International Journal of Fashion Design, Technology and

Education

ISSN: 1754-3266 (Print) 1754-3274 (Online) Journal homepage: http://www.tandfonline.com/loi/tfdt20

Factors that influence consumers’ purchase

intention of smart closets

Anna Perry

To cite this article: Anna Perry (2016): Factors that influence consumers’ purchase intention
of smart closets, International Journal of Fashion Design, Technology and Education, DOI:
10.1080/17543266.2016.1202332

To link to this article: http://dx.doi.org/10.1080/17543266.2016.1202332

Published online: 27 Jun 2016.

Submit your article to this journal

Article views: 5

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=tfdt20

Download by: [Tulane University] Date: 09 July 2016, At: 11:49

 INTERNATIONAL JOURNAL OF FASHION DESIGN, TECHNOLOGY AND EDUCATION, 2016
http://dx.doi.org/10.1080/17543266.2016.1202332

Factors that influence consumers’ purchase intention of smart closets

Anna Perry
Design and Merchandising, Colorado State University, Fort Collins, CO, USA

ABSTRACT ARTICLE HISTORY

The current study incorporated the theory of planned behaviour to examine how various factors Received 15 March 2016
influence consumers’ purchase intention of smart closets. An online survey was conducted with Accepted 13 June 2016
433 participants. The findings indicated that aesthetics was insignificant, function was important,
KEYWORDS
while the key determinant of purchase intention was compatibility via attitude and social Attitude; subjective norm;
influence. To create a positive attitude, function and compatibility should be improved. To perceived compatibility;
enhance social influence, compatibility and technology rapidly changing played key roles. perceived performance; fear
Consumers’ confidence in using smart closets was increased as performance was higher, while it of obsolescence
decreased as technology was rapidly moving. The current study, for the first time, indicated that
fear of obsolescence is not always an unfavourable condition. Although rapidly changing
technology impeded self-confidence in using smart closets, it also increased social influence,
Downloaded by [Tulane University] at 11:49 09 July 2016

which in turn enhanced purchase intention.

1. Introduction ease or difficulty of performing a behaviour (Ajzen,

Selection of clothing is a daily routine but not an easy
routine. People often experience difficulties in finding
and matching clothing for the right occasion (Al-Omar
et al., 2013). Smart closets can address the issues of
managing clothing inventory, finding the right matches,
recommending appropriate styles, managing laundry,
connecting with fashion vendors, and borrowing apparel
via social network (Al-Omar et al., 2013). However, little
is known about how consumers perceive and respond
to smart closets. Business managers cannot assume
that consumers are willing to purchase smart closets.
Understanding the factors that influence consumers’
willingness to purchase is important for design and
implementation of smart closets. Therefore, the purpose
of this study was to explore how various factors influence
consumers’ purchase intention of smart closets.
2. Literature review
2.1. Theoretical background
The theory of planned behaviour proposed that human
behaviours are guided by three factors: attitude, subjec-
tive norm, and behavioural control (Ajzen, 1991, 2002)
(Figure 1). Attitude refers to the degree to which a person
has a favourable evaluation of a behaviour, subjective
norm refers to the perceived social pressure from signifi-
cant others, and behaviour control refers to the perceived
1991).
A large number of studies have empirically proven the
relations between the three factors and behavioural
intentions (e.g. Ajjan & Hartshorne, 2008; Lam, Cho,
& Qu, 2007). Because the three factors are important
influences on behaviour intention, it is important to
understand which antecedents may affect the three fac-
tors. Researchers have extended the theory with various
antecedents (Figure 1). For example, Ajzen (2002) pro-
posed behavioural beliefs, normative beliefs, and control
beliefs as antecedents of attitude, subjective norm, and
behaviour control, respectively. Taylor and Todd
(1995) proposed that usefulness, ease of use, and com-
patibility are antecedents of attitude. Other empirical
studies have also chosen different antecedents to fit indi-
vidual content; for example, in an information technol-
ogy study, antecedents of attitude are IT beliefs and
technology fit (Lam et al., 2007). However, in an online
shopping research, antecedents of attitude are privacy
and security (Vijayasarathy, 2004).
In order to fit with the current research content of
smart closets, all antecedents were carefully selected.
Smart closets are virtual technologies and have both
functional and aesthetic purposes. Therefore, a technol-
ogy variable (fear of obsolescence), a functional variable
(perceived performance), and an aesthetic variable
(perceived aesthetics) were selected (Grewal, Gotlieb, &
Marmorstein, 1994; Hwang, 2014; Venkatesh & Brown,

