Content uploaded by Te-Lin Chung
Author content
All content in this area was uploaded by Te-Lin Chung on Apr 15, 2020
Content may be subject to copyright.
Journal of Fashion Marketing and Management
Evaluating garments in Augmented Reality when shopping
online
Journal:
Journal of Fashion Marketing and Management
Manuscript ID
JFMM-05-2018-0077.R4
Manuscript Type:
Original Article
Keywords:
Augmented Reality, Apparel, Virtual, Fit and size, The Stimulus-
Organism-Response model, Product performance
Journal of Fashion Marketing and Management
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
1Evaluating garments in Augmented Reality when shopping online
2Introduction
3 According to a recent report (Narvar, 2017), online shoppers returned apparel more than goods
4 from any category (43%); 70% of apparel is returned due to being the wrong size or color. In
5 online shopping environments, simulated reality enables consumers to “test drive” products
6 during the pre-purchase stage and decreases product returns (Edvardsson et al., 2005). Especially
7 in fashion, such simulation systems provide companies substantial opportunities by
8 compensating for the lack of experiential shopping through enriching product information with
9 interactive visual cues (Fiore and Jin, 2003).
10 Image interactive technologies (IIT) are website features designed to simulate actual
11 product experiences by enabling online shoppers to (a) view products from different angles; (b)
12 change design features; and (c) see how apparel products look on their bodies/avatars to
13 understand garment fit and appearance (Fiore and Jin, 2003; Fiore et al., 2005; Merle et al.,
14 2012). IIT creates a feeling of presence in online environments, fully immersing shoppers in the
15 environment and enabling interaction. Immersion in an online environment is an important
16 aspect that generates a psychological condition, which is necessary for experiencing presence
17 when there are only visual clues for making purchase decisions (Steuer, 1992; Witmer and
18 Singer, 1998).
19 For online apparel shopping, there are two distinct IIT approaches for virtually trying
20 garments. One approach requires customizing virtual avatars from an existing library of
21 parametric models to represent shoppers’ body measurements and shapes as closely as possible,
22 then trying digital garments on these avatars. Physical and mechanical properties of garments can
23 be modeled three-dimensionally (3D), allowing shoppers to view garments in transparent or
Page 1 of 35 Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
24 tension modes and visually judge where the garment is tight or loose (triMirror, n.d). 3D
25 garments can be also generated from two-dimensional (2D) photographs for virtual try-on and
26 size recommendations (Metail, n.d.). However, such environments with both virtual garments
27 and virtual bodies can create artificial settings and make it difficult for consumers to make real-
28 world connections to the product (Azuma, 1997). As Li et al. (2001) indicated, presenting
29 products in their environmental context is important. Consumers prefer to see products within
30 their intended context, such as “the ring displayed on a hand or the laptop computer presented in
31 an office setting” (Li et al., 2001, p. 28). Therefore, using Augmented Reality (AR) for virtual
32 try-on is another approach gaining popularity when shopping online, as consumers can see
33 garments or accessories on their bodies without spending time customizing avatars.
34 AR technology integrates computer-generated sensory information with a physical
35 environment in real-time. Pine and Korn (2011) described AR as using digital information “to
36 enhance, extend, edit, or amend the way we experience the real world” (p. 36). AR systems
37 appear 3D and can apply to all senses (Azuma, 1997). In order to operate an AR application,
38 users must have access to a display device with a video camera (e.g., smartphones, tablets,
39 computers, or mirror-looking screens). On this display device, users can see their environment
40 while computer-generated images of products are placed on top of the view in real-time
41 (Carmigniani et al., 2011). From this perspective, AR can provide shoppers with an experience
42 that resembles physical interaction (Verhagen et al., 2014) and can potentially compensate for
43 the lack of experiential information in online settings (Kang, 2014; Lee, 2012), thus bridging the
44 gap between online and offline shopping (Huang et al., 2011; Lu and Smith, 2007).
45 Previously, researchers focused on the development, usability, and user acceptance of the
46 AR technology (Chang et al., 2013; Huang et al., 2011; Kang, 2014; Lu and Smith, 2007; Rese
Page 2 of 35Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
47 et al., 2017), AR’s experiential value at the pre-purchase stage (Bulearca and Tamarjan, 2010;
48 Huang and Liu, 2014; Kang, 2014), and AR’s impact on purchase intentions (Beck and Crié,
49 2018; Huang and Liu, 2014; Rese et al., 2017). However, no researcher, to our knowledge, has
50 specifically developed a study to compare consumers’ perceptions of using AR for evaluating
51 garment sizes, fit, and product attributes when shopping online with their responses towards the
52 physical garments once they were “ordered and received.” Therefore, the purpose of the present
53 study was to examine consumers’ perceptions of a garment’s size, fit, product performance,
54 attitudes towards the product, and purchase intentions when using AR virtual try-on in an online
55 shopping context as compared to when physically trying on the real garment. The focus of the
56 present research was women. Specifically, we aimed to understand whether AR virtual try-on
57 could provide a comparable representation of physically trying on a garment in terms of fit, size
58 and product performance, and if there would be a difference between AR virtual try-on and
59 physical try-on regarding their impact on attitudes towards the apparel product and purchase
60 intentions.
61 Literature Review
62 A brief overview of AR apparel applications in online environments
63 In AR environments, apparel applications range from overlaying 2D static front images
64 of garments on the real-time static image of the viewer’s body (e.g., Webcam Social Shopper by
65 Zugara) to 3D, which is simultaneous rendering or dynamic fitting of the garment around the
66 viewer's body to simulate garment drape as the viewer moves (e.g., Magic Mirror). In both cases,
67 AR imagery allows viewers to see immediately how clothes would look on them (Batista, 2013;
68 Huang and Liu, 2014; Pachoulakis and Kapetanaki, 2012). The experience with an AR garment
69 is very similar to holding a garment up to oneself in front of a mirror (Schwartz, 2011). AR
Page 3 of 35 Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
70 applications in the apparel industry are usually developed for websites and mobile devices so that
71 customers can virtually try on clothing and accessories (Carmigniani et al., 2011; Pachoulakis
72 and Kapetanaki, 2012). Zugara, FittingBox, MemoryMirror, and Magic Mirror are some of the
73 developers that provide AR applications to fashion brands. In 2017, Gap collaborated with San
74 Francisco-based start-up Avametric and launched a digital dressing room with AR, where
75 shoppers can create their avatars and try garments on (Avametric, n.d.). In 2018, Amazon
76 patented a magic mirror that uses AR to superimpose garment images to users’ reflections in the
77 mirror in real-time, which can help with an AR-enabled shopping experience on Amazon.com
78 (Boyle, 2018).
79 Although only a few apparel retailers have experimented with AR in their online stores,
80 more should consider using the potential of AR technologies to support consumers’ online
81 shopping (Pantano et al., 2017). Benefits of using AR include (a) providing shoppers with digital
82 help and increased likelihood of exploring more garments, (b) suggesting clothing based on user
83 preferences or fashion trends, (c) reducing the number of returned items, and (d) low technology
84 barriers (Chitrakorn, 2018). Challenges of using AR in online shopping are related to whether
85 these tools can assist shoppers with understanding product performance when making purchase
86 decisions (Pantano et al., 2017).
