
measures univariate analysis of variance was conducted to examine if people fixated
longer and more often (in terms of longer duration and larger number of the fixations)
at some body sites (i.e., shoulder) than other body sites (i.e., neck). Simple linear
regression was used to test relationships among body satisfaction, body discrepancy,
concern with fit, confidence in judgment, and purchase intent (See Figure 4.1, See
Table 4. 1). A .05 level of significance was set for all analyses of dependent variables.
Figure 4.1 Hypotheses tested in this study
Body image
Discrepancy
Body
Satisfaction
Visual Attention
Fixation
Duration
Fixation
Number
Concern
with Fit
Confidence
of Judgment
Purchase
Intention H1
H2
H3
H4 H5
H6
H7
Body
Sites
H8
57
Table 4.1 Statistic analysis
Hypotheses & Variables Statistical Test
H1 IV: Body Image discrepancy Simple Linear regression to test the
relationship between body image
discrepancy and body cathexis DV: Body Cathexis
H2 IV: Body Image discrepancy Simple Linear regressions to test the
relationship of body image discrepancy
to duration of fixations and number of
fixations
DV: Duration of Fixations
Number of fixations
H3 IV: Body Cathexis Simple Linear regressions to test the
relationship of body cathexis to
duration of fixations and number of
fixations
DV: Duration of Fixations
Number of fixations
H4 IV: Duration of Fixations
Number of fixations
Simple Linear regression will be used
to test the relationship among duration
of fixations, number of fixations and
concern with fit, because duration of
fixations and number of fixations were
highly correlated.
DV: Concern with fit
H5 IV: Concern with fit Simple Linear regression to test the
relationship between concern with fit
and confidence in fit judgment DV: Confidence in fit judgment
H6 IV: Concern with fit Simple Linear regression to test the
relationship between concern with fit
and purchase intention DV: Purchase intention
H7 IV: Confidence of judgment Simple Linear regression to test the
relationship between confidence in fit
judgment and purchase intention DV: Purchase intention
H8 IV: Body sites (i.e. neck) One-way repeated-measures
ANOVA to test if people fixated longer
and more often at some body sites (i. e.
shoulder) than other body sites (i.e.
neck).
DV: Duration of Fixations
Number of fixations
58
Demographics
The demographic data included self-reported age, weight and height. The
mean age of participants (N = 45) was 21 years, with a range of 17 to 25 years. All
participants were female undergraduate students who majored in either fashion design
or fashion merchandising. The mean weight was 115lbs with a range of 98lbs to 155
lbs. The height ranged from 5ft 5in to 5 ft 9in, with an average of 5ft 6in (see Table
4.2).
Table 4.2 Demographics: Age, Weight, and Height
Demographics
of participants
Female participants
N = 45
Min. Max. Mean SD
Age 17 25 21 2.31
Weight (lbs) 98 155 115 4.52
Height (ft) 5 ft 5in 5 ft 9in 5 ft 6in 3.94
Descriptive Statistics for Variables
•Body Cathexis
Participants’ body cathexis was measured using the scale revised from the
body cathexis scale used in Labat and Delong (1990). The original body cathexis scale
used in Labat and Delong’s research (1990) accessed satisfaction with 19 body sites. In
the present study, satisfaction with 9 relevant body sites (neck, shoulder, bust, back,
arm, waist, abdomen, hip, and buttock) was summed to form an indicator of overall
59
body cathexis. The nine body sites were selected because they were highly related to
blazer fit and blazers were chosen as the stimuli in this study. Scores could range from
9 to 81. Higher scores indicate more satisfaction with body parts. The actual range of
overall satisfaction scores was from 30 to 75 (M = 52.05). Among the individual body
sites, participants’ satisfaction scores were relatively high for the neck (M = 6.89); and
relatively low for the abdomen (M = 4.97) (see Table 4.3).
Table 4.3 Descriptive Statistics for Body cathexis and sub-items
Measures Range Female participants
N = 45
Min. Max. Mean S.D.
Total Body Cathexis 9-81 30 75 52.05 11.03
Cathexis of Neck 1-9 4 9 6.89 1.78
Cathexis of Shoulder 1-9 3 9 6.87 1.73
Cathexis of Bust 1-9 1 9 5.61 2.31
Cathexis of Back 1-9 3 9 5.89 1.64
Cathexis of Arm 1-9 1 9 5.32 2.13
Cathexis of Abdomen 1-9 1 9 4.97 2.32
Cathexis of Waist 1-9 1 9 5.76 2.11
Cathexis of Hip 1-9 1 9 5.05 1.75
Cathexis of Buttock 1-9 2 9 5.68 1.85
•Body Discrepancy
The discrepancy between participants’ actual and ideal bodies was
measured by a revised version of Body-image Ideals Questionnaire items (Cash &
Szymanski, 1995). The original BIQ assessed individual’s self-ideal discrepancy on 10
60
appearance characteristics (height, skin complexion, hair texture and thickness, facial
features, muscle tone and definition, body proportions, weight, chest size, physical
strength, and physical coordination). For the present research, the BIQ scale was
modified to tap respondent’s self-ideal discrepancy on nine body sites (neck, shoulder,
bust, arm, back, waist, hip, buttock, and abdomen). The nine relevant body sites were
chosen because female consumers’ perception of these body sites played important
roles in the process of judging fit of a garment (Ashdown & O'Connell, 2006). In the
present study, body discrepancy scale measured participants’ discrepancies between
their ideal and real status of each body site using a 4-point scale: 0 (exactly as I am),
+1 (almost as I am), +2 (fairly unlike me), and +3 (very unlike me). The importance of
each body site toward their appearance was also measured using a 4-point scale: 0 (not
important), 1(some what important), 2 (moderately important), and 3 (very important).
The final body self-discrepancy was calculated by multiplying the discrepancy scores
by the importance scores and then summing the scores.
Hence, the final body self-discrepancy score had a possible range from 0
to 81. Higher scores indicate greater discrepancy. In the present study, participants’
body self-discrepancy scores ranged from 6 to 39 (M = 19.76). The results indicated
that the participants had a small discrepancy between their ideal body and actual body.
Among the individual body sites, neck had a relatively low discrepancy score (M =
0.33); abdomen had a relatively higher discrepancy score (M = 3.96) (see Table 4.4).
61
Table 4.4 Descriptive Statistics for Body self-discrepancy
Measures Range Female participants
N = 45
Min. Max. Mean S.D.
Total Body discrepancy 0-81 3 39 19.6 9.57
Discrepancy on Neck 0-9 0 2 .33 1.78
Discrepancy on Shoulder 0-9 0 6 .84 1.73
Discrepancy on Bust 0-9 0 9 2.64 2.31
Discrepancy on Back 0-9 0 9 1.27 1.64
Discrepancy on Arm 0-9 0 9 3.02 2.13
Discrepancy on Abdomen 0-9 0 9 3.96 2.32
Discrepancy on Waist 0-9 0 9 2.53 2.11
Discrepancy on Hip 0-9 0 9 2.51 1.75
Discrepancy on Buttock 0-9 0 9 2.53 1.85
•Visual Attention
Two major parameters of visual attention, the number of visual fixations
and duration of fixations on visual stimuli were recorded by the eye tracker in less than
20 minutes. Duration time, measured in milliseconds, refers to the amount of time that
a person’s eye stays on a specific spot. Number of fixations refers to the number of
times the eye located or hit on the same spot. In this study, the visual stimuli (front-
view and back-view photos of a human model wearing blazers of different sizes) were
divided into nine areas of interest (AOI) based on the relevant body sections, including
neck, shoulder, bust, waist, arm, back, abdomen, hip, and buttock (See Appendix A).
Participant’s duration time and number of fixations for the AOIs were hypothesized to
be significantly different from each other, for example, waist as an AOI might draw
significantly longer duration and more fixations than abdomen as an AOI. Participant’s
overall duration time and number of fixations were hypothesized to be affected by
62
participant’s body image ideals discrepancy, and body satisfaction, and influence
participants’ concern with fit.
The duration and number of fixations on the AOIs of front-view and back-
view photos of each blazer (size 2, 4, 6, and 8) were summed respectively. Photos of
blazer of size 4 had the longest duration of fixations, ranged from 2252 to 16882 (M =
6271) milliseconds, and the largest number of fixations, ranged from 9 to 92 (M =
34.9). Photos of blazer of size 8 had the shortest duration of fixations, ranged from
1656 to 15553 (M = 5307), and the largest number of fixations, ranged from 7 to 66
(M = 22.0) (see Table 4.5). The results indicated that the blazer of size 4 received the
most visual attention from participants while the blazer of size 8 received the least
visual attention.
Table 4.5 Descriptive Statistics for duration and numbers of fixations
Visual
stimuli
Duration of fixations* Number of fixations
Min. Max. Mean S.D. Min. Max. Mean S.D.
size 2 2252 16882 6271.8 3322.8 8 53 24.6 10.6
size 4 2292 21128 9430.4 4711.6 9 92 34.9 17.2
size 6 1853 14136 6244.8 2960.2 9 48 24.3 9.9
size 8 1656 15553 5307.4 3386.5 7 66 22.0 12.7
*Duration of fixation was measured in milliseconds.
•Concern with Fit
The Concern with Fit scale, which included three items adapted from
Rosa, Garbarino, Malter (2006) and Kim and Damhorst (2010), measured consumers’
about whether the size of garments fit their bodies and looked good on their bodies.
63
The three items used a 5-point scale. The scores for the three items were summed and
averaged to form an index yielding a range from 1 to 5. Higher scores indicate
participants had a higher level of concern with the fit of the blazer of a specific size.
Among the 4 sizes of the blazers, participants indicated high level of
concern toward the size 8 blazer (M = 3.8) and low level of concern toward the size 4
blazer (M = 2.2) (see Table 4.6).
Table 4.6 Descriptive Statistics for concern with fit
Measures Female participants
N = 45
Min. Max. Mean S.D.
Concern with fit for size 2 1 5 3.0 1.7
Concern with fit for size 4 1 5 2.2 1.1
Concern with fit for size 6 1 5 3.0 1.4
Concern with fit for size 8 1 5 3.8 1.2
•Confidence in Judgment
Confidence in judgment scale, which includes two items revised from
Laroche, Kim and Zhou (1996), measures consumers’ certainty regarding the fit of a
garment and consumers’ certainty regarding choosing the best-fitting garment. The
original two items used in Laroche et al. (1996) tapped consumers’ confidence in
brand judgment. In the present study, these items were modified to access consumers’
confidence in garment fit judgment. The two items used a 5-point scale. The scores of
the two items were averaged to yield a potential range from 1 to 5. Higher scores
indicate higher certainty. Participants’ responses toward the four sizes of blazers were
very close (see Table 4.7).
64
Table 4.7 Descriptive Statistics for Confidence of judgment
Measures Female participants
N = 45
Min. Max. Mean S.D.
Confidence of judgment for size 2 1 5 3.9 1.1
Confidence of judgment for size 4 1 5 4.0 .78
Confidence of judgment for size 6 1 5 3.7 .92
Confidence of judgment for size 8 2 5 4.0 .75
•Purchase Intention
Respondents reported their willingness to buy the blazer from online
retailers. The item was adopted from Kim and Lennon (2007) with a range from 1 to 5.