CONTACT Anna Perry anna.perry@colostate.edu Design and Merchandising, Colorado State University, 1574 Campus Delivery, Gifford 313, Fort Collins,
CO 80523-1574, USA
© The Textile Institute and Informa UK Ltd 2016
 2 A. PERRY
Figure 1. The theory of planned behaviour (Ajzen, 1991).
2001). In addition, smart closets are a new phenomenon
and unknown for most consumers. It is important to
know whether smart closets are suitable for consumers’
lives. Thus, a suitability variable (perceived compatibil-
ity) was also selected (Ko, Sung, & Yun, 2009).
Downloaded by [Tulane University] at 11:49 09 July 2016
2.2. Perceived aesthetics and attitude
Perceived aesthetics refers to the degree to which the
product is perceived as having a pleasing appearance
(Lamb & Kallal, 1992). Aesthetics significantly influences
consumers’ perceptions of products (Mumcu & Kimzan,
2015). Product aesthetic experience evokes positive
emotions, which in turn create positive attitudes
(Wang, Hong, & Lou, 2010). Empirical studies have
shown that aesthetics affects consumers’ attitude towards
retail brands (Page & Herr, 2002), visual objects (Bar &
Neta, 2006), and advertisements (Clow, James, Kranen-
burg, & Berry, 2006). It is important to understand
how aesthetics is related to attitude towards buying
smart closets. This knowledge is useful for product devel-
opers to decide how much effort they should put into
designs. Therefore, the following hypothesis was
proposed:
H1: Perceived aesthetics is positively related to attitude
towards buying.
2.3. Perceived compatibility, attitude, and
subjective norm
Perceived compatibility refers to the degree to which the
product is perceived as appropriate with one’s existing
needs, lifestyles, and technology (Ko et al., 2009). Ajjan
and Hartshorne (2008) suggested that when compatibil-
ity increases, the attitude is likely to become more
positive. Empirical studies have indicated that consu-
mers’ perceptions of compatibility significantly predicted
their attitude towards online shopping (Vijayasarathy,
2004), mobile payment services (Schierz, Schilke, &
Wirtz, 2010), software development (Hardgrave, Davis,
& Riemenschneider, 2003), and Internet technology
(Ajjan & Hartshorne, 2008).
Perceived compatibility may also influence subjective
norm. Both information resources (e.g. social media) and
significant others (e.g. peers and family) are important
influences on subjective norm (Bhattacherjee, 2000).
Information influences usually affect early adopters,
whereas significant others affect later adapters (Brown,
Venkatesh, & Bala, 2006). Because smart closets are a
relatively new phenomenon, it was expected that consu-
mers are more affected by information resources than
significant others. Thus, the current study only included
information influences as antecedents of subjective
norm. Compatibility is product information which may
be described on product instructions or social media.
Therefore, as an information influence, compatibility
possibly may affect subjective norm. In addition, both
compatibility and subjective norm share a key concept:
appropriateness. Compatibility emphasises that a pro-
duct fits well with lifestyle. Subjective norm emphasises
a behaviour fits well with others’ expectations. A more
appropriate product is more likely to fit with others’
expectations. Therefore, compatibility possibly positively
influences subjective norm.
It has been well known that compatibility plays a key
role in selecting mates and friends (Levinger & Rands,
1985). However, it is unsure how importantly that com-
patibility plays a role in selecting technology products
when compared with other factors. To help industry
understand the importance of compatibility and create
more consumer-friendly smart closets, it is necessary to
know whether compatibility is related to attitude and
subjective norm. Therefore, the following relationships
were hypothesised:
H2: Perceived compatibility is positively related to atti-
tude towards buying.
H3: Perceived compatibility is positively related to sub-
jective norm.
 INTERNATIONAL JOURNAL OF FASHION DESIGN, TECHNOLOGY AND EDUCATION 3

2.4. Perceived performance, attitude, and self- delivers a social pressure information: the technology is
efficacy out-of-date soon; therefore, you should catch up to the
new technology trend in order to fit with others. For
Perceived performance refers to the effect when a pro-
example, when a majority of consumers are using smart-
duct performs as expected (Grewal et al., 1994). Product
phones and continuing to update their phones, individ-
performance appears to play a key role in influencing
uals who still have old cell phones may perceive social
attitude (Manaktola & Jauhari, 2007). If a product is per-
pressure to buy new smartphones in order to fit with
ceived as not able to perform as anticipated, consumers’
others.
attitude towards buying is usually low (Akturan &
Fear of obsolescence may also affect consumers’ self-
Tezcan, 2002). If the product is perceived as being able
efficacy. When technology quickly becomes out-of-date
to perform as well as expected, the attitude towards buy-
and new versions continue updating, an unsure situation
ing is also good.
is created. Consumers are not sure what type of functions
Product performance also appears to play an important
have been changed and whether their abilities are
role in increasing self-efficacy, which refers to the degree
enough to perform a task by using the newest technol-
to which one person has the ability to perform an action
ogy. The unsure situation decreases consumers’ confi-
(Bandura, 1977). Performance is one of the most impor-
dence in using a technology.
tant sources of efficacy information (Bandura, 1977). If
It is important to explore whether fear of obsolescence
a product is perceived as more performable, then consu-
is indirectly related to purchase intention through sub-
mers may have more confidence in their ability to perform
Downloaded by [Tulane University] at 11:49 09 July 2016