87 Conceptual framework
88 To explore how consumer perception of apparel products and behavioral intentions
89 would be impacted by AR in online apparel shopping, the Stimulus–Organism–Response (S–O–
90 R) model (Mehrabian and Russell, 1974) was selected. The model proposes that environmental
91 stimuli are associated with behavioral responses, and that environmental stimuli (S) affect
92 organisms (O). Response (R) is the result of the internal (cognitive or emotional) process of the
Page 4 of 35Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
93 organism in the form of approach or avoidance behaviors (Eroglu et al., 2003; Fiore and Kim,
94 2007; Mehrabian and Russell, 1974; Prashar, Sai, and Parsad, 2017; Watson, Alexander, and
95 Salavati, 2018). The S-O-R was applied by numerous researchers to understand the influence of
96 new retail technologies on consumers’ affective and behavioral responses when shopping online
97 (Eroglu et al., 2001; Fiore and Kim, 2007; Prashar et al., 2017; Watson, et al., 2018; Wu et al.,
98 2013) and is considered as a robust model (Watson et al., 2018). Past studies found that AR
99 creates rich sensory experiences and influences mental imagery, resulting in positive emotional
100 and behavioral responses (Park and Yoo, 2020; Watson et al., 2018). In the present study, the S-
101 O-R model was used as a foundation to examine hypotheses. The hypotheses were developed
102 based on the elements in the model: stimuli (i.e., AR and physical try-on), organism (i.e.,
103 telepresence as an internal state) and responses to the stimuli (i.e., attitudes towards the product
104 and purchase intentions)
105 Understanding Stimulus: Perception of apparel products in online AR environments
106 There are only a few studies investigated how garments and fashion accessories would
107 be perceived in online AR environments. In a study conducted by Chang et al. (2013), a real-
108 time 3D dynamic fitting room was developed by using AR and Microsoft Kinect, through which
109 sensors were used to automatically measure participants’ sizes. Findings showed that sizes based
110 on the Kinect measurements were close to participants’ claimed sizes, indicating the potential of
111 using AR for online apparel shopping. Verhagen et al. (2014) examined the differences among
112 three different eyeglass presentation formats (picture, 360-spin application, and AR try-on) on
113 the Ray-Ban website. They found that AR can make users feel significantly more “locally
114 present” as compared to seeing pictures or 360-spin formats of the products, suggesting that
115 retailers who sell products that consumers need to try on before buying can use AR technologies.
Page 5 of 35 Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
116 Online shoppers tend to perceive that products they see on a website may not look, feel,
117 or fit the same as the products they find in a brick-and-mortar store (Yu et al., 2012). For apparel
118 products, these perceptions manifest themselves as risks related to product performance based on
119 three main attributes: visual, tactile, and trial (Yu et al., 2012). Therefore, it is important to
120 measure whether apparel products received after using an AR application for online shopping are
121 close to shoppers’ expectations. Suh and Chang (2006) argued that online shopping
122 environments lead to a discrepancy between online products (pre-purchase) and physical
123 products (post-purchase) as consumers can not touch or try-on the online products. According to
124 the authors, this would result in either finding physical products more satisfactory (i.e., positive
125 disconfirmation), or the opposite (i.e., negative disconfirmation).
126 Previous AR studies examined size selections (Chang et al., 2013), technology
127 acceptance (Pantano et al., 2017), telepresence (Verhagen et al., 2014), interactivity and
128 vividness (Yim et al., 2017), and perceived tactile sensations (Overmars and Poels, 2015) in
129 various AR settings. However, none of these studies specifically addressed if and how shoppers
130 fit into the sizes selected by using an AR application, and if AR products’ expected performance
131 matches actual products’ performance once the online order is received. Therefore, the present
132 study’s results would be beneficial for the researchers when examining AR in online shopping,
133 and help retailers increase benefits and overcome challenges by providing them with
134 experimental data. In light of these needs, the following hypothesis was examined:
135 H1: AR virtual try-on will provide a comparable representation of physical try-on in
136 terms of (a) finding the right size, (b) evaluating fit, and (c) evaluating product performance.
137 Organism: Telepresence
Page 6 of 35Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
138 Research on IIT-supported environments shows that shoppers can see digital product
139 attributes through a variety of rich visual cues, using their gestures to control the environment
140 (Huang and Lui, 2014; Merle et al., 2012). Because of this increased interaction, products are
141 experienced in “the mind’s eye,” which can potentially provide accurate sensory information
142 (e.g., touch, taste, and smell) based on real-world experiences with similar products (Schlosser,
143 2003). Telepresence is defined as a consumer’s sense of being present in a virtual environment,
144 such as an online store, where consumers could browse and shop as they would in a brick-and-
145 mortar location (Mollen and Wilson, 2010; Shih, 1998). Lim and Ayyagari (2018) described it as
146 “the perception of direct product experience simulated through a medium” (p.361). Telepresence
147 provides a good basis to understand consumers’ immersion and information processing in the
148 online AR context, as literature has found that telepresence is crucial for consumer immersion in
149 virtual environments (Steuer, 1992). Sense of telepresence is created by the quality and quantity
150 of simulated sensory information in the virtual space (Fiore et al., 2005), particularly the
151 perceived interactivity and virtuality, both characteristics that set AR apart from more traditional
152 forms of online shopping (Javornik, 2016). Coming into contact with digital products in AR can
153 enrich product experiences. Additionally, consumers perceive AR products as tangible and
154 attractive (Verhagen et al., 2014).
155 Response: Attitudes towards apparel products and purchase intentions
156 AR online shopping experiences can result in positive attitudes toward products and
157 increased purchase intentions (Verhagen et al., 2014). Yim et al. (2017) found that AR-based
158 product presentations were superior to conventional web-based product presentations since they
159 offer higher immersion, media novelty, and media enjoyment, and increase attitude toward
160 medium and purchase intention. If a website utilizes AR, consumers become more curious about
Page 7 of 35 Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
161 the products, tend to patronize the website, and eventually purchase the products (Beck and Crié,
162 2018). Beck and Crié (2018) validated their findings by conducting tests with two products
163 (garments and eyeglasses) with both student and consumer samples. However, their study was an
164 online study and not set in a laboratory environment where participants could use the same
165 system and try-on the physical garment and accessory. To our knowledge, no studies empirically
166 compared consumers’ attitudes towards the product and purchase intentions between AR and
167 real-world conditions. Therefore, we proposed the following hypothesis:
168 H2: AR virtual try-on will have a comparable effect to physical try-on in terms of users’
169 (a) attitudes towards the apparel product and (b) purchase intentions.
170 A recent study by Kim et al. (2017) found that the use of AR is positively related to
171 enhanced telepresence, which in turn contributes to attitude toward the technology, and purchase
172 intention of products. In the study, researchers suggested that in comparison to virtual reality
173 (VR)-based presentations (i.e., wearing sunglasses on a 3D virtual model), AR-based
174 presentations (i.e., using a webcam to see themselves wearing sunglasses) were more likely to
175 stimulate presence, thus leading to stronger purchase intentions. Other researchers also supported
176 that telepresence increases attitudes towards products (Debbabi et al., 2013) and purchase
177 intentions (Song et al., 2007; Watson et al., 2018). When comparing a VR interface to 2D photos
178 and a video interface, Suh and Chang (2006) found that higher levels of telepresence (i.e.,
179 manipulated as VR in their study) increased positive attitudes toward the product, which was a
180 computer desk. However, they did not find any direct association of purchase intentions with
181 telepresence. Similar to previous studies which examined a variety of products from accessories
182 to make-up in both VR and AR settings (Park and Yoo, 2020; Yim et al., 2017; Watson et al.,
183 2018), telepresence in AR apparel virtual try-on may increase attitudes and purchase intentions
Page 8 of 35Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
184 toward the product. Thus, to understand the impact of telepresence on attitudes towards the
185 apparel product and purchase intentions, we proposed the hypothesis presented below:
186 H3: When using AR virtual try-on, individuals who have a higher telepresence will have
187 greater (a) attitudes towards the apparel product and (b) purchase intentions than those who have
188 a lower telepresence.