Higher scores indicate higher willingness to make a purchase from online retailers.
Participants indicated a relatively high level of purchase intent to buy a blazer of size 4
(M = 3.5) and relatively low level of purchase intent toward the blazer of size 8 (M =
2.3) (see Table 4.8).
Table 4.8 Descriptive Statistics for Purchase intent
Measures Female participants
N = 45
Min. Max. Mean S.D.
Purchase intent for size 2 1 5 2.7 1.3
Purchase intent for size 4 1 5 3.5 .99
Purchase intent for size 6 1 5 2.7 1.2
Purchase intent for size 8 1 5 2.3 1.2
65
Hypotheses Testing
Hypothesis 1: Participants’ total body image discrepancy will be
negatively related to their total body satisfaction.
Linear regression was used to examine the relationship between body
satisfaction and body image discrepancy. The analysis indicated that body image
discrepancy was a significant predictor of body satisfaction, (1,43) 26.4, .001F p .
Body image discrepancy accounted for 38% ( 2 .38R ) of the variation in body
satisfaction. The two variables were negatively related, .62 . Hence, participants
who had a large discrepancy between their ideal body and actual body felt less
satisfied toward their body. Therefore, H1 was supported.
Hypothesis 2a: Participants’ total body image discrepancy will be
positively related to the total duration of fixations over the human models when
making judgments of fit. (i.e., Consumers will look longer at the human model if
they have large body image discrepancy.)
Hypothesis 2b: Participants’ total body image discrepancy will be
positively related to the total number of fixations over the human model when
making judgments of fit. (i.e., Consumers will look more often at the human model
if they have large body image discrepancy.)
For hypothesis 2a, simple linear regression was used to examine the
relationship between total body image discrepancy and total duration of fixations on
the human model. The analysis indicated that body image discrepancy was a
significant predictor for overall duration of fixations on the human model,
(1,43) 9.2, .01F p (for size 2), (1,43) 7.5, .05F p (for size 4),
66
(1,43) 16.7, .01F p (for size 6), (1,43) 15.8, .01F p (for size 8). Body image
discrepancy accounted for 17~32% ( ) of the variation in overall
duration of fixations on the human model. The two variables were positively
related, .42 ~ .56 . Hence, participants who had a large discrepancy between their
ideal body and actual body looked longer at the human model when they made
judgments of fit. Therefore, Hypothesis 2a was supported.
For Hypothesis 2b, simple linear regression results indicated that the total
body image discrepancy positively predicts the overall number of fixations on the
human model (see Table 4.9). (1,43) 6.01, .05F p (for size 2),
(1,43) 6.42, .05F p (for size 4), (1,43) 14.13, .01F p (for size 6),
(1,43) 17.32, .01F p (for size 8). Body image discrepancy accounted for 14~32%
( 2 .17 ~ .32R ) of the variation in fixations on the human model. The two variables
were positively related, .38 ~ .57 . Therefore, Hypothesis 2b was supported as
well. Participants who had a large self-ideal body discrepancy looked more often at the
human model when they made judgment of fit.
Table 4.9 Linear Regression Results for Hypotheses 2
DV= Duration of Fixation DV= Number of Fixation
Size2 Size4 Size6 Size8 Size2 Size4 Size6 Size8
IV= Body
image
discrepancy
F 9.21 7.52 16.72 15.84 6.01 6.42 14.13 17.32
.45 .42 .56 .55 .38 .39 .53 .57
.20 .17 .32 .31 .14 .15 .28 .32
.00 .10 .00 .00 .02 .02 .00 .00
67
Hypothesis 3a: Participants’ total body satisfaction will be negatively
related to the total duration of fixations over the human model when making
judgments of fit. (i.e., Consumers will look longer at the human model if they are
less satisfied with their bodies.)
Hypothesis 3b: Participants’ total body satisfaction will be negatively
related to the total number of fixations over the human model when making
judgments of fit. (i.e., Consumers will look more often at the human model if they
are less satisfied with their bodies.)
For Hypothesis 3a, simple linear regression analysis was conducted for
each size of blazer to examine the relationship between total body satisfaction and
overall duration of fixations on the human model. The results showed that body
satisfaction significantly predicted the duration of fixations on the human model,
(1,43) 5.23, .05F p (for size 2), (1,43) 7.81, .01F p (for size 4),
(1,43) 7.42, .05F p (for size 6), (1,43) 4.82, .05F p (for size 8). Total body
satisfaction accounted for 12~18% ( 2 .12 ~ .18R ) of the variation in duration of
fixations. The two variables were negatively related, .42 ~ .34 . Hence,
participants who were less satisfied with their body looked longer at the human model
when they made judgments of fit. Therefore, Hypothesis 3a was supported.
For Hypothesis 3b, simple linear regression analysis (see Table 4.10) was
conducted for each size of blazer to examine the relationship between total body
satisfaction and overall number of fixations on the human model. The results showed
that body satisfaction significantly predicted the number of fixations on the human
model, (1,43) 4.23, .05F p (for size 2), (1,43) 6.12, .05F p (for size 4),
68
(1,43) 6.51, .05F p (for size 6), (1,43) 4.22, .05F p (for size 8). Total body
satisfaction accounted for 11~15% ( 2 .11~ .15R ) of the variation in number of
fixations. The two variables were negatively related, .39 ~ .32 . Hence,
participants who were less satisfied with their body looked more often at the human
model when they made judgments of fit. Therefore, Hypothesis 3b was supported.
Table 4.10 Linear Regression Results for Hypotheses 3
DV= Duration of Fixation DV=Number of Fixation
Size2 Size4 Size6 Size8 Size2 Size4 Size6 Size8
IV= Body
satisfaction
F 5.23 7.81 7.42 4.82 4.23 6.12 6.51 4.22
-.36 -.42 -.41 -.34 -.32 -.38 -.39 -.32
.13 .18 .17 .12 .11 .14 .15 .11
.03 .01 .01 .03 .04 .02 .02 .04
Hypothesis 4-a: Duration of fixations on human model will be
negatively related to concern with fit (i.e., Consumers will have less concern with fit
when they get more information by looking longer at the photos of human model
wearing blazers.)
Hypothesis 4-b: Number of fixations on human model will be negatively
related to concern with fit (i.e., Consumers will have less concern with fit when they
get more information by looking more often at the photos of human model wearing
blazers.)
First, Pearson correlation coefficients between two independent variables,
overall number of fixations and overall duration of fixations, were calculated. The
69
results showed that these two independent variables were highly correlated, r>.85,
p<.001. The multi-collinear recommendation for computing multiple linear regressions
was violated. Hence, the Hypotheses 4a and 4b were tested individually using simple
linear regression instead of multiple linear regression.
For Hypothesis 4a, simple linear regression was used to examine the
relationship between the total duration of fixations on the human model and concern
with fit. Participants’ duration of fixations on the human model and concern with fit
were repeatedly measured for each size of blazer, the linear regression between these
two variables were repeatedly computed for each size of blazer. The results revealed
no significant relationship between the total duration of fixations and concern with
fit, (1,43) .38, .05F p (for size 2), (1,43) 2.15, .05F p (for size 4),
(1,43) .17, .05F p (for size 6), (1,43) 3.91, .05F p (for size 8). Total duration of
fixations only accounted for 1~8% ( 2 .01~ .08R ) of the variation in concern with fit
(see Table 4.11). Therefore, Hypothesis 4a was not supported.
For Hypothesis 4b, simple linear regression was calculated to examine the
relationship between the total number of fixations on the human model and concern
with fit. Participants’ total number of fixations on the human model and concern with
fit were repeatedly measured for each size of blazer, the linear regression between
these two variables were respectively computed for each size of blazer. The results
showed no significant relationship between the total number of fixations and concern
with fit, (1,43) .47, .05F p (for size 2), (1,43) 1.74, .05F p (for size 4),
(1,43) .20, .05F p (for size 6), (1,43) 4.83, .05F p (for size 8). The F ratio
between concern with fit and total number of fixations for all sizes of blazers were
70
bigger than .05. The overall number of fixations only accounted for 1~10%
( 2 .01~ .10R ) of the variation in concern with fit for all sizes (see Table 4.11).
Therefore, Hypothesis 4b was not supported.
Table 4.11 Linear Regression Result for Hypothesis 4
IV: Duration of Fixation IV: Number of Fixation
Size2 Size4 Size6 Size8 Size2 Size4 Size6 Size8
DV:
Concern
with fit
F .38 2.15 .17 3.91 .47 1.74 .20 4.70
.09 .218 -.06 -.29 .10 .20 -.07 -.32
.01 .05 .01 .08 .01 .04 .01 .10
.54 .15 .68 .05 .50 .20 .66 .05
Hypothesis 5: Consumers’ concern with garment fit will be negatively
related to their confidence in fit judgments.
For Hypothesis 5, simple linear regression was computed to test the
relationship between the concern with fit and confidence in fit judgment. Participants’
concern with fit and confidence in fit judgments were repeatedly measured for each
size of blazer, hence the linear regression between these two variables were
respectively computed for each size of blazer. The results indicated no significant
relationship between concern with fit and confidence in fit
judgments, (1,43) 1.32, .05F p (for size 2), (1,43) 1.74, .05F p (for size 6),
(1,43) 2.39, .05F p (for size 8). Noticeably, participants’ concern with fit for blazer
size 4 was significantly related to their confidence in fit judgment for the same size
blazer, (1,43) 18.24, .01F p , .55 . However, the concern with fit for size 2, 6,
71
and 8 only accounted for 3~5% ( 2 .03 ~ .05R ) of the variation in corresponding
confidence in fit judgment (see Table 4.12). Therefore, Hypothesis 5 was not
supported.
Table 4.12 Linear Regression Result for Hypothesis 5
DV=Confidence in fit judgment
Size2 Size4 Size6 Size8
IV=Concern with fit
F 1.32 18.24 1.74 2.39
-.17 -.55 -.20 .23
.03 .30 .04 .05
.26 .00 .19 .13
Hypothesis 6: Consumers’ concern with fit will be negatively related to
their purchase intent.
For Hypothesis 6, linear regression results (see Table 4.13) indicated that
concern with fit was a significant predictor for purchase intent,
(1,43) 13.01, .01F p (for size 2), (1,43) 15.32, .01F p (for size 4),
(1,43) 30.71, .01F p (for size 6), (1,43) 32.23, .01F p (for size 8). Concern with
fit accounted for 23~43% ( 2 .23 ~ .43R ) of the variation in purchase intention. These
two variables were negatively related .48 (for size 2), .51 (for size 4),
.65 (for size 6), .65 (for size 8). Participants who were more concerned
about the fit toward a size of garment correspondingly had lower purchase intent for
this size. Therefore, H6 was supported.