189 Methods
190 Data were collected with a one-factor (i.e., garment) within-subject quasi-experimental
191 study using repeated measures in two conditions: virtually trying-on condition using the AR
192 technology vs. physically trying-on condition. Within-subject design was selected because it
193 allowed researchers to remove subject-to-subject variation from the analysis of the relative
194 effects of different treatments (Seltman, 2015). It is important to “consider the context” when
195 deciding whether a between or within subject design should be selected (Charness, Gneezy and
196 Kuhn, 2012, p.2). Therefore, to be able to create conditions similar to the real-world, and not
197 conflict with the practice of online shopping (i.e., evaluating a garment on a website first and
198 ordering it for physical try-on), we let the participants try on the same dress in AR first and did
199 not reverse the order.
200 In this study, variance in participants’ body shapes and sizes is controlled with the within-
201 subject research design. Each participant was asked to try on the virtual dress using the AR
202 technology (Treatment 1); rate the perceived product performance, fit and size, attitudes and
203 purchase intentions towards the dress; and then physically try on the same dress (Treatment 2) in
204 a dressing room in the research lab after “ordering it online and receiving it via mail.” In a
205 similar vein, after viewing the product (a computer table) in an online store by using three
206 different viewing formats, Suh and Chang’s (2006) instructed their study participants to go to a
Page 9 of 35 Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
207 separate room to view the physical product and compare it to the online product. After Treatment
208 2 in our study, participants answered the same set of questions that they did after Treatment 1.
209 The two treatments in this study design were fundamentally different, as Treatment 1 required
210 participants to evaluate the garment solely based on the visual images presented by the AR
211 technology, whereas Treatment 2 allowed participants to evaluate the garment by seeing,
212 touching, and wearing it. More details of the experiment procedure are discussed in the following
213 sections.
214 Participants
215 Female college students age 18 and above were targeted as participants as they use the
216 Internet for apparel shopping, are technology-savvy, and adopt new product visualization
217 technologies easily (Yu et al., 2012). Compared to men, women examine garments in more detail
218 and tend to have more difficulty in selecting clothing items for themselves when shopping online
219 (Hansen and Jensen, 2008). After receiving approval from the Institutional Review Board (IRB),
220 undergraduate and graduate female participants were recruited via a large Midwestern
221 university’s mass-emailing service. A total of 87 participants from a variety of majors voluntarily
222 participated in the study. Participants’ mean age was 25.6 years old (SD = 6.08). The majority of
223 the participants were European-American (n = 65, 74.7%), followed by Asian/Asian-American
224 (n = 9, 10.3%), other (n = 6, 6.9%), Latino/Hispanic (n = 4, 4.6%), and African-American (n = 3,
225 3.4%). Most participants (89.7%) indicated that they had bought apparel online. To increase
226 participation, each participant was offered a $5 gift card as incentive.
227 Stimulus
228 In this study, a dress was used to develop the treatments. For Treatment 1, the dress
229 image was used in the AR technology for virtual try-on; for Treatment 2, the real dress was used
Page 10 of 35Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
230 for physical try-on. Based on their pilot study finding using a convenience sample of 52 female
231 undergraduate students, Kim and Damhorst (2010) suggested dresses as the most frequently
232 purchased garment type among students. Huang and Liu (2014) also found that dresses were
233 among the top-three garments to be tried-on in AR; women spent the longest time on dresses
234 when using AR for virtual try-ons. Considering computer monitor limitations, full-length dresses
235 and pants are usually not easily seen with an online AR application. Also, full-length sleeves add
236 an extra variable to control the believability of the AR simulation.
237 In order to select the stimulus, six dress images of various knee-length, short-
238 sleeved/sleeveless dress styles were evaluated by five women in a pre-test. The pre-test
239 examined the garment style’s attractiveness, fashionability, and likeability, using a 7-point
240 Likert-type scale adapted from Park (2009). The average ratings of the six dresses were as
241 follows: 4.47 (short-sleeved fitted dress), 3.53 (sleeveless fitted dress), 5.33 (short-sleeved
242 shift/A-line dress), 3.60 (sleeveless shift/A-line dress), 5.20 (sleeveless fit-and-flare dress), and
243 4.13 (short-sleeved fit-and-flare dress). Following suggestions from Kim and Lennon (2008) and
244 Park (2009), the short-sleeved fit-and-flare dress with a neutral rating was chosen to limit the
245 garment style’s potential effect on the variables. The dress was purchased from a mass-retailer in
246 sizes from XS (0-2) to XL (16-18). To eliminate the confounding effect of brand name, brand
247 labels were removed from the dresses.
248 To create the dress image for Treatment 1, a photo of the dress in size medium was taken
249 on an appropriate dress form. The dress form and background components were then erased in
250 Adobe Photoshop. The final image was uploaded in PNG format to the AR developer’s (Zugara)
251 server (Figure 1). In the online AR application, the computer’s webcam captured participants’
252 body images, and the front-view of the stimulus was displayed on their body in 2D. Participants
Page 11 of 35 Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
253 were able to adjust the size and placement of the AR dress by moving their hands in the air and
254 “clicking” the control buttons shown on the screen without needing a mouse or a keyboard. This
255 way, the computer screen became an interactive mirror without necessitating a high-tech kiosk
256 (Figure 1).
257 Insert Figure 1 here
258 Since Treatment 2 was the real dress for physical try-on and should remain exactly the
259 same as the Treatment 1 dress to avoid any variations other than the AR versus physical try-on
260 conditions, the researchers intentionally made no adjustment to the dress. The researchers
261 nonetheless made sure that all sizes of the dress were available in the lab for Treatment 2.
262 Experimental procedures
263 Participants were invited to our research laboratory and received instructions about the
264 task and the procedures. Informed consent forms were filled out at the beginning. Participants
265 were instructed to use an iMac on which the AR application website specifically developed for
266 this study was available. To virtually try on the AR dress (Treatment 1), participants stood 4-5
267 steps in front of the computer screen with a built-in video camera, so they could see their bodies
268 at least from head to calf. After the virtual try-on experience, participants were asked to complete
269 an online questionnaire on a separate laptop. The questionnaire measured fit and size perceptions
270 of the dress, product performance perceptions, telepresence, attitudes towards the dress, and
271 purchase intentions. The respondents also indicated the size of the dress that would fit them.
272 Next, participants physically tried on the real dress in the sizes they indicated previously
273 (Treatment 2), then answered a second online questionnaire regarding the real dress with the
274 same measurement instruments, except telepresence, as the previous questionnaire. We
275 implemented “time-off” in two ways to minimize potential confounding effects of the within-
Page 12 of 35Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
276 subjects design: (1) after the first questionnaire, participants went to the changing room to try on
277 the real dress and (2) we added three open-ended questions about participant reflections on AR
278 (not included to the present study) between the questionnaires to make it hard to remember
279 repeated questions between Treatment 1 and Treatment 2.