72
Table 4.13 Linear Regression Result for Hypotheses 6
DV= Purchase Intention
Size2 Size4 Size6 Size8
IV=Concern with
fit
F 13.01 15.32 30.71 32.23
-.48 -.51 -.65 -.65
.23 .26 .42 .43
.001 .000 .000 .000
Hypothesis 7: Consumers’ confidence with fit judgments will be
positively related to their purchase intent.
For Hypothesis 7, simple linear regression was conducted to test the
relationship between confidence in fit judgment and purchase intention. Participants’
confidence in fit judgments and purchase intention were repeatedly measured for each
size of blazer, hence the linear regression between these two variables was respectively
computed for each size of blazer as well. The results indicated no significant
relationship between confidence in fit judgment and purchase
intention, (1,43) 2.43, .05F p (for size 2), (1,43) 1.60, .05F p (for size 6),
(1,43) 2.89, .05F p (for size 8). Noticeably, participants’ confidence in fit
judgment for blazer of size 4 was positively ( .39 ) related to their purchase
intention for the same size blazer, (1,43) 7.93, .05F p . For size 2, 6, and 8,
participants’ confidence in fit judgments only accounted for 4~6% ( 2 .04 ~ .06R ) of
the variation in corresponding purchase intention (see Table 4.14). Therefore, H7 was
not supported.
Table 4.14 Linear Regression Result for Hypotheses 7
DV: Purchase Intention
73
Size2 Size4 Size6 Size8
IV: Confidence in
fit judgment
F 2.48 7.93 1.60 2.89
.23 .39 .189 -.251
.05 .16 .04 .06
.12 .01 .21 .10
Hypothesis 8a - There is a significant difference among the body sites of
the human model on duration of fixations when judging the garment fit;
Hypothesis 8b - There is a significant difference among the body sites of
the human model on number of fixations when judging the garment fit..
One-way repeated-measures ANOVA were used to examine whether the
duration of fixations and numbers of fixations varied across the body sites (e.g., neck
versus waist). One-way repeated-measures ANOVA tests were conducted for front
view and back view of four sizes of blazers. Specifically, for the front view, the
dependent variables were duration of fixations and number of fixations, while the
independent variable was body site with 7 levels (neck, shoulder, bust, arm, waist,
abdomen, and hip). For the back view, the dependent variables were duration of
fixations and number of fixations, while the independent variable was body site with 6
levels (neck, shoulder, back, arm, waist, and buttock).
For Hypothesis 8a, the one-way repeated-measures ANOVA results
showed that consumers’ duration of fixations significantly varied across the body sites,
F(3.75, 164.831) = 21.41, p < .001(for front view of size 2), F(2.62, 115.41)=54.42, p
1 According to Kerr, Hall and Kozub (2002), when the sphericity assumption was not
met, the “Greenhouse-Geisser” correction was applied to the One-way ANOVA and
changed the degrees of freedom.
74
< .001 (for front view of size 4), F(2.85, 125.23)=41.00, p < .001 (for front view of
size 6), F(2.52, 111.00)=47.64, p < .001 (for front view of size 8), F(3.46, 152.05) =
17.62, p < .001 (for back view of size 2), F(3.87, 170.41)=27.35 , p < .001 (for back
view of size 4), F(2.85, 125.46)=33.43 , p < .001 (for back view of size 6,) F(3.624,
159.45)=18.36, p < .001 (for back view of size 8). The pairwise comparison results
also indicated that participants’ duration of fixations at bust and waist were
significantly longer than those at other body sites for front views of all sizes. There
was no significant difference between duration of fixations at bust and waist. Hence,
bust and waist are the hot-spots for front view of the blazer, where participants fixated
longest when they made fit judgments. Similarly, duration of fixations at back, waist,
and buttock were significantly longer than those at other body sites for back views of
all sizes. There was no significant difference among back, waist, and buttock. Hence,
back, waist and buttock were the hot-spots for back view of blazer (See Figure 4.2a &
b and Table 4.16). Therefore, H8a was supported.
For Hypothesis 8b, the one-way repeated-measures ANOVA results(See
Table 4.15) showed that consumers’ number of fixations significantly changed across
the body sites, F(3.32, 146.24) = 42.52, p < .001 (for front view of size 2), F(3.43,
151.04) = 52.88, p < .001 (for front view of size 4), F(3.09, 135.76) = 50.17, p < .001
(for front view of size 6), F(2.85, 125.22) = 54.11, p < .001 (for front view of size 8),
F(3.75, 164.83) = 21.41, p < .001 (for back view of size 2), F(3.19, 140.29) = 32.57, p
< .001 (for back view of size 4), F(3.34, 146.94)=37.99, p < .001 (for back view of
size 6,) F(2.52, 111.00) = 47.64, p < .001 (for back view of size 8).
75
The pairwise comparison results also indicated that participants’ number
of fixations at bust and waist were significantly larger than those at other body sites for
front views of all sizes. There was no significant difference between number of
fixations at bust and waist. Hence, bust and waist are the hot-spots for front view of
blazer, where participants fixated most often when they made fit judgments. Similarly,
number of fixations at back, waist, and buttock were significantly larger than those at
other body sites for back views of all sizes. There was no significant difference among
back, waist, and buttock. Hence, back, waist and buttock were the hot-spots for the
back view of the blazer (See Figure 4.3a & b and Table 4.16). Therefore, H8b was also
supported.
Table 4.15 One-way repeat measures ANOVA Result for Hypothesis 8
Hypothesis 8a
DV = duration of fixations;
IV = body sites with 7 levels (neck, shoulder, bust, arm, waist, abdomen, and hip)
Front
view
Size 2 F(2.33, 102.44) = 31. 19 p < .001
Size 4 F(2.62, 115.41)= 54.42 p < .001
Size 6 F(2.85, 125.23)= 41.00 p < .001
Size 8 F(2.52, 111.00)= 47.64 p < .001
DV = duration of fixations;
IV = body sites with 6 levels (neck, shoulder, back, arm, waist, and buttock).
Back
view
Size 2 F(3.46, 152.05) = 17.62 p < .001
Size 4 F(3.87, 170.41)= 27.35 p < .001
Size 6 F(2.85, 125.46)= 33.43 p < .001
Size 8 F(3.62, 159.45)= 18.36 p < .001
Hypothesis 8b
DV = number of fixations
IV = body sites with 7 levels (neck, shoulder, bust, arm, waist, abdomen, and hip)
Front Size 2 F(3.32, 146.24) = 42.52 p < .001
76
view Size 4 F(3.43, 151.04)= 52.88 p < .001
Size 6 F(3.09, 135.76)= 50.17 p < .001
Size 8 F(2.85, 125.22)= 54.11 p < .001
DV = number of fixations
IV = body sites with 6 levels (neck, shoulder, back, arm, waist, and buttock).
Back
view
Size 2 F(3.75, 164.83) = 21.41 p < .001
Size 4 F(3.19, 140.29)= 32.57 p < .001
Size 6 F(3.34, 146.94)= 37.99 p < .001
Size 8 F(2.52, 111.00)= 47.64 p < .001
Table 4.16 Means of duration of fixations and number of fixations
Hypothesis
8a
Mean of duration of fixations
neck shoulde
r
bust arm waist abdome
n
hip
Fron
t
view
size 2 53.58 473.96 1319.6
0
462.87 1968.3
3
407.49 307.44
size 4 107.67 876.64 1415.5
1
468.13 2833.8
9
468.91 458.42
size 6 89.96 500.87 1382.9
3
228.0 1379.0
0
217.96 231.16
size 8 38.93 315.40 863.71 199.82 1461.7
6
182.13 165.69
Hypothesis
8a
Mean of duration of fixations
neck shoulder back arm waist buttock
Back
view
size 2 66.02 225.11 515.22 189.98 650.22 543.82
size 4 116.00 346.78 698.33 277.33 950.49 948.36
size 6 10.20 168.87 840.69 137.20 686.29 742.36
size 8 51.84 221.47 596.16 195.76 577.22 653.42
Hypothesis
8b
Mean of number of fixations
Neck Shoulder Bust Arm Waist Abdomen hip
Fron
t
size 2 .27 1.60 5.20 1.98 6.80 1.76 1.24
size 4 .56 3.22 5.67 1.89 8.98 1.82 2.09
77
view size 6 .42 1.78 5.13 1.09 5.29 1.07 .87
size 8 .18 1.27 3.58 .87 5.80 .80 .73
Hypothesis
8b
Mean of number of fixations
neck shoulder back arm waist buttock
Back
view
size 2 .18 .91 1.84 .82 2.64 2.13
size 4 .44 1.22 2.53 1.09 3.60 3.84
size 6 .04 .64 3.07 .67 2.64 2.98
size 8 .16 .73 2.38 .87 2.40 2.91
Note: unit of duration of fixations is millisecond
78
Figure 4.2a Effect of body sites on duration of fixations in Front view
Figure 4.2b Effect of body sites on duration of fixations in back view
79
Figure 4.3a Effect of body sites on number of fixations in Front View
80
Figure 4.3b Effect of body sites on number of fixations in back view
81
Table 4.17 Summery of Hypotheses Testing Results
Hypothesis 1: Participants’ total body image discrepancy will be negatively related
to their total body satisfaction.
IV= body
image
discrepancy
DV= body
satisfaction
F (1,43) = 26.4 p <
.001
β = -.62 R2 = .38
Hypothesis 2-a: Participants’ total body image discrepancy will be positively
related to the total duration of fixations over the human models when making
judgments of fit.
IV = body
image
discrepancy
DV := overall
duration of
fixation
F (1,43) = 9.21a
p < .01 β = .45 R2 = .20
F (1,43) = 7.52b p < .05 β = .42 R
2 = .17
F (1,43) = 16.72c p < .01 β = .56 R
2 = .32
F (1,43) = 15.84d p < .01 β = .55 R
2 =
.31
Hypothesis 2-b: Participants’ total body image discrepancy will be positively
related to the total number of fixations over the human model when making
judgments of fit.
IV=body
image
discrepancy
DV= overall
number of
fixations
F (1,43) = 6.01a
p < .05 β = .38 R2 = .14
F (1,43) = 6.42.b p < .05 β = .39 R
2 = .15
F (1,43) = 14.13c p < .01 β = .53 R
2= .28
F (1,43) = 17.32d p < .01 β = .57 R
2= .32
Hypothesis 3-a: Participants’ total body satisfaction will be negatively related to
the total duration of fixations over the human model when making judgments of fit.
IV= body
cathexis
DV= overall
duration of
Fixations
F (1,43) = 5.23a
p < .05 β = - .36 R2 = .13
F (1,43) =7.81 b p < .05 β = -.42 R
2 = .18
F (1,43) =7.42 c p < .05 β = -.41 R
2=.17
F (1,43) =4.82 d p < .05 β = -.34 R
2= .12
Hypothesis 3-b: Participants’ total body satisfaction will be negatively related to
the total number of fixations over the human model when making judgments of fit.
IV= body
cathexis
DV= overall
number of
fixations
F (1,43) = 4.23a
p < .05 β = -.32 R2 = .11
F (1,43) = 6.12.b p < .05 β = -.38 R
2 = .14
F (1,43) = 6.51c p < .05 β = -.39 R
2 =.15
F (1,43) = 4.22d p < .05 β = -.32 R
2= .11
Hypothesis 4a: Duration of fixations on human model will be negatively related to
concern with fit.