280 Survey instruments and data analysis procedures
281 Items used to assess dress fit involving thirteen areas (e.g., neck, busk, waist) except
282 buttocks were adapted from the fit scales developed by Song and Ashdown (2012) using a 5-
283 point scale that was anchored at too loose/long/wide (1), excellent fit (3), and too
284 tight/short/narrow (5). Product performance in regards to both treatments was measured with the
285 product performance risk scale adapted from Yu et al. (2012) using a 7-point Likert-type scale,
286 which was anchored at not sure at all (1) and very sure (7). A question, “How sure are you about
287 the apparel product’s attributes to perform satisfactorily to your needs?”, was asked to measure
288 three dimensions (visual, tactile, and trial) at ten sub-dimensions (visual: style, fabric, color,
289 details, coordination with other items; tactile: touch and feel, weight of garment; and trial: fit,
290 comfort, and appearance on body) (Yu et al., 2012). Telepresence was measured using the scale
291 adapted from Song et al.’s (2007) five-item, 7-point Likert-type scale, which was anchored at
292 strongly disagree (1) and strongly agree (7). The items asked if the application “…lets me easily
293 visualize what the real dress is like,” “…gives me as much sensory information about the dresses
294 as I would experience in a store,” “…creates a product experience similar to the one I would
295 have when shopping in a store,” “…allows me to interact with the dresses as I would in the
296 store,” and “…provides accurate sensory information about the dresses.”
297 Attitudes toward the AR and real dresses were evaluated using six items with a 7-point
298 Likert-type scale from Holbrook and Batra (1987) and Bruner (1998) for the following question:
Page 13 of 35 Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
299 “Please tell us about your overall thoughts and feelings about the dress: The dress is...” The
300 items were anchored at very dislikable (1)/very likeable (7), very unattractive (1)/very attractive
301 (7), very bad (1)/very good (7), very unfavorable (1)/very favorable (7), very unpleasant (1)/very
302 pleasant (7), very unappealing (1)/very appealing (7). Purchase intention was measured by a
303 scale originated from MacKenzie et al. (1986) and used by Yu et al. (2012): very improbable
304 (1)/very probable (7); very unlikely (1)/very likely (7); and very impossible (1)/very possible (7).
305 The instrument was pilot tested with five students. Some phrases in the instructions and the
306 questionnaire were further edited for clarity. Cronbach’s alphas of all scales were greater than
307 0.8. Table 1 shows the constructs, their items, and the reliability scores. To examine hypotheses,
308 paired-sample t-tests and multivariate analysis of variance (MANOVA) were conducted in SPSS
309 26.
310 Insert Table 1 about here
311 Results
312 Results for H1
313 Finding the right size. No statistical difference was found between the virtually tried-on AR
314 dress sizes (M = 2.69; SD = 1.16) and physically tried-on dress sizes (M = 2.65; SD = 1.23)
315 (t(86) = -1.75; p = .41). Therefore Hypothesis 1a was supported. Table 2 shows the distribution
316 of best-fitting dress sizes for AR (Treatment 1) and real (Treatment 2) dresses.
317 Insert Table 2 about here
318 Evaluating Fit. Before analyzing the overall fit of the dress using the product performance
319 construct, fit was evaluated in more detail by looking at thirteen locations on both AR and real
320 dresses. In general, all areas evaluated were perceived to be close to 3 (excellent fit). As shown
321 in Table 3, there were no significant differences in fit at four areas (neck, shoulder width, sleeve
Page 14 of 35Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
322 opening, and volume/fullness in skirt) between AR and real dresses. However, results showed
323 significant differences between these two conditions in nine areas out of thirteen. Participants
324 perceived looser fit around bust (ΔM=.28, SD= 1.00, t(84) = 2.58, p < .05) and wider shoulder
325 (ΔM=.22, SD= .93, t(86) = 2.18, p < .05) when virtually trying on the AR dress. However, areas
326 such as waist (ΔM = -.39, SD = .97, t(86) =-3.76, p < .001), abdomen (ΔM=-.26, SD=.88, t(83)
327 =-2.37, p < .01), and hip (ΔM=-.43, SD=.83, t(85) = -4.79, p < .001) were perceived tighter when
328 using AR. When using AR, lengths were perceived to be longer at the following areas: sleeve
329 (ΔM= .29, SD= .85, t(85) = 3.16, p < .01), torso (ΔM= .26, SD= .92, t(86) = 2.68, p < .01), skirt
330 (ΔM=.37, SD=.94, t(86) =-3.64, p < .001), and overall dress length (ΔM=.42, SD=.76, t(84)
331 =5.13, p < .001). Therefore Hypothesis 1b was partially supported.
332 Insert Table 3 about here
333 Evaluating product performance. Out of ten attributes that investigated participants’ perceived
334 AR dress performance and real dress performance, nine items were found to be significantly
335 different. Average ratings for the attributes related to tactile properties (i.e., touch and feel:
336 M=2.55, weight: M=2.81) were lower than neutral (4) for the AR dress (see Table 1). When
337 wearing the real dress, participants thought that nine dress attributes performed significantly
338 better than when using AR: style (ΔM = .60; SD =1.69; t(86) = 3.30; p < .01), fabric (ΔM = 2.01;
339 SD = 2.15; t(85) = 8.68; p < .001), coordination with other items (ΔM = .48; SD = 1.45; t(85) =
340 3.04; p < .001), details (ΔM = .96; SD = 1.64; t(86) = 5.50; p < .001), touch and feel (ΔM = 3.34;
341 SD = 2.17; t(86) = 14.32; p < .001), weight (ΔM = 3.21; SD = 2.05; t(84) = 14.42; p < .001), fit
342 (ΔM = 1.16; SD = 1.85; t(85) = 5.82; p < .001), comfort (ΔM = 2.53; SD = 2.17; t(84) = 10.75; p
343 < .001), and appearance on the body (ΔM = .46; SD = 2.09; t(85) = 2.06; p < .05). Color (ΔM
Page 15 of 35 Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
344 =.41; SD = 1.93; t(85) = 1.12; p > .05) was not significantly different between the two conditions
345 (Table 4). Therefore Hypothesis 1c was partially supported.
346 Insert Table 4 about here
347 Results for H2
348 A paired-sample t-test was conducted to test Hypotheses 2a (H2a) and 2b (H2b).
349 Attitudes towards both AR and real dress were favorable and above 5. Purchase intentions were
350 moderately positive in both conditions as well (Table 1). Participants had significantly more
351 favorable attitudes towards the real dress (M = 5.60, SD = 1.14) than the AR dress (M = 5.25,
352 SD = 1.00) (∆M = .35, SD = 1.15, t(85) = 2.62, p < .05). Participants indicated greater purchase
353 intentions during physical try-on (M = 4.74, SD = 1.71) as compared to AR try-on (M = 4.27, SD
354 = 1.70) (∆M = .47, SD = 1.94, t(85)=2.32, p < .05). Thus, H2a and H2b were not supported.
355 Results for H3
356 In order to examine Hypothesis 3, participants were split into two groups based on the
357 mean value of telepresence (M=3.70). Participants who indicated telepresence 3.70 and higher
358 on average were assigned to the high telepresence group (n=39), whereas participants who had
359 telepresence level lower than 3.70 were placed in the low telepresence group (n=48). Results of
360 MANOVA showed that the high telepresence group tended to have more positive attitudes (M
361 Low Telepresence= 5.04 (SD = .14), M High Telepresence=5.50 (SD =.16)) and greater purchase intentions
362 to the apparel product when using AR (M Low Telepresence= 3.88 (SD = .24), M High Telepresence=4.77
363 (SD =.27)) than the low telepresence group [F (1, 83) = 3.15; p <.05, partial η2= .071]. Thus,
364 Hypothesis 3a and 3b were supported.
Page 16 of 35Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
365 Discussion
366 In this present study, we investigated if AR can help online apparel shoppers order the
367 right size, obtain clues about fit and product performance by judging visual clues, and determine
368 if virtually trying on an AR garment is the same as physically trying it on in regard to attitudes
369 towards the garment and purchase intentions. Additionally, we examined how high and low
370 telepresence levels of the participants in the AR condition impact their attitudes and purchase
371 intentions. Understanding how online AR technologies affect consumers’ perceptions of
372 garments can help brands develop new ways to reduce consumers’ regret caused by post-
373 purchase expectation disconfirmation. In our study, the majority of the participants were able to
374 select their sizes correctly and did not need to try on a different size once they “received” the
375 garment “via mail” after “ordering it online.” Narvar’s (2017) survey shows that online apparel
376 shoppers make bracket purchases, which means buying multiple versions of an item (size-, color-
377 and style-wise) to see which they prefer, with the intention of returning the rest. In this regard,
378 using AR would be very beneficial for retailers to implement as it gives shoppers more
379 confidence in the sizes they want to try at home and reduces bracket purchases that increase re-
380 shelving and shipping costs. Implications also include using this technology in a physical store
381 environment to help shoppers quickly sort through styles to find what they like.
382 In regard to perceiving fit, our study found that participants were able to approximate
383 how garment parts would fit (loose, tight, or right) when using an AR application. Therefore, our
384 results imply that AR virtual try-on can give shoppers visual clues on the garment fit. However,
385 the type of AR technology (3D overlay vs. 2D overlay), interactivity speed, and quality of the
386 AR images impacts the level of visual information (realistic vs. graphic) that shoppers receive
387 (Yim et al., 2017). In our study, when compared to the real dress, fit of the AR dress was
Page 17 of 35 Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
388 perceived to be significantly looser at the bust and tighter at the waist, abdomen, and hip. The
389 discrepancies found around the waist, abdomen, and hip are considered plausible because the AR
390 dress was a superimposed, static 2D image on the body, not stretching at these areas. Our finding
391 on the fit perception at bust was unexpected. This result may have arisen from a possible
392 difference between the bust measurements of the dress form, which was used to create the AR
393 dress stimuli, and study participants. Moreover, lengths of sleeve, torso, skirt, and overall dress
394 were perceived to be significantly longer in AR. Holding a garment up to oneself and assessing
395 its length may be different than wearing it. After wearing the garment, the third dimension
396 (depth) adjusts the garment length on the body and the garment becomes shorter than its flat
397 form against the body.
398 Although in our study only one type of stimulus (dress) was used, the findings may
399 inform improvement of AR applications to help consumers evaluate fit. For example, garment
400 pictures may be improved by taking several pictures of the same product depending on different
401 body types or sizes (petite, regular, tall, and plus) to reduce the discrepancy. The images can be
402 adapted to each user’s body shape, to the extent that the materials’ elasticity allows. While
403 retailers do not have much control over how AR technology improves, the findings can help
404 inform consumers of possible discrepancies, allowing for more accurate decisions regarding fit.
405 Almost all of the product-performance-related items, except color, were perceived
406 significantly different in two conditions. Attributes such as style, fabric, coordination with other
407 items, details, touch and feel, weight, overall fit, comfort, and appearance on body were
408 perceived to perform better for the real dress. The average ratings for tactile attributes (e.g.,
409 touch and feel, comfort, and weight) were closer to the unsatisfactory side of the scale when
410 using AR. In AR, users cannot account for tactile attributes such as touch and feel, comfort and
Page 18 of 35Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
411 weight. Nonetheless, results related to specific visual characteristics (i.e., style, detail, and
412 coordination with other items), were above the neutral level when using AR, showing that AR
413 visuals were satisfactory to help participants understand these attributes when shopping online.
414 The findings showed that physical try-on condition affected consumers to have higher
415 attitudes and purchase intentions compared to AR try-on condition. An explanation could be that
416 consumers still prefer and make decisions based on the actual tactile experience they gained from
417 physical try-on. However, it does not mean that AR is useless to consumers. As the findings
418 from H1 indicated, AR does provide good visual information that can increase consumers’
419 attitudes and purchase intentions. Furthermore, participants with higher telepresence level were
420 found to have more positive attitudes towards the dress and greater purchase intentions when
421 using AR as compared to the participants with low telepresence level. This finding adds to the
422 existing literature (Debbabi et al. 2013; Kim et al, 2017; Suh and Chang, 2006) of how varying
423 levels of telepresence affect attitudes towards the product and purchase intentions from the
424 apparel field’s point-of–view.
425 Conclusion
426 In the present study, the S-O-R model was used as a theoretical framework to investigate
427 how AR products, most specifically a dress, and AR try-on would be perceived by consumers in
428 comparison to physical interaction with the dress. For this purpose, we used an online shopping
429 scenario that allowed our participants to experience the AR product, “order it” to see the physical
430 dress, and decide if they want to keep it after physical try-on (i.e., purchase intention of the real
431 dress). In addition to contributing to the academic field of AR product presentation in online
432 shopping, our findings offer several implications for research and society/practice. Theoretical
433 contributions of this research imply that although physical try-on plays an important role in
Page 19 of 35 Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
434 apparel purchase decisions, AR stimulus can provide information that helps consumers make
435 accurate decisions particularly regarding apparel sizes and visual characteristics when shopping
436 online. Therefore AR can supplement, rather than replace, the physical try-on experience.
437 Attitudes towards the dress and intentions to purchase the dress when using the AR technology
438 were above the mid-point, and close to the attitudes and intentions measured in the physical try-
439 on condition. Additionally, participants with higher telepresence levels were likely to have
440 higher attitudes and purchase intentions as compared to the participants with low telepresence.
441 These findings suggest that AR can be instrumental in introducing a certain style, building
442 positive attitudes towards products, and driving sales when the consumers perceive a certain
443 level of telepresence.
444 Although adopting AR to provide more information about the product on an e-commerce
445 environment would be an expensive investment (Plotkina and Saurel, 2019), our findings imply
446 that retailers can benefit from using AR technology to increase consumer interest in their
447 products. Retailers need to understand the potentials of AR technologies and work with
448 technology developers to push the limits to enhance shopper experiences. As suggested by
449 Pantano et al. (2017), fashion retailers who want to implement AR systems should be aware of
450 the recent progresses as well as drawbacks in technology, taking part in the innovation process
451 rather than passively adopting the offered technology. AR technology is an untapped area in
452 apparel, and its potential in conveying reliable information when shopping online needs to be
453 examined more closely.