82
IV = overall
duration of
fixations
DV = concern
with fit
F (1,43) = .38a
p > .05 β = .09 R2 = .01
F (1,43) = 2.15b p > .05 β = .218 R
2 = .05
F (1,43) = .17c p > .05 β = -.06 R
2= .01
F (1,43) = 3.91d p > .05 β = -.29 R
2= .08
Hypothesis 4b: Number of fixations on human model will be negatively related to
concern with fit.
IV = overall
number of
fixations
DV = concern
with fit
F (1,43) = .47a
p > .05 β = .10 R2 = .01
F (1,43) = 1.74b p > .05 β = .20 R
2 = .04
F (1,43) = .20c p > .05 β = -.07 R
2 =.01
F (1,43) = 4.70d p > .05 β = -.32 R
2 = .10
Hypothesis 5: Consumers’ concern with garment fit will be negatively related to
their confidence in fit judgments.
IV= concern
with fit
DV= confidence
in fit judgment
F (1,43) = 1.32a
p >.05 β = -.17 R2 = .03
F (1,43) = 18.24b p < .01 β = -.55 R
2 = .30
F (1,43) = 1.74c p > .05 β = -.20 R
2= .04
F (1,43) = 2.39d p > .05 β = .23 R
2= .05
Hypothesis 6: Consumers’ concern with fit will be negatively related to their
purchase intent.
IV= concern
with fit
DV= purchase
intention
F (1,43) = 13.01a
p < .01 β = -.48 R2 = .23
F (1,43) = 15.32.b p < .01 β = -.51 R
2 = .26
F (1,43) = 30.71c p < .01 β = -.65 R
2= .42
F (1,43) = 32.23d p < .01 β = -.65 R
2= .43
Hypothesis 7: Consumers’ confidence with fit judgments will be positively related
to their purchase intent.
IV=
confidence in
judgment
DV= purchase
intention
F (1,43) = 2.48a
p > .05 β = .23 R2 = .05
F (1,43) = 7.93b p < .05 β = .39 R
2 = .16
F (1,43) = 1.60c p > .05 β = .19 R
2=.04
F (1,43) = 2.89d p > .05 β = -.25 R
2= .06
Hypothesis 8-a: There is a significant difference among duration of fixations over
the body sites of the human model when judging the garment fit.
IV=
body
sites (i.e.
neck)
DV= duration
of Fixations
Front
view
F(3.75, 164.83) = 21.41a
p < .001
F(2.62, 115.41)= 54.42b p < .001
F(2.85, 125.23)= 41.00c p < .001
F(2.52, 111.00)= 47.64d p < .001
83
IV=
body
sites (i.e.
neck)
DV= duration
of Fixations
Back
View
F(3.46, 152.05) = 17.6 a
p < .001
F(3.87, 170.41)= 27.35b p < .001
F(2.85, 125.46)= 33.43c p < .001
F(3.62, 159.45)= 18.36d p < .001
Hypothesis 8-b: There is a significant difference among number of fixations over
the body sites of the human model when judging the garment fit.
IV=
body
sites (i.e.
neck)
DV= number
of fixations
Front
View
F(3.32, 146.24) = 42.52 a
p < .001
F(3.43, 151.04)= 52.88 b
p < .001
F(3.09, 135.76)= 50.17 c p < .001
F(2.85, 125.22)= 54.11 d
p < .001
IV=
body
sites (i.e.
neck)
DV= number
of fixations
b
Back
view
F(3.75, 164.83) = 21.41 a p < .001
F(3.19, 140.29)= 32.57 b
p < .001
F(3.34, 146.94)= 37.99 c p < .001
F(2.52, 111.00)= 47.64 d
p < .001
Note a is for size 2
b is for size 4
c is for size 6
d is for size 8
84
Chapter 5
DISCUSSION AND CONCLUSIONS
The purposes of this study were to: examine the relationship between
female consumers’ body image discrepancy and body satisfaction, which was
addressed in Hypothesis 1; examine the effects of female consumers’ body image
discrepancy and body satisfaction on their visual perception of garment fit (focus on
how long and how often they look at the garments), which was addressed in
Hypotheses 2 and 3; examine the relationships among female consumers’ visual
perception, concern with garment fit, confidence in fit judgment and the purchase
intent, which were addressed in Hypotheses 4, 5, 6, and 7; examine the effect of body
sites on female consumers’ visual perception of garment fit, which was addressed in
Hypothesis 8. Generally, this research examined the effect of subjective factors (body
image discrepancy, body satisfaction) and objective factors (body sites) on consumers’
visual perceptions (duration of fixations and number of fixations) of garments and the
effect of visual perception on consumers’ concern with fit judgment, confidence in
judgment, and purchase intent for the garment.
Eight hypotheses were investigated in this study: 1) Hypothesis 1-
Participants’ total body image discrepancy will be negatively related to their total body
satisfaction; 2) Hypothesis 2a -Participants’ total body image discrepancy will be
positively related to the total duration of fixations over the garment when making
85
judgments of fit; Hypothesis 2b -Participants’ total body image discrepancy will be
positively related to the total number of fixations over the garment when making
judgments of fit; 3) Hypothesis 3a - Participants’ total body satisfaction will be
negatively related to the total duration of fixations over the garment when making
judgments of fit; Hypothesis 3b - Participants’ total body satisfaction will be
negatively related to the total number of fixations over the garment when making
judgments of fit; 4)Hypothesis 4a - Duration of fixations on a human model will be
negatively related to concern with fit; Hypothesis 4b - Number of fixations on a human
model will be negatively related to concern with fit; 5) Hypothesis 5 - Consumers’
concern with garment fit will be negatively related to their confidence in fit judgments;
6) Hypothesis 6 - Consumers’ concern with fit will be negatively related to their
purchase intent; 7) Hypothesis 7 - Consumers’ confidence in fit judgments will be
positively related to their purchase intent; 8) Hypothesis 8a - There is a significant
difference among the body sites of the human model on duration of fixations when
judging the garment fit; Hypothesis 8b - There is a significant difference among the
body sites of the human model on number of fixations when judging the garment fit.
Discussion
It was predicted in Hypothesis 1 that there is a negative relationship
between college women’s total body image discrepancy and their body satisfaction. As
expected, the results of the current study supported this hypothesis. Female
Hypothesis 1: College women’s total body image discrepancy will be
negatively related to their total body satisfaction.
86
participants who had a larger discrepancy between their ideal body and their actual
body were prone to be less satisfied with their body. This finding was consistent with
previous studies (Altabe & Thompson, 1992; Jacobi & Cash, 1994; Jung, Lennon &
Rudd, 2001).
According to Higgins, Klein, and Strauman (1987), discrepancy might
exist between the actual self and the idealized self that people would like to be; the
discrepancy may lead to an unpleasant internal psychological state (e.g., dissatisfied
with body) and negative behavior (e.g., eating disorder). When a college woman buys
clothes, she tends to compare the image of her body reflected in a retail store mirror
with the ideal image or expectation that exists in her mind. Attractive physical
appearance is one of the salient attributes people would like to possess (Connell, et al.,
2006). The ideal image in college women’s minds is influenced by the cultural ideals
for beauty, which set the standards for people to evaluate their body and physical
appearance (Irving, 1990; Richins, 1991). Monteath and McCabe (1997) found that
college women are extremely sensitive to cultural ideals of beauty in western culture,
and showed greater levels of body dissatisfaction when their perceived bodies deviated
from the cultural standard.
87
Hypothesis 2a: Participants’ total body image discrepancy will be positively related
to the total duration of fixations over the human models when making judgments of
fit.
Hypothesis 2b: Participants’ total body image discrepancy will be positively related
to the total number of fixations over the human model when making judgments of
fit.
Hypotheses 2a and 2b predicted that participants’ total body image
discrepancy would be positively related to their total duration and number of fixations
over the human model when they make judgments of fit. As expected, these
hypotheses were supported in the current study.
The female participants who had greater body image discrepancy tended to
look longer and more frequently over the human model who wore the garment, when
they were making judgments of fit of the garment. The results are consistent with the
findings in Ju and Johnson (2010), who used the eye tracking technology to measure
the extent to which young women focused their attention on human models in fashion
advertisements. They found that young women who reported themselves frequently
engaged in comparing with fashion models, tended to look longer and more frequently
at human models in fashion advertisements. Monteath and McCabe (1997) indicated
that college women were exposed to cultural ideals (i.e. fashion magazines or
advertisements) and often compared themselves with these idealized images of
thinness and beauty. It is reasonable to expect that women whose perceived bodies
deviated from their idealized bodies would spend longer comparing themselves with
either a fashion model in a magazine or an internalized ideal body in their memory. In
terms of finding a well-fitted garment, they might be more worried about if the
88
garment on the model’s body looks close to the ideal image which she would like to
present and hence take longer to confirm it.
Hypothesis 3a: Participants’ total body satisfaction will be negatively related to
the total duration of fixations over the garment when making judgments of fit.
Hypothesis 3b: Participants’ total body satisfaction will be negatively related to
the total number of fixations over the garment when making judgments of fit.
Hypotheses 3a and 3b predicted that participants’ total body satisfaction
would be negatively related to their total duration and number of fixations over the
human model when they making judgment of fit.
The female participants who were less satisfied with their bodies tended to
look longer and more frequently over the human model, when they were making
judgments of garment fit. This result could be explained from the perspective that
internal affective states influence visual information processing. Kuo, Hsu and Day
(2009) used the duration of fixations and number of fixations obtained from eye-
tracking technology to indicate people’s cognitive effort in processing information for
making judgments or decisions. They found that respondents under negative mood
were prone to make more cognitive effort (longer duration of fixations and greater
number of fixations) in making a judgment than their counterparts under positive
mood. Schwarz, Bless and Bohner (1991) proposed that people in a bad mood exerted
more cognitive effort in processing relevant information than those in a good mood,
because they were more motivated than their counterparts. It is reasonable to infer that
those women who were less satisfied with their body might go through a relatively
89
negative psychological state when they were making a fit judgment of a garment, and
hence fixate longer and more frequently on the human model.
Hypothesis 4a: Duration of fixations on human model will be negatively related to
concern with fit (i.e., Consumers who fixate longer at the photos of human model
wearing blazers have less concern with fit.)
Hypothesis 4b: Number of fixations on human model will be negatively related to
concern with fit (i.e., Consumers fixate more often at the photos of human model
wearing blazers have less concern with fit.)
Hypotheses 4a and 4b predicted that female consumers, who fixated
longer and more frequently on the visual image of the human model for evaluating fit
of a garment, would have less concern with fit. Consumers’ concern with fit refers to
the perceived risk related to fit. When consumer can gain more information about fit of
a garment, the perceived risk about fit or concern with fit would be reduced (Bhatnagar
et al., 2000; Forsythe et al., 2006; Park et al., 2005). According to visual attention
literature, people tend to fixate on the informative areas of a visual stimulus
(Rayner, 2009), and the information from fixations were integrated in visual short-
term memory (VSTM) (Hollingworth, Richard & Luck, 2008). People might gain
more information about the fit of a garment, when they fixate longer and more
frequently on the visual image of a human model wearing the garment. Previous
researchers (Kim & Lennon, 2010; Park & Stoel, 2002) found that the amount of
information could reduce the perceived risk for online shopping. Because concern
with fit was defined as perceived risk about fit for online shopping (Kim & Damhorst,
2010), it is reasonable to postulate that consumers might have less concern with fit
90
after they obtain more information about fit by fixating longer and more often on the
visual image of a human model wearing the garment.
However, these hypotheses were not supported for all sizes of blazers
tested in this study. The overall duration of fixations and number of fixations had no
significant relationship with concern with fit for all sizes of blazers. There are several
reasons for failing to support these hypotheses. First, only three items, “The blazer of
this size may not fit me”, “The blazer of this size may not look good on me”, “I may
have a hard time picturing myself wearing the garment” were adopted from the
original concern with fit scale used in Kim and Damhorst (2010)’s research. (The
reason is too many items presented on the screen of an eye tracker might make
participants get bored and sabotage the reliability of the whole experiment.) These
three items may be inadequate to assess consumers’ entire concern with fit. Second,
people might have different efficiency in perceiving visual information. Although
some participants fixated at the visual stimuli longer and more often, they are not able
to gain larger amount of information. Researchers found that experts tended to be
more efficient in extracting useful information for making a decision than amateurs
because they used less fixation and shorter duration on a visual image (Humphrey &
Underwood, 2009). Some participants might know more about how to judge fit of
garment, because they majored in apparel design, hence they could fixate less often on
the visual image and still extract enough information and be less concerned about fit.
In addition, other factors might also interrupt the possible relationship between
consumers’ visual attention and concern with fit, such as mood, apparel involvement,
91
and patience. Lastly, the sample size of this study may be too small to include enough
variation for supporting this hypothesis.
Hypothesis 5: Consumers’ concern with garment fit will be negatively related to
their confidence in fit judgments.
Hypothesis 5 predicted that consumers who were more concerned with fit
of a garment would be less confident with their fit judgment of the garment. However,
the hypothesis was not supported. In the current study, the construct of confidence in
fit judgment included two aspects: the knowledge confidence and choice confidence.
The knowledge confidence refers to participants’ confidence about their previous fit
evaluation, and the choice confidence is about how certain the participants are that the
garment would fit their bodies. The concern with fit taps the perceived risk for buying
a fit garment. The confidence of judgment includes the item related to consumers’ past
shopping experiences, while the concern with fit only taps the current potential risk.
The timing discrepancy might make these two variables unrelated to each other. It is
possible that consumers who were confident about their fit evaluation in the past
shopping experiences but still think it is very risky to purchase a garment online.
Secondly, the visual stimuli used in this study were blazers, which are
relatively expensive. Although consumers thought the risk for buying a blazer online is
financially large, they might still be confident about their fit judgment. Hence, the
confident of fit judgment was not able to predict the concern with fit. Finally, the
sample size of this study may be too small to generate a significant regression result.
Hypothesis 6: Consumers’ concern with fit will be negatively related to their
purchase intent.
92
Hypothesis 6 predicted that consumers who were more concerned with fit
of a garment would be less likely to buy this garment. As expected, the hypothesis was
supported. This result is consistent with previous studies (Kim & Damhorst, 2010,
Rosa et al, 2006). Concern with fit is one of the major reasons for the high return rate
of clothes bought online, because the garments cannot be inspected by consumers
before the purchase (Beck, 2003). Concern with fit refers to the perceived risk related
to fit of a garment. Because consumers evaluate fit of garment based on the visual and
tactile feedback, the lack of such information increases the perceived risk for online
apparel shopping and hence reduces purchase intention.
This result provides implications for online apparel retailers. The result
indicated that when consumers choose clothes online, they are worried about if the
garment would look good on them and they might have difficulty in visualizing the fit
of the garment. Hence, there is a need for online apparel retailers to develop a better
visual presentation tool for apparel to facilitate consumers’ visualization of how the
garment would look on their body to increase their willingness to shop online.
Hypothesis 7: Consumers’ confidence in fit judgments will be positively related to
their purchase intent.
Fit is a critical attribute to be considered when buying a garment
(Ashdown & O'Connell, 2006). When consumers were not able to make a confident
judgment about fit of the garment, they may think it is too risky to buy (Mollenkopf et
al., 2007). Hence, Hypothesis 7 predicted that consumers who were more confident in
their fit judgment of a garment would show a higher level of intention to buy the
garment. However, the hypothesis was not supported. This result is not consistent with
93
previous research that intention to buy a specific product would be positively affected
by a consumer's confidence in judging the attributes of a product (Cox & Stuart, 1964;
Howard & Sheth ,1969; Laroche et al., 1996; Peterson & Pitz, 1988; Urbany et al.
1989). College students sampled for this study might have less interest in buying a
blazer compared with working professionals, hence their purchase intention could be
very low even though they could confidently judge whether a blazer is properly fitted.
Consumers’ confidence in judgment ratings were influenced by their
previous experiences. Consumers who had bought properly fit clothes online would
be more likely to trust their evaluation of fit of garments; hence they would have high
level of confidence in fit judgment. However, there are some other factors which affect
consumers’ willingness to buy a garment, such as color, style, and fabric. For example,
consumers who were confident about their fit evaluation would not buy a garment, if
they did not like the style. Finally, the sample size of this study may be too small to
generate a significant regression result.
Hypothesis 8-a: There is a significant difference among the body sites of the human
model on duration of fixations. (Consumers fixate longer at some body sites than
other body sites when they are judging fit of a garment.)
Hypothesis 8-b: There is a significant difference among the body sites of the human
model on number of fixations. (Consumes fixate more often at some body sites than
other body sites when they are judging fit of a garment.)
Hypotheses 8a and 8b predicted that there was a significant difference in
the visual attention (measured by duration of fixations and number of fixations) placed
on the various body sites of the human model, which meant consumers tended to fixate
94
longer and more often at some body sites than other body sites when they were judging
fit of a garment. As expected, these hypotheses were supported. Specifically, in the
front view of human model wearing the garment, the body site waist received
significantly more visual attention (longer duration and larger number of fixations)
than all the other body sites. In addition, the bust was another body site that received a
large number of fixations and that people looked at for a long time. In the back view of
the human model, back, waist and buttocks were the three body sites on which people
placed a lot of attention. The result in current study is explainable in the current
theoretical framework about selective visual attention.
In the current selective visual attention literature, there are two major
theoretical systems about the strategies people use to control their eye movements and
focus their attention: top-down control and bottom-up control (Duchowski, 2007). The
top-down control is typically found in a visual search or visual recognition situation,
which means people are requested to find or identify a specific object in a visual field.
In this situation, the internal representation or the features of the object usually were
stored in the long-term working memory, people’s attention is directed to a specific
portion of a visual field by the signals arising from working memory (Moores, Laiti, &
Chelazzi, 2003). The bottom-up control was widely found in various visual attention
tasks, including reading and visual search (Keith, 2009). Todd and Kramer (1993)
mentioned that people’s visual attention is likely to fixate at the salient features
including edges, corners, special colors, but not plain surfaces.
In the current research, participants were requested to make judgments
about whether the garment would fit by looking at the visual image of a human model
95
wearing the garment. Apparel experts and researchers have identified five major
features for evaluating garment fit: grain, set, line, balance, and ease. (Erwin, Kinchen,
& Peters, 1979). Although consumers might not know these five rules for evaluating
fit, they have their own knowledge and understanding about the “fit” concept. For
example, “set” was defined as absence of wrinkles on the garment (Erwin et al., 1979),
which may be intuitively adopted by consumers as their criteria for judgment of
garment fit, hence, wrinkles would be an salient feature which attract consumers
attention when they evaluate garment fit. Applying a top-down control view
(Humphrey & Underwood, 2009) to an apparel fit situation, consumers’ knowledge
and understanding about garment fit stored in their memory would likely direct their
visual attention to fixate at some specific area. In current study, the participants tended
to fixate at the waist and bust for evaluating the fit of a blazer, perhaps because their
previous knowledge and understanding about fit drive them to obtain information from
these body sites. Applying a bottom-up control view (Treisman, 1982) to an apparel fit
situation, the body sites, such as the waist and bust, of a blazer contains many
construction details and draping effects, such as button, pocket, hem, and wrinkles,
which are likely to be salient features that attract people’s attention (Figure. 5.1).
Generally, some body sites, such as waist, bust, and buttocks might be
selected as areas of interest in the garment fit evaluation process, on which people
would exert more visual attention, because salient features for fit judgment (i.e.,
wrinkles) often appear in these body sites. This result has implications for the online
apparel retailers. Online apparel retailers could increase the accuracy of the visual
image of the garment at some specific areas such as waist and bust to highlight the
96
construction details and draping effects, which would attract consumers’ attention and
help them evaluate the fit of the garment. For example, online apparel retailers could
make bust and waist of a blazer have higher zooming rate while other body sites (e.g.,
neck and arm) have lower zooming rate.
Figure 5.1 Construction details and draping effects might draw attention
Conclusions and Implications
The results of the present study indicated: 1) there is significant
relationship between female college consumers’ body image discrepancy and their
body satisfaction-Hypothesis 1 was supported; 2) female consumers’ body image
97
discrepancy and body satisfaction significantly predicted their visual attention over the
garments (how long and how often they looked at the garments) when making garment
fit judgments-Hypotheses 2 and 3 were supported; 3) no relationships between female
consumers’ visual attention and their concern with garment fit-Hypothesis 4 was not
supported; 4) there is no significant relationship between female consumers’ concern
with garment fit and their confidence in fit judgment-Hypothesis 5 was not supported;
5) there is no significant relationship between female consumers’ concern with fit and
purchase intention- Hypothesis 7 was not supported; 6) female consumers’ concern
with fit was a significant predictor of their purchase intent for the garment- Hypothesis
6 was supported; 7) Consumers attend to some human body sites more than others-
Hypothesis 8 was supported.
Fit is a critical issue in buying a closely fitting garment, such as blazer,
especially when the shopping behavior occurs online. Consumers need visual feedback
and tactile feedback to make a judgment about garment fit; however, the tactile
information is unavailable and the visual information is very limited in the context of
online apparel shopping. Consumers obtain the visual information for fit judgments
from the visual display of garments presented on the website and from previous
experience with the brand, while in the real-store environment consumers are able to
perceive the visual information from a mirror. The current study used the eye-tracking
technology to record consumers’ visual attention in terms of fixations when they look
at the visual image of a human model wearing garments of different sizes (to simulate
the situation that consumers look at the mirror when they try on garments of different
sizes). This study analyzed the effects of subjective factors (body image discrepancy
98
and body cathexis) and objective factors (body sites) on consumers’ visual attention
during the process of perceiving information for making a judgment of garment fit;
examined the impact of consumers’ visual attention on their concern with fit; and
analyzed relationships among consumers’ concern with fit, confidence of judgment
and purchase intention. The results of this study have some managerial implications
for online apparel retailers.