454 Some limitations of the present study must be addressed. First, the use of a student
455 population reduces the generalizability of the study findings. Additionally, the vast majority of
456 these women were in the XS-M category with very few women in the larger sizes. The product
Page 20 of 35Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
457 chosen for this experiment is another limitation. As Kim and Lennon (2008) indicated, when a
458 single stimulus is used, the results cannot be generalized to other stimuli. Different garment types
459 in varying silhouettes and fit must be considered for further study. Future studies should apply
460 our methodology to other product categories and compare the results. Technology accounts for a
461 third limitation. The AR garments were 2D and did not wrap around the body. Lu and Smith
462 (2007) mentioned that AR system rendering should be improved to merge digital and real
463 environments in a realistic way. This would improve vividness, i.e., “the ability of a technology
464 to produce a sensorially rich mediated environment” (Steuer 1992, p. 80). As Suh and Chang
465 (2006) suggested, focus should be on improving IIT interfaces to generate higher telepresence
466 levels, so that consumers’ perceptions of products in online stores can be improved. Plotkina and
467 Saurel (2019) argue that AR-based tools for trying on garments virtually are not “sufficiently
468 technologically advanced” yet. Their study compared a mobile application with AR try-on to a
469 mobile commerce interface that presented fashion models similar to the consumers. Plotkina and
470 Saurel (2019) found that their female participants preferred traditional pictures. The present
471 study used the AR provider’s server; therefore, fiber, fabric information, and price were not
472 included on the website. Participants did not get clues on whether the fabric would stretch when
473 wearing the dress, or if the dress was affordable. Written explanations would encompass the
474 limitations and overcome picture-related misconceptions. As Kim and Lennon (2008) suggested,
475 detailed verbal descriptions are important to enhance consumer understanding of the product and
476 positively influence their purchase decisions. Additionally, collecting information from males
477 would be a good idea, as they have different preferences and were reported to be less confident
478 when selecting clothes without advice from a knowledgeable person (Hansen and Jensen, 2009;
479 O’Cass, 2004). The present study examined the influence of AR, which was developed by a
Page 21 of 35 Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
480 specific technology provider, on telepresence based on the S-O-R model. Future studies should
481 consider other providers as well as advancing AR functions, and using additional theoretical
482 models to compare the effects of different AR try-on conditions on telepresence.
483 Another limitation was the experimental design. Because a within-subject experimental
484 design was selected, with a possible carry-over effect, it is possible that attitudes and purchase
485 intentions were higher in the physical try-on condition due to the participants’ learning of the
486 product, which might have been reinforced by experiencing the same product twice, first
487 virtually and second physically. Future research should look at conducting a between-subject
488 experiential design. Future studies should also investigate factors such as visual imagery on AR
489 fitting experience to better differentiate its competitive advantage as compared to virtual try-on
490 based on parametric models. Finally, future researchers should examine the perceived value of
491 AR fitting systems and their influence on consumer experience. Although AR fitting has
492 limitations on providing accurate fit information, based on our study, the unique interactive
493 features may contribute to consumers’ perceived value of the shopping experience.
494 References
495 Avametric. (2018), available at: https://www.avametric.com/
496 Azuma, R. (1997), “A survey of augmented reality”, Presence-Teleoperators and Virtual
497 Environments, Vol. 6, No. 4, pp. 355-85.
498 Beck, M., and Crié, D. (2018), “I virtually try it … I want it! Virtual fitting room: a tool to
499 increase on-line and off-line exploratory behavior, patronage and purchase intentions”,
500 Journal of Retailing and Consumer Services. Vol. 40, pp. 279-86.
Page 22 of 35Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
501 Boyle, A. (2018), “Amazon’s blended-reality mirror shows you wearing virtual clothes in virtual
502 locales”, available at: https://www.geekwire.com/2018/amazon-patents-blended-reality-
503 mirror-shows-wearing-virtual-clothes-virtual-locales/ (accessed 15 March 2019).
504 Bruner, G. C. (1998), “Standardization and justification: do ad scales measure up?”, Journal of
505 Current Issues and Research in Advertising, Vol. 20 No. 1, pp. 1-18.
506 Bulearca, M., and Tamarjan, D. (2010), “Augmented reality: a sustainable marketing tool?”,
507 Global Business and Management Research: An International Journal, Vol. 2 No. 2/3,
508 pp. 237-52.
509 Carmigniani, J., Furht, B., Anisetti, M., Ceravolo, P., Damiani, E., and Ivkovic, M. (2011),
510 “Augmented reality technologies, systems and applications”, Multimedia Tools and
511 Applications, Vol. 51 No. 1, pp. 341-77.
512 Cass, M. (2017), “Streetwear brands are tapping the creative power of AR”, available at:
513 https://www.jwtintelligence.com/2017/08/augmented-fashion/ (accessed 15 March 2019).
514 Chang, H., Li, Y., Chen, H., Feng, S., and Chien, T. (2013), “A dynamic fitting room based on
515 Microsoft Kinect and augmented reality technologies”, In Kurosu, M. (Ed.), Human-
516 Computer Interaction. Interaction Modalities and Techniques, Vol. 8007, pp. 177-85.
517 Charness, G., Gneezy, U., and Kuhn, M. (2012). “Experimental methods: Between-subject and
518 within-subject design”, Journal of Economic Behavior and Organization, Vol. 81 No.1,
519 pp. 1-8.
520 Chitrakorn, K. (2018), “5 technologies transforming retail in 2018”, available at:
521 https://www.businessoffashion.com/articles/fashion-tech/5-technologies-transforming-
522 retail (accessed 15 March 2019).
Page 23 of 35 Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
523 Debbabi, S., Baile, S., Des Garets, V., and Roehrich, G. (2013), “The impact of telepresence in
524 an online ad on forming attitudes towards the product: the relevance of the traditional
525 experiential approach”, Recherche Et Applications En Marketing, Vol. 28 No. 2, pp. 3-
526 24.
527 Edvardsson, B., Enquist, B., and Johnston, R. (2005), “Cocreating customer value through
528 hyperreality in the prepurchase service experience”, Journal of Service Research, Vol. 8
529 No. 2, pp. 149-61.
530 Eroglu, S. A., Machleit, K. A., and Davis, L. M. (2001). “Atmospheric qualities of online
531 retailing: A conceptual model and implications.” Journal of Business Research, Vol. 54
532 No 2, pp. 177–184.
533 Eroglu, S., Machleit, K., and Davis, L. (2003). “Empirical testing of a model of online store
534 atmospherics and shopper responses.” Psychology & Marketing, Vol. 20, No. 2, pp. 139–
535 150.
536 Fiore, A. M., and Jin, H. (2003), “Influence of image interactivity on approach responses
537 towards an online retailer”, Internet Research, Vol. 13 No. 1, pp. 38-48.
538 Fiore, A., and Kim, J. (2007). “An integrative framework capturing experiential and utilitarian
539 shopping experience.” International Journal of Retail & Distribution Management, Vol.
540 35 No. 6, pp. 421-442.
541 Fiore, A. M., Kim, J., and Lee, H. (2005), “Effect of image interactivity technology on consumer
542 responses toward the online retailer”, Journal of Interactive Marketing, Vol. 19 No. 3,
543 pp. 38-53.
544 FittingBox. (n.d.), available at: http://www.fittingbox.com/
Page 24 of 35Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
545 Hansen, T., and Jensen, J. (2009), “Shopping orientation and online clothing purchases: the role
546 of gender and purchase situation”, European Journal of Marketing, Vol. 43 No. 9/10, pp.
547 1154-70.
548 Holbrook, M. B., and Batra, R. (1987), “Assessing the role of emotions as mediators of
549 consumer responses to advertising”, Journal of Consumer Research, Vol. 14, pp. 404–20.
550 Huang, S., Yang, Y., and Chu, C. (2011), “Human-centric design personalization of 3D glasses
551 frame in markerless augmented reality”, Advanced Engineering Informatics, Vol. 26 No.
552 1, pp. 35-45.