The research finds that consumers’ body image discrepancy was highly
related to their body cathexis, which means people whose perceived body deviates
from their ideal body tend to be less satisfied with their body. This result is also
supported by body image research (Thompson et al., 1999). The research also found
that consumer’s body image discrepancy and body satisfaction could influence their
visual attention in perceiving information for evaluating fit. Consumers who have
large body image discrepancies and are less satisfied with their bodies were prone to
fixate longer and more often at the visual image of a human model when they made a
judgment of fit. According to self-discrepancy theory, consumers who have large body
image discrepancy and high level of body dissatisfaction may experience a negative
psychological state. Kuo, Hsu, and Day’s research (2009) indicated that individuals
under negative emotion would exert more cognitive effort (measured by duration of
fixations and number of fixations) on gathering information to make a decision.
Hence, consumers who have large body image discrepancies and a high level of body
dissatisfaction would expend more cognitive effort (in terms of longer duration of
fixations and larger number of fixations) on visual images of garment to make a
judgment of garment fit. This result also supports the theory that people’s affective
99
state influences their response to the external environment (Eroglu, Machleit, & Davis,
2001).
The other important finding of this research is that consumers’ visual
attention is drawn to certain body sites of a human model when making a judgment of
fit of a blazer. In the front view of the model, the waist and bust are the two highly
fixated on areas (with long duration and large number of fixations), while waist and
buttocks are highly fixated on in the back view of the model. The reasons that people
fixate longer and more often at these specific areas can be explained using from two
perspectives: top-down and bottom up processing. From the top-down view, fit
knowledge, which consumers learn from previous shopping experience and now stored
in their memory, directs their attention to the body sites (i.e. waist) which may contain
important information for evaluating fit. From the bottom-up view, the visual image of
body sites contain salient features (i.e. edge, corners, special color) which tend to stand
out and draw people’s attention. No matter which perspective is used to explain the
results, some body sites (i.e., waist, bust, buttocks) are very important in consumers’
visual perception of garment fit. Online apparel retailers, who want to facilitate
consumers’ fit judgments, should make the visual display of the garment especially
clear and accurate at these body sites.
The current study did not find a significant relationship between
consumers’ visual attention and concern with fit. Perhaps the instrument used in this
study is not sensitive enough to assess the concept of concern with fit. In future
studies, the researcher could include more items to obtain a better measure of concern
with fit. This research does confirm that consumers’ concern with fit has a negative
100
impact on their purchase intention. The result supports the previous research that fit is
one of the major reasons that keep consumers away from buying clothes online. For
the online apparel retailers, there is a urgent need to develop new online visualization
tools to reduce consumers’ concern with fit in order to facilitate online apparel
shopping.
Better online virtual models and more efficient online fitting tools could
be the methods to fulfill this goal. Some companies have developed new online fitting
technology. Alcatel-Lucent, Hachette Filipacchi Media U.S., Inc., Samsung, [TC]²,
TelePresence Tech, VisionMAX, and Vidyo have collaborated to develop an
interactive Virtual Personal Stylist platform, which offers an accurate and reliable
online fitting experience with high bandwidth networks. Consumers’ personal avatar
with accurate body measurements can be created in an in-store [TC] ² 3D scanning
booth. Consumers’ scanning profile is then stored in an online data storage and sharing
application (a cloud) and can be retrieved in retail stores via an interactive “mirror,”
kiosk, or mobile device to try on clothes. This platform will allow a consumer to mix-
and-match a large variety of garments in a very short time without repeatedly taking
off their clothes ([TC] ² Technology Communicator, 2011). The findings of the current
study suggest that the draping effect of garments of different sizes needs to be
displayed on the virtual avatar to provide more accurate online fitting information. In
addition, the accuracy (i.e., the amount of pixels per inch) of the visual display of
garments at waist, bust, and buttocks should be increased to draw consumers’ attention
and provide more visual information to facilitate consumers’ judgments of fit.
101
Limitations
Although the research found some useful results, there are some
limitations in this research. First, many body related variables besides body image
discrepancy and body cathexis should be included in this research. For example, body
boundary aberration was a significant predictor for consumers’ concern with fit in
online shopping (Rao et al., 2006), which means body boundary aberration might also
be an important factor in visual perception of garment fit.
Second, although 4 sizes of garments are used in this research, the visual
attention differences caused by the size differences were not examined. In the present
study, because the within-subjects design was used and the sequence of presenting
garment stimuli was not randomized, there could be an order effect. This order effect
could mask the visual attention differences induced by the differences in garment
sizes. In future study, the garment size effect could be examined using a between-
subject design.
Third, the sample number of this study was small; it might be the reason
why some hypotheses could not be supported in this research. In future research, more
participants should be recruited to make the research more generalizable.
Lastly, the eye tracker used in this study was a desk-mounted setting,
which means participants needed to sit in front of a computer screen to view the visual
stimuli. In the future study, if researchers are able to use a head-mounted eye-tracker,
which is commercially available, they might be able to simulate an online fitting room
more accurately. Participants can wear the head-mounted eye-tracker and inspect their
looks in a mirror. This setting would allow researchers to obtain more accurate results.
102
Appendices
APPENDIX A- GARMENT STIMULI
Figure A.1 Front-view and back view of blazer of size 2
103
Figure A.2 Front-view and back view of blazer of size 4
104
Figure A.3 Front-view and back view of blazer of size 6
105
Figure A.4 Front-view and back view of blazer of size 8
106
APPENDIX B - BODY CATHEXIS SCALE AND BODY IMAGE
DISCREPANCY SCALES
Body cathexis scale
Use this 1 to 9 scale to indicate how dissatisfied or satisfied you are with each of the
following areas or aspects of your body:
1 2 3 4 5 6 7 8 9
Very
Dissatisfied
Neither
Satisfied
Nor
Dissatisfied
Very
Satisfie
d
______ 1. How dissatisfied or satisfied you feel about your neck?
______ 2. How dissatisfied or satisfied you feel about your shoulder
______ 3. How dissatisfied or satisfied you feel about your bust ?
______ 4. How dissatisfied or satisfied you feel about your back?
______ 5. How dissatisfied or satisfied you feel about your arm?
______ 6. How dissatisfied or satisfied you feel about your abdomen
______ 7. How dissatisfied or satisfied you feel about your waist?
______ 8. How dissatisfied or satisfied you feel about your hip?
______ 9. How dissatisfied or satisfied you feel about your buttocks?
107
Body Image Discrepancy Scale
Use this 0 to 3 scale to indicate how you feel about each areas of your body.
1-a. My ideal neck is:
0 1 2 3
Exactly As I Am Almost As I
Am
Fairly
Unlike Me
Very Unlike
Me
1-b. How important to your appearance if your neck is ideal?
0 1 2 3
Not Important Somewhat
Important
Moderately
Important
Very
Important
2-a. My ideal shoulder is:
0 1 2 3
Exactly As I Am Almost As I
Am
Fairly
Unlike Me
Very Unlike
Me
2-b. How important to your appearance if your shoulder is ideal?
0 1 2 3
Not Important Somewhat
Important
Moderately
Important
Very
Important
3-a. My ideal back is:
0 1 2 3
Exactly As I Am Almost As I
Am
Fairly
Unlike Me
Very Unlike
Me
3-b. How important to your appearance if your back is ideal?
0 1 2 3
Not Important Somewhat
Important
Moderately
Important
Very
Important
4-a. My ideal bust is:
0 1 2 3
108
Exactly As I Am Almost As I
Am
Fairly
Unlike Me
Very Unlike
Me
4-b. How important to your appearance if your bust is ideal?
0 1 2 3
Not Important Somewhat
Important
Moderately
Important
Very
Important
5-a. My ideal arm is:
0 1 2 3
Exactly As I Am Almost As I
Am
Fairly
Unlike Me
Very Unlike
Me
5-b. How important to your appearance if your arm is ideal?
0 1 2 3
Not Important Somewhat
Important
Moderately
Important
Very
Important
6-a. My ideal waist is:
0 1 2 3
Exactly As I Am Almost As I
Am
Fairly
Unlike Me
Very Unlike
Me
6-b. How important to your appearance if your waist is ideal?
0 1 2 3
Not Important Somewhat
Important
Moderately
Important
Very
Important
109
APPENDIX C- CONCERN WITH FIT, CONFIDENCE IN FIT JUDGMENT,
AND PURCHASE INTENT SCALES
Concern with fit and size scale
Please rate the extent to which you agree with the following statement using a 5
point scale (1=strongly disagree, 5=strongly agree)
Strongly
Disagree
1
2
3
4
Strongly
Agree
5
__________1. The blazer of this size may not fit me.
__________2. The blazer of this size may not look good on me.
__________3. I may have a hard time picturing myself wearing the garment.
Confidence of Judgment scale
not
confident/certain
at all
1
2
3
4
very
confident
5
______1. Please rate how confident you are about your fit evaluation of garments.
______2. To which extent you are certain about this size of blazer is fit for you.
Purchase intention scale
Strongly
Disagree
1
2
3
4
Strongly
Agree
5
________1. I would like to buy a blazer of this size from this retailer.
110
APPENDIX D- PERMISSION FOR USING PHOTOS OF HUMAN MODEL
111
APPENDIX E UNIVERSITY OF DELAWARE IRB APPROVAL
112
REFERENCES
Alexander,M., Connell, L. J., & Presley, A. B.(2005) Clothing fit preferences of young
female adult consumers. International Journal of Clothing Science and
Technology, 17(1), 52-64.
Altabe, M. N., & Thompson, J. K. (1992). Size estimation vs. figural ratings of body
image disturbance: Relation to body dissatisfaction and eating dysfunction.
International Journal of Eating Disorders, 11, 397-402.
Ashdown, S. P. & Delong, M. (1995). Perception testing of apparel ease variation.
Applied Ergonomics, 26(1),47-54.
Ashdown, S. P. & Dunne, L. (2006). A Study of Automated Custom Fit: Readiness of
the Technology for the Apparel Industry. Clothing and Textiles Research
Journal, 24(2), 121-136.
Ashdown, S.P. & DeLong, M. (1995). Perception testing of apparel ease variation.
Applied Ergonomics, 26(1), 47-54.
Ashdown, S.P. &DeLong, M. (2007). Size in clothing ready-to-wear. NY.
Ashdown,S.P., & O'Connell, E. K.(2006).Comparison of test protocols for judging the
fit of mature women's apparel. Clothing and Textiles Research Journal, 24(2),
137-146.
Beck, B. (2003). Key strategic issues in online apparel retailing. Retrieved March 6,
2010, from http://www.techexchange.com/thelibrary/online_fit.html
Bennett, P. D. & Harrell, G. D. (1975). The Role of confidence in understanding and
predicting buyers' attitudes and purchase intentions. Journal of Consumer
Research, 2(2), 110-117.
Bhatnagar, A., Misra, S., Rao, H.R., (2000). On risk, convenience, and Internet
shopping behavior. Communications of the ACM, 43 (1), 98–114.
Blackwell, R.D., Miniard, P.W. and Engel, J.F. (2001), Consumer behavior. The
Dryden Press, Orlando, FL.
113
Brown, P. & Rice, J. (2001). Ready-To-Wear apparel analysis (Third Edition). Upper
Saddle River, New Jersy: Prentice-Hall, Inc.
Brown, T. A., Cash, T. F. & Mikulka, P. J. (1990). Attitudinal body-image assessment:
Factor analysis of the Body Self-Relations Questionnaire. Journal of
Personality Assessment, 55, 135-144.
Cash, T. F. (1990). The psychology of physical appearance: Aesthetics, attributes, and
images. In T.F. Cash & T. Pruzinsky (Eds.), Body images: Development,
deviance and change (pp. 51-79). New York: Guilford Press.
Cash, T. F. (1995). Developmental teasing about physical appearance: Retrospective
descriptions and relationships with body image. Personality and Social
Behavior: An International Journal, 23, 123-130.
Cash, T. F., & Deagle, E.A. (1997). The nature and extent of body-image disturbances
in anorexia nervosa and bulimia nervosa: A meta-analysis. International
Journal of Eating Disorders, 22, 107-125.
Cash, T. F., & Szymanski, M. L. (1995). The development and validation of the Body-
Image Ideals Questionnaire. Journal of Personality Assessment, 64, 466-477.
Cash, T.F., & Strachan, M.D. (1999). Body images, eating disorders, and beyond. In
R. Lemberg (Ed.), Eating disorders: A reference sourcebook (pp. 27-36).
Phoenix, AZ: Oryx Press.
Cash, T.F., Novy, P.L., & Grant J.R. (1994). Why do women exercise? Factor analysis
and further validation of the Reasons for Exercise Inventory. Perceptual &
Motor Skills, 78, 539-544.
Chau, P.Y.K., Tam, G.A.K.Y., 2000. Impact of information presentation modes
on online shopping: an empirical evaluation of a broadband interactive
shopping service. Journal of Organizational Computing and Electronic
Commerce, 10 (1), 1–22.
Chen, W. & Swalm, R. (1998). Chinese and American college students’ body-image:
Perceived body shape and body affect. Perceptual & Motor Skills, 87, 395-403.
114
Chen-Yu, H. J., and Kincade, D. H. (2001). Effects of product image at three stages of
the consumer decision process for apparel products: Alternative evaluation,
purchase and post purchase. Journal of Fashion Marketing and Management,
5(1), 29-43.
Citrin, A., Stern, D., Spangenberg, E., Clark, M. (2003). Consumer need for tactile
input. An internet retailing challenge. Journal of Business Research, 56(11),
915-922.
Connell, L. J., Ulrich, P. V., Brannon, E. L., Alexander, M. & Presley, A. B. (2006).
Body shape assessment scale: Instrument development for analyzing female
figures. Clothing and Textiles Research Journal, 24, 80-95.
Cox, D. F. & Stuart, U. R. (1964). Perceived risk and consumer decision making-
the case of telephone shopping. Journal of Marketing Research, 1, 32-39.
Delong, M., Kim, S. H., & Larntz, K. (1993). Perceptions of garment proportions by
female observers. Perceptual & Motor Skills, 76, 811-819.
Duchowski, A. T. (2007). Eye tracking methodology: Theory and practice. London:
Springer.
Elliot, S., Fowell, S., 2000. Expectations versus reality: a snapshot of consumer
experiences with Internet retailing. International Journal of Information
Management, 20, 323–336.
Eroglu, S. A., Machleit, K. A., & Davis, L. M. (2001) Atmospheric qualities of online
retailing: A conceptual model and implications. Journal of Business Research,
54, 177-184.
Erwin, M., Kinchen, L., & Peters, K. (1979). Clothing for moderns (6th ed.).
Englewood Cliffs, NJ: Prentice Hall.
Fan, J., Yu, W. & Hunter, L. (2004). Clothing appearance and fit: Science and
technology. Boca Raton: CRC Press LLC.
Feather,B., Herr,D., & Ford, S.(1997). Black and white female athletes' perceptions of
their bodies and garment fit. Clothing and Textiles Research Journal, 15(2):
125 -128.
115
Ferwerda, J.A (2003) Three varieties of realism in computer graphics. Proceedings
SPIE Human. Vision and Electronic Imaging, 3, 290-297.
Ferwerda, J. A., Westin, S. H., Smith, R. C., Pawlicki, R. (2004). Effects of rendering
on shape perception in automobile design. ACM Symposium on Applied
Perception in Graphics and Visualization, 1, 107-114.
Fitzgibbon, M. L., Blackman, L. R., & Avellone, M. E. (2000). The relationship
between body image discrepancy and body mass index across ethnic groups.
Obesity Research, 8(8), 582-589.
Forsythe, S., Liu, C., Shannon, D., & Gardner, L. C. (2006). Development of a scale to
measure the perceived benefits and risks of online shopping. Journal of
Interactive Marketing, 20, 55-75.
Forrester Research. (March, 2010). Forrester forecast: Double digit growth for online
retail in US and Western Europe. Retrieved April 13, 2010 from
http://www.businesswire.com/portal/site/home/permalink/?ndmViewId=.
Garner, D. & Garfinkel, P.(1981). Body image in anorexia nervosa: Measurement
theory and clinical implications. International Journal of Psychiatry and
Medicine, 2 (11), 263-284.
Garner, D.M., Olmstead, M.A., & Polivy, J. (1983). Development and validation of a
multidimensional eating disorder inventory for anorexia nervosa and bulimia.
International Journal of Eating Disorders, 2, 15-34.
Goldsberry, E., Shim, S. & Reich, N. (1996). Women 55 years and older: Overall
satisfaction and dissatisfaction with the fit of ready-to-wear. Clothing and
Textiles Research Journal, 14(2), 121-132.
Ha, Y., Kwon, W., & Lennon, S. J. (2007). Online visual merchandising (VMD) of
apparel web sites. Journal of Fashion Marketing and Management, 11(4), 477-
493.
Hagen, M. (1986). Varieties of Realism. Cambridge University Press.
Hammond, J. & Kohler, K.(2001) E-commerce in the textile and apparel industries. In
tracking a transformation: E-commerce and the terms of competition in
Industries. The BRIE-IGCC Economy Project. Brookings Institute.
116
Häubl, G. & Trifts, V. (2000). Consumer decision making in online shopping
environments: The effects of interactive decision aids. Marketing Science,
19(1), 4-21.
Heinberg, L. J., & Thompson, J. K. (1995). Body image and televised images of
thinness and attractiveness: A controlled laboratory investigation. Journal of
Social and Clinical Psychology, 14, 325-338.
Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect.
Psychological Review, 94(3), 319-340.
Higgins, E. T., Klein, R., & Strauman, T. (1987). Self-concept discrepancy theory: A
psychological model for distinguishing among different aspects of depression
and anxiety. Social Cognition, 3, 51-76.
Hollingworth, A., Richard, A. M., & Luck, S. J. (2008).Understanding the function
of visual short-term memory: Transsaccadic memory, object correspondence,
and gaze correction. Journal of Experimental Psychology: General, 137, 163-
181.
Horrigan, J. B. (2008, February). Online shopping: Internet users like the convenience
but worry about the security of their financial information. Retrieved April 25,
2010, from http://www.pewinternet.org/Reports/2008/Online-Shopping.aspx
Howard, J. A. & Sheth, J. N. (1969). The theory of buyer behavior. New York: John
Wiley and Sons, Inc.
Humphrey, K., & Underwood, G. (2009). Domain knowledge moderates the influence
of visual saliency in scene recognition. British Journal of Psychology, 100,
377–398.
Irving, L. (1990). Mirror images: Effects of the standard of beauty on the self- and
body-esteem of women exhibiting varying levels of bulimic symptoms.
Journal of Social and Clinical Psychology, 9(2), 230-242.
Istook, C.(2008). Three-dimensional body scanning to improve fit, In C. Fairhurst
(Eds.), Advances in apparel production, Cambridge, England : Woodhead
Publishing. 94-116
117
Istook, C., & Hwang, S. (2001). 3D body scanning systems with application to the
apparel industry. Journal of Fashion Marketing and Management, 5(2), 120-
132.
Jacob, R. J. K., & Karn, K. S. (2003). Eye tracking in Human-Computer Interaction
and usability research: Ready to deliver the promises, In J. Hyona, R. Radach,
& H. Deubel (Eds.), The mind’s eye: cognitive and applied aspects of eye
movement research (573-605). Amsterdam: Elsevier.
Jacobi, L., & Cash, T.F. (1994). In pursuit of the perfect appearance: Discrepancies
among self- and ideal-percepts of multiple physical attributes. Journal of
Applied Social Psychology, 24, 379-396.
Jarvenpaa, S.L., Todd, P.A., 1997. Consumer reactions to electronic shopping on the
Word Wide Web. International Journal of Electronic Commerce, 1 (2), 59–
88.
Ju, H. W., & Johnson, K. K. P. (2010). Fashion advertisements and young women:
Determining visual attention using eye tracking. Clothing and Textiles
Research Journal, 28(3), 159-173.
Jung, J., & Lennon, S. J. (2003). Body image, appearance self-schema, and media
images. Family and Consumer Sciences Research Journal, 32(3), 27-51.
Jung, J., Lennon, S. J., & Rudd, N. A.. (2001). Self-schema or self-discrepancy?
Which best predicts body image? Clothing and Textiles Research Journal, 19,
171-184.
Kanwisher, N. & Driver, J. (1992).Objects, Attributes, and Visual Attention: Which,
what, and where. Psychological Science, 1(1), 26-31.
Kerr, A. W., Hall, H. K., Kozub, S. A. (2002). Doing statistics with SPSS.
Trowbridge, Wiltshire: The Cromwell Press.
Khakimdjanova, L., and Park, J.H.(2005). Online visual merchandising practice of
apparel e-merchants. Journal of Retailing and Consumer Services, 12, 307-
318.
118
Kim, Dong-Eun (2009). Apparel fit based on viewing of 3D virtual models and living
models. International Textile and Apparel Association 2009 Preliminary
Conferences. Retrieved from
http://dha.design.umn.edu/programs/grad/documents/Poster_DongEunKim-
1_000.pdf
Kim, H., & Lennon, S. J. (2010). E-atmosphere, emotional, cognitive, and behavioral
responses. Journal of Fashion Marketing and Management, 14(3), 412-428.
Kim, H.J.and Damhorst, M. L. (2010). The relathionship of body-related self-
discrepancy to body dissatisfaction, apparel involvement, concerns with fit and
size of garments, and purchase intention in online apparel shopping. Clothing
and Textiles Research Journal, 28(4), 239-254.
Kim, J., Fiore, A. M., & Lee, H.-H. (2007). Influences of online store perception,
shopping enjoyment, and shopping involvement on consumer patronage
behavior toward an online retailer. Journal of Retailing and Consumer
Services, 14(2), 95-107.