553 Huang,T., and Liu, F. H. (2014), “Formation of augmented-reality interactive technology's
554 persuasive effects from the perspective of experiential value”, Internet Research, Vol. 24
555 No. 1, pp. 82-109.
556 Javornik, A. (2016), “Augmented reality: research agenda for studying the impact of its media
557 characteristics on consumer behavior”, Journal of Retailing and Consumer Services, Vol.
558 30, pp. 252-261.
559 Kang, J. (2014), “Augmented reality and motion capture apparel e-shopping values and usage
560 intention”, International Journal of Clothing Science and Technology, Vol. 26 No. 6, pp.
561 486-99.
562 Kim, H., and Damhorst, M. (2010), “The relationship of body-related self-discrepancy to body
563 dissatisfaction, apparel involvement, concerns with fit and size of garments, and purchase
564 intentions in online apparel shopping”, Clothing and Textiles Research Journal, Vol. 28
565 No. 4, pp. 239-54.
Page 25 of 35 Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
566 Kim, H. C., and Hyun, M. Y. (2016), “Predicting the use of smartphone-based Augmented
567 Reality (AR): does telepresence really help?” Computers in Human Behavior, Vol. 59,
568 pp. 28-38.
569 Kim, M., and Lennon, S. (2008), “The effects of visual and verbal information on attitudes and
570 purchase intentions in Internet shopping”, Psychology & Marketing, Vol. 25 No. 2, pp.
571 146-78.
572 Kim, S., Baek, T. H., and Kim, S. H. (2017), “The effect of presence on consumers' responses to
573 virtual mirror technology”, Paper presented at the 2017 International Textile and Apparel
574 Association Conference, 14-18 November 2017, St. Petersburg, FL.
575 Lee, K.-Y. (2012), “Consumer processing of virtual experience in e-commerce: a test of an
576 integrated framework”, Computers in Human Behavior, Vol. 28 No. 6, pp. 2134-42.
577 Li, H., Daugherty, T., and Biocca, F. (2001), “Characteristics of virtual experience in electronic
578 commerce: a protocol analysis”, Journal of Interactive Marketing, Vol. 15 No. 3, pp. 13-
579 30.
580 Li, H., Daugherty, T., and Biocca, F. (2003), “The role of virtual experience in consumer
581 learning”, Journal of Consumer Psychology, Vol. 13 No. 4, pp. 395-407.
582 Lim, J. and Ayyagari, R. (2018). “Investigating the determinants of telepresence in the e-
583 commerce setting.” Computers in Human Behavior. Vol. 8, pp. 360-371.
584 Lu, Y., and Smith, S. (2007), “Augmented reality e-commerce assistant system: trying while
585 shopping”, In Jacko, J. (Ed.), Human-Computer Interaction. Interaction Platforms and
586 Techniques, Vol. 4551, pp. 643–52.
Page 26 of 35Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
587 MacKenzie, S. B. Lutz, R. J., and Belch, G. E. (1986), “The role of attitude toward the ad as a
588 mediator of advertising effectiveness: a test of competing explanations”, Journal of
589 Marketing Research, Vol. 23 No. 5, pp. 130–43.
590 Mehrabian, A.and Russell, J. (1974). An approach to environmental psychology. Cambridge:
591 M.I.T. Press.
592 Magic Mirror. (n.d.), available at: http://www.magicmirror.me/
593 Memory Mirror. (n.d.), available at: http://memorymirror.com/
594 Merle, A., Senecal, S., and St-Onge, A. (2012), “Whether and how virtual try-on influences
595 consumer responses to an apparel web site”, International Journal of Electronic
596 Commerce, Vol. 16 No. 3, pp. 41–64.
597 Metail. (n.d.), available at: http://www.metail.com/
598 Mollen, A., and Wilson, H. (2010), “Engagement, telepresence and interactivity in online
599 consumer experience: reconciling scholastic and managerial perspectives”, Journal of
600 Business Research, Vol. 63 No. 9-10, pp. 919-925.
601 Narvar. (2017), “Making Returns a Competitive Advantage”, available at:
602 http://see.narvar.com/rs/249-TEC-
603 877/images/Narvar_Consumer_Survey_Returns_June2017.pdf (accessed 15 March
604 2019).
605 Overmars, S., and Poels, K. (2015), “Online product experiences: the effect of simulating
606 stroking gestures on product understanding and the critical role of user control”,
607 Computers in Human Behavior, Vol. 51 No. A, pp. 272–84.
608 O'Cass, A. (2004), “Fashion clothing consumption: antecedents and consequences of fashion
609 clothing involvement”, European Journal of Marketing, Vol. 38 No. 7, pp. 869–82.
Page 27 of 35 Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
610 Pachoulakis, I., and Kapetanakis, K. (2012), “Augmented reality platforms for virtual fitting
611 rooms”, International Journal of Multimedia & Its Applications, Vol. 44, pp. 35-46.
612 Pantano, E., Rese, A., and Baier, D. (2017), “Enhancing the online decision-making process by
613 using augmented reality: a two-country comparison of youth markets”, Journal of
614 Retailing and Consumer Services, Vol. 38, pp. 81-95.
615 Park, M. J. (2009), “The roles of information load and information quality in online apparel
616 shopping”, Journal of the Korean Home Economics Association, Vol. 47 No. 9, pp. 101–
617 10.
618 Park, M. and Yoo, J. (2020). “Effects of perceived interactivity of augmented reality on
619 consumer responses: A mental imagery perspective”, Journal of Retailing and Consumer
620 Services, Vol. 52. Available from: doi: 10.1016/j.jretconser.2019.101912 [Accessed 15th
621 August 2019].
622 Pine, B. J., and Korn, K. (2011). Infinite Possibility: Creating Customer Value on the Digital
623 Frontier, Berrett-Koehler Publishers, San Francisco, CA.
624 Plotkina, D. and Saurel, H. (2019). “Me or just like me? The role of virtual try-on and physical
625 appearance in apparel M-retailing.” Journal of Retailing and Consumer Services, Vol. 51,
626 pp. 362-377.
627 Prashar, S., Sai Vijay, T., and Parsad, C. (2017).”Effects of online shopping values and website
628 cues on purchase behaviour: A study using S-O-R framework.” Vikalpa, Vol. 42 No.1,
629 pp. 1-18.
630 Rese, A., Baier, D., Geyer-Schulz, A., and Schreiber, S. (2017), “How augmented reality apps
631 are accepted by consumers: a comparative analysis using scales and
632 opinions”, Technological Forecasting & Social Change, Vol. 124, pp. 306-19.
Page 28 of 35Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
633 Steuer, J. (1992), “Defining virtual reality: dimensions determining telepresence”, Journal of
634 Communication, Vol. 42 No. 4, pp. 73-93.
635 Schlosser, A. (2003), “Experiencing products in the virtual world: the role of goal and imagery in
636 influencing attitudes versus purchase intentions”, Journal of Consumer Research, Vol. 30
637 No. 2, pp. 184-98.
638 Schwartz, A. M. (2011), “Augmenting purchase intent: an empirical study on the effects of
639 utilizing augmented reality in online shopping”, Unpublished Master’s thesis, University
640 of California, Riverside, CA, available at:
641 http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1858976
642 Seltman, H. J. (2015), “Within-Subjects Designs”, In Experimental Design and Analysis,
643 available at http://www.stat.cmu.edu/~hseltman/309/Book/chapter14.pdf
644 Shih, C.-F. (1998) "Conceptualizing consumer experiences in cyberspace", European Journal of
645 Marketing, Vol. 32 No. 7/8, pp. 655-663.
646 Song, H., and Ashdown, S. (2010), “An exploratory study of the validity of visual fit assessment
647 from three-dimensional scans”, Clothing and Textiles Research Journal, Vol. 28 No. 4,
648 pp. 263-78.