Kim, J-H., Kim, M., & Lennon, S. J. (2007). Information components of apparel retail
websites: Task relevance approach. Journal of Fashion Marketing and
Management, 11(4), 494-510.
Kim, M., & Lennon, S. J. (2008). The effects of visual and verbal information on
attitudes and purchase intentions in Internet shopping. Psychology &
Marketing, 25, 146-178.
Kinsbourne, M., & Lempert, H. (1980). Human figure representation by blind
children. Journal of General Psychology, 102, 201–209.
Knoblich, G., Thornton,I., Grosjean, M. & Shiffar,M. (2006). Human body perception
from the inside out, Oxford University Press, Oxford, UK.
Kozar, J. M. (2010). Women’s responses to fashion media images: a study of female
consumers aged 30–59. International Journal of Consumer Studies, 34(3),272-
278.
Kozar, J. M., and Damhorst, M. L.(2009). Comparison of the ideal and real body as
women age: Relationships to age identity, body satisfaction and
importance,and attention to models in advertising. Clothing & Textiles
Research Journal, 27(3), 197-210.
119
Kozar,J. M. & Damhorst, M. L.(2008). Older women's responses to current fashion
models. Journal of Fashion Marketing and Management, 12(3),338-350.
Kuo, F. Y., Hsu, C. W. & Day, R. F. (2009). An exploratory study of cognitive effort
involved in decision under Framing: An application of the eye-tracking
technology. Decision Support System, 48, 81-91.
LaBat, K. L. & DeLong, M. R. (1990). Body cathexis and satisfaction with fit of
apparel. Clothing and Textiles Research Journal, 8(2), 43-48.
Laroche,M., Kim, C., Zhou,L. (1996). Brand familiarity and confidence as
determinants of purchase intention: An empirical test in a multiple brand
context. Journal of Business Research, 37, 115-120.
Lee, E. J., and Park, J.K.(2009). Online service personalization for apparel shopping.
Journal of Retailing and Consumer Services, 16, 83-91.
Lee, G.-G., & Lin, H.-F. (2005). Customer perceptions of e-service quality in online
shopping. International Journal of Retail & Distribution Management, 33(2),
161-176.
Lim, H. S. (2009). Three dimensional virtual try-on technologies in the achievement
and testing of fit for mass customization. Unpublished Doctor dissertation ,
North Carolina State University, Raleigh.
Mahoney, E. R. & Finch, M. D. (1976). The dimensionality of body-cathexis. Journal
of Psychology, 92, 277-279.
Markee, N., Carey, I., & Pedersen, E.(1990). Body cathexis and clothed body cathexis:
Is there a diff erence? Perceptual and Motor Skills, 70, 1239-1244.
McBurney, D. & Collins, V., 1977. Introduction to sensation/perception, Prentice-Hall
Inc., Englewood Cliffs, NJ.
McQuitty, M. A., & Peterson, R. T. (2000). Selling home entertainment on the
Internet: An overview of a dynamic market place. Journal of Consumer
Marketing, 17(3), 233-248.
Mollenkopf, D., Rabinovich, E., Laseter, T. & Boyer, K. (2007) Managing internet
product returns: A focus on effective service management. Decision Sciences,
38(2), 2007
120
Monteath, S., & McCabe, M. (1997). The influence of societal factors on female body
image. Journal of Social Psychology, 137(6), 708-727.
Moores, E., Laiti, L., & Chelazzi, L.(2003). Associative knowledge controls
deployment of visual selective attention. Nature Neuroscience, 6(2), 182-189.
Nguyen, H. T., Isaacowitz, D. M., & Rubin, P. A. D. (2009). Age and fatigue related
markers of human faces: An eye tracking study. Opthamology, 115, 355-360.
Park, J. H., & Stoel, L. (2002). Apparel shopping on the Internet: Information
availability on US apparel merchant web sites. Journal of Fashion Marketing
and Management, 6(2), 158-176.
Park J. H., Lennon, S. J., & Stoel, L. (2005). Online product presentation: Effects on
mood, perceived risk, and purchase intention. Psychology & Marketing, 22(9),
695-719.
Park, J. H., Stoel, L., & Lennon, S. J. (2008). Cognitive, affective, and conative
responses to visual simulation: The effects of rotation in online product
presentation. Journal of Consumer Behaviour, 7, 72-87.
Park, J., & Stoel, L. (2005). Effects of brand familiarity, experience and information
on online apparel purchase. International Journal of Retail & Distribution
Management, 33(2), 148-160.
Peterson, D. K., and Pitz, G. F. (1988). Confidence, uncertainty, and the use of
information. Journal of Experimental Psychology: Learning, Memory, and
Cognition, 14, 85-92.
Petrie, T. A., Tripp, M. M., & Harvey, P. (2002). Factorial and construct validity of the
body parts satisfaction scale-revised: An examination of minority and
nonminority women. Psychology of Women Quarterly, 26, 213–221.
Pieters, R. and Warlop, L. (1999).Visual attention during brand choice: The impact of
time pressure and task motivation. International Journal of Research in
Marketing, 16, 1–16.
Pieters, R., & Wedel, M. (2004). Attention capture and transfer in advertising: Brand,
pictorial, and text-size effects. Journal of Marketing, 68(2), 36-50.
121
Pisut, G. & Connell, L. J. (2007). Fit preferences of female consumers in the USA.
Journal of Fashion Marketing and Management, 11(3), 366-379.
Rayner Keith (2009). Eye movements and attention in reading, scene perception, and
visual search. The Quarterly Journal of Experimental Psychology, 62(8), 1457-
1506.
Rosa, J. A., Garbarino, E. C., & Malter, A. J. (2006). Keeping the body in mind: The
influence of body esteem and body boundary aberration on consumer beliefs
and purchase intention. Journal of Consumer Psychology, 16, 79-91.
Rosen, G., & Ross, A. (1973). The relationship of body image to self concept.
Unpublished paper, University of Pittsburgh, Pittsburgh.
Rudd, N. A., & Lennon, S. J. (1994). Linkages between attitudes toward gender roles,
body satisfaction, self-esteem and appearance management behaviors in
women. Family and Consumer Science Research Journal, 23, 94-117.
Schwarz,N.,Bless,H. & Bohner, G. (1991). Mood and persuasion: affective states
influence the processing of persuasive communications. Advances in
Experimental Social Psychology, 24, 161–199.
Secord, P. F. and Jourard, S. M. (1953). The appraisal of body-cathexis: Body cathexis
and the self. Journal of Counseling Psychology, 17(5), 343-347.
Shim, S., Eastlick, M. A., Lotz, S. L., & Warrington, P. (2001). An online pre-
purchase intentions model: The role of intention to search. Journal of
Retailing, 77(3), 397-416.
Shim, S., Kotsiopulos, A. & Knoll, D. S. (1991). Body cathexis, clothing attitude, and
their relations to clothing and shopping behavior among male consumers.
Clothing and Textiles Research Journal, 9(3), 35-44.
Sicilia, M. & Ruiz, S. (2010). The effect of web-based information availability on
consumers' processing and attitudes. Journal of Interactive Marketing, 24, 31–
41.
Slyke, C.V., Comunale, C.L., Belanger, F., 2002. Gender differences inperceptions of
Web-based shopping. Communications of the ACM, 45 (7), 82–86.
122
Sniezek, J. A. (1992). Groups under uncertainty: An examination of confidence in
group decision making. Organizational Behavior and Human Decision
Processes, 52, 124-155.
Sniezek, J. A., and Henry, R. A. (1989). Accuracy and confidence in group judgment.
Organizational Behavior and Human Decision Processes, 43, 1-28.
Spitz, H., Stark, L., & Noton, D (1971). Scanpaths and pattern recognition. Science,
New Series, 173(3998), p. 753
Strauman, T. J., & Higgins, E. T. (1987). Automatic activation of self-discrepancies
and emotional syndromes: When cognitive structures influence affect. Journal
of Personality and Social Psychology, 53, 1004-1014.
Sundstedt, V., Debattista, K., Longhurst, P., Chalmers, A., & Troscianko, T. (2005).
Visual attention for efficient high-fidelity graphics. Spring Conference on
Computer Graphics. Retrieved from http://
www.cs.bris.ac.uk/home/veronica/Documents/SCCG05_sundstedt.pdf
Szymanski, D.M. & Hise, R.T. (2000). E-satisfaction: an initial examination. Journal
of Retailing, 76(3), 309-322.
[TC]² (2010, March). [TC]² [On-line]. Available: Retrieved March 16, 2010, from
http://www.tc2.com/products
[TC]² Technology Communicator.(2011). Virtual personal stylist concept using
[TC]²’s 3D scanning showcased at NG CONNECT AT CES 2011. Retrieved
from http://www.tc2.com/newsletter/2011/012611.html
Then, N.K. & DeLong, M.R., 1999. Apparel shopping on the web. Journal of
Family and Consumer Sciences, 91 (3), 65–68.
Thompson, J. K. (1990). Body image disturbance: Assessment and treatment.
Elmsford, NY: Pergamon Press.
Thompson, J. K., Heinberg, L. J., Altabe, M., & Tantleff-Dunn, S. (1999). Exacting
Beauty: Theory, assessment, and treatment of body image disturbance.
Washington, DC: American Psychological Association.
123
Todd, S., & Kramer, A. F. (1993). Attentional guidance in visual attention. Paper
presented at 37th Annual Meeting of Proceedings of the Human Factors and
Ergonomics Society.
Torralba, A., Oliva, A., Castelhano, M. S., & Henderson, J. M. (2006). Contextual
guidance of eye movements and attention in real-world scenes: The role of
global features in object search. Psychological Review, 113, 766-786.
Treisman, A. (1982). Perceptual Grouping and Attention in Visual Search for Features
and for Objects. Journal of Experimental Psychology: Human Perception and
Performance, 8(2), 194-214.
Urbany, L. E., Dickson, P. R., Wilkie, W. L.(1989). Buyer uncertainty and information
search. Journal of Consumer Research, 16(2), 208-215.
US Census Bureau News (2010). Quarterly retail e-commerce sales 4th quarter 2009.
Retrieved from: http://www.census.gov/retail/mrts/www/data/html/09Q4.html
Volino, P., Cordier,C. & Magnenat-Thalmann, N. (2005). From early virtual garment
simulation to interactive fashion design. Computer-Aided Design, 37, 593-608.
Wendel, G. & Lester, D. (1988). Body-cathexis and self-esteem. Perceptual & Motor
Skills, 67, 538.
Williamson, D. A., Gleaves, D. H., Watkins, P. C., and Schlundt, D. G. (1993).
Validation of self-ideal body size discrepancy as a measure of body
dissatisfaction. Journal of Psychopathology and Behavioral Assessment, 15(1),
57-68.
Workman, J. E. & Lentz, E. S. (2000). Measurement specifications for manufacturers’
prototype bodies, Clothing and Textiles Research Journal, 18(4), 251-259.
Yoo, W. S., Lee, Y. J., & Park, J. K. (2010). The role of interactivity in e-tailing:
Creating value and increasing satisfaction. Journal of Retailing and Consumer
Services, 17(2), 89-96.
Related Documents