649 Song, K., Fiore, A. M., and Park, J. (2007). “Telepresence and fantasy in online apparel
650 shopping experience”, Journal of Fashion Marketing and Management: An International
651 Journal, Vol. 11 No. 4, pp. 553-570.
652 Suh, K., and Chang, S. (2006). “User interfaces and consumer perceptions of online stores: The
653 role of telepresence.” Behaviour & Information Technology, Vol. 25 No. 2, pp. 99-113.
654 triMirror. (n.d.), available at: http://www.trimirror.com/en/about/
Page 29 of 35 Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
655 Yim, MYC., Chu, SC. and Sauer, P.L. (2017), “Is augmented reality technology an effective tool
656 for e-commerce? An interactivity and vividness perspective”, Journal of Interactive
657 Marketing, Vol. 39, pp. 89-103.
658 Yu, U., Lee, H., and Damhorst, M. (2012), “Exploring multidimensions of product performance
659 risk in the online apparel shopping context: visual, tactile, and trial risks”, Clothing and
660 Textiles Research Journal, Vol. 30 No. 4, pp. 251-66.
661 Verhagen, T., Vonkeman, C., Feldberg, F. and Verhagen, P. (2014), “Present it like it is here:
662 creating local presence to improve online product experiences”, Computers in Human
663 Behavior, Vol. 39, pp. 270-80.
664 Watson, A., Alexander, B., and Salavati, L. (2018). “The impact of experiential augmented
665 reality applications on fashion purchase intention.” International Journal of Retail &
666 Distribution Management, Available from: doi:10.1108/IJRDM-06-2017-0117 [Accessed
667 15th August 2019].
668 Witmer, B., and Singer, M. (1998), “Measuring presence in virtual environments: a presence
669 questionnaire”, Presence: Teleoperators and Virtual Environments, Vol. 7 No. 3, pp.
670 225-40.
671 Wu, J., Won J., Hae, K.J., Damminga, C., Kim, H-Y., and Johnson, K. (2013). “Fashion product
672 display: An experiment with Mockshop investigating colour, visual texture, and style
673 coordination.” International Journal of Retail & Distribution Management, Vol. 41 No.
674 10, pp. 765-789.
675 Workman, J. (2010), “Fashion consumer groups, gender, and need for touch”, Clothing and
676 Textiles Research Journal, Vol. 28 No. 2, pp. 126-39.
677 Zugara. (n.d.), available at: http://www.zugara.com
Page 30 of 35Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
Figure 1. Representation of the AR interface showing the controls and the dress stimulus
Page 31 of 35 Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
Table 1.
The Descriptive Statistics of Survey Instruments
AR dress
Real dress
Constructs
n
M (SD)
α
n
M (SD)
α
Product performance
Style
87
5.10 (1.39)
N/A
87
5.70 (1.37)
N/A
Fabric
86
3.73 (1.90)
N/A
86
5.74 (1.24)
N/A
Color
86
4.79 (1.50)
N/A
86
5.20 (1.59)
N/A
Coordination with other items
86
5.22 (1.50)
N/A
86
5.70 (1.23)
N/A
Details
87
4.37 (1.41)
N/A
87
5.33 (1.38)
N/A
Touch and feel
87
2.55 (1.84)
N/A
87
5.89 (1.09)
N/A
Weight of garment
85
2.81 (1.91)
N/A
85
6.02 (.99)
N/A
Overall Fit
86
4.20 (1.71)
N/A
86
5.36 (1.59)
N/A
Comfort
85
3.38 (2.01)
N/A
85
5.91 (1.14)
N/A
Appearance on the body
86
4.63 (1.66)
N/A
86
5.09 (1.71)
N/A
Telepresence
86
3.70 (1.25)
.88
N/A
Attitude towards the dress
86
5.25 (1.00)
.94
86
5.60 (1.14)
.96
Purchase Intention
86
4.27 (1.70)
.96
86
4.74 (1.71)
.96
Notes. M = mean, SD = standard deviation, α = Cronbach’s alpha
7-point Likert-type scales were used to measure the constructs/ items
Page 32 of 35Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
Table 2.
Distribution of AR (virtual try-on) and Real (physical try-on) Dress Sizes
Which dress size fit you best?
XS
S
M
L
XL
XXL
XS
11
0
0
0
0
0
S
2
27
0
0
0
0
M
0
4
24
4
0
0
L
0
0
2
5
0
0
XL
0
0
0
0
3
1
What size do you
think you should wear
for this dress?
XXL
0
0
0
0
0
3
Page 33 of 35 Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
Table 3.
Fit Comparisons at Thirteen Areas
Fit location
Real
dress fit
M (SD)
AR
dress fit
M (SD)
ΔM (SD)
df
t
AR dress
fits…than
real dress
Neck
2.44 (.54)
2.48 (.65)
-.04 (.60)
85
-.54
-
Bust
2.76 (.68)
2.48 (.85)
.28 (1.00)
84
2.58*
Looser
Waist
3.02 (.65)
3.41 (1.00)
-.39 (.97)
86
-3.76***
Tighter
Abdomen
3.14 (.60)
3.40 (.78)
-.26 (.88)
83
-2.73**
Tighter
Hip
2.98 (.34)
3.41 (.77)
-.43 (.83)
85
-4.79***
Tighter
Armhole
3.34 (.61)
3.11 (.89)
.23 (1.05)
83
1.86
-
Shoulder width
2.95 (.61)
2.73 (.86)
.22 (.93)
86
2.18*
Wider
Sleeve opening
3.25 (.51)
3.06 (.76)
.19 (.88)
85
1.97
-
Volume/fullness in
skirt
3.06 (.47)
3.06 (.56)
.00 (.68)
86
.00
-
Sleeve length
3.27 (.50)
2.98 (.76)
.29 (.85)
85
3.16**
Longer
Torso length
3.20 (.63)
2.94 (.85)
.26 (.92)
86
2.68**
Longer
Skirt length
3.17 (.77)
2.80 (.73)
.37 (.94)
86
3.64***
Longer
Overall dress length
3.14 (.64)
2.72 (.67)
.42 (.76)
84
5.13***
Longer
Notes. ΔM =Real dress fit-AR dress fit.
A 5-point scale, anchored at too loose/long/wide (1), excellent fit (3), and too tight/short/narrow
(5), was used to measure the items.
*p < .05. ** p < .01. *** p < .001.
Page 34 of 35Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Journal of Fashion Marketing and Management
EVALUATING GARMENTS IN AUGMENTED REALITY WHEN SHOPPING ONLINE
Table 4.
Product Performance Comparison of Ten Apparel Attributes
ΔM
SD
df
t
Style
.60
1.69
86
3.30**
Fabric
2.01
2.15
85
8.68***
Color
.41
1.93
85
1.12
Coordination with other items
.48
1.45
85
3.04**
Details
.96
1.64
86
5.50***
Touch and feel
3.34
2.17
86
14.32***
Weight of garment
3.21
2.05
84
14.42***
Overall Fit
1.16
1.85
85
5.82***
Comfort
2.53
2.17
84
10.75***
Appearance on the body
.46
2.09
85
2.06*
Note. ΔM =Real dress performance-AR dress performance
*p < .05. **p < .01. ***p < .001.
Page 35 of 35 Journal of Fashion Marketing and Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