jective norm and self-efficacy. Theoretically, this knowl-

a task by using the product. Empirical study has supported
edge would clarify the fear of obsolescence concept and
the above proposition by showing a positive performance
its relation with other variables. Practically, understand-
enhanced self-efficacy, while a negative performance
ing the relationship would help marketers develop better
impeded self-efficacy (Krueger & Dickson, 1994).
business strategies. As the above literature indicated, fear
It is important to know how good the smart closets’ per-
of obsolescence was expected to have a positive influence
formance is. This knowledge would help designers decide
on subjective norm, but negatively influence self-efficacy:
how much effort they should put into improving func-
tions. Therefore, the following hypotheses were proposed: H6: Fear of obsolescence is positively related to subjec-
tive norm.
H4: Perceived performance is positively related to atti- H7: Fear of obsolescence is negatively related to self-
tude towards buying. efficacy.
H5: Perceived performance is positively related to self-
efficacy.
2.6. Attitude, subjective norm, and purchase
intention
2.5. Fear of obsolescence, subjective norm,
and self-efficacy Based on the theory of planned behaviour, individuals’
beliefs are expected to influence their attitude towards
Fear of obsolescence refers to the degree to which rapidly
a behaviour, which in turn affects intention to perform
changing technology is associated with fear of technol-
the behaviour (Ajzen, 1991). In addition, to fit in with
ogy becoming out-of-date (Venkatesh & Brown, 2001).
important others, individuals may perform certain beha-
Fear of obsolescence was proposed to directly link with
viours (Vijayasarathy, 2004). Empirically, numerous
technology adoption: because consumers are afraid of
studies have proven the link between attitude and inten-
technology obsolescence, and they lack self-confidence
tion, and between subjective norm and intention (e.g.
in technology, they do not want to purchase it (Venka-
Ajjan & Hartshorne, 2008; Lam et al., 2007; Vijayasar-
tesh & Brown, 2001). However, some empirical studies
athy, 2004). Thus, the following hypotheses were
have rejected the direct link, for example, in a household
proposed:
technology study (Brown et al., 2006) and in a wireless sen-
sor network service study (Lin, Lu, & Hsieh, 2011). In H8: Attitude towards buying is positively related to pur-
addition, fear of technology obsolescence is a general belief chase intention.
of technology rather than a specific belief of smart closets. H9: Subjective norm is positively related to purchase
intention.
Therefore, the current study proposed that fear of obsoles-
cence may indirectly influence purchase intention via
2.7. Self-efficacy and purchase intention
mediators, such as subjective norm and self-efficacy.
As an information resource, fear of obsolescence may In the current study, self-efficacy was used to substitute
influence subjective norm. Fear of obsolescence may be the behaviour control variable from the theory of
perceived from mass media or significant others, and it planned behaviour. Justifications were described as the
 4 A. PERRY

Figure 2. Proposed model.

following. Firstly, self-efficacy and behaviour control are 3.2. Measures

Downloaded by [Tulane University] at 11:49 09 July 2016

very similar concepts: both are concerned with perceived

All measures were adapted from previous studies and
ability to perform a behaviour (Ajzen, 2002). Therefore,
were slightly modified to fit with the current research con-
many empirical studies replaced behaviour control with
tent, smart closets (Appendix). Each measure included
self-efficacy (e.g. Ajjan & Hartshorne, 2008; Lam et al.,
three items, which were assessed by a seven-point Likert
2007). Secondly, some empirical studies have indicated
scale anchored by 1 (strongly disagree) and 7 (strongly
that self-efficacy, rather than behaviour control, was
agree). Appendix also indicated the adaption resources
directly related to behaviour intention (e.g. Vijayasar-
and Cronbach’s α values of all measures, which ranged
athy, 2004). Thirdly, because smart closets may be a
from .75 to .94.
complex technology, especially for novice consumers,
researchers suggested including a self-confidence vari-
able, self-efficacy (Compeau & Higgins, 1995). 3.3. Participants
When consumers have confidence that they are
Empirical studies have shown that research data col-
capable of using a technology, they will be more likely
lected from MTurk in various fields are valid and reliable
to use it (Sun, Wang, Guo, & Peng, 2013). The link
(Buhrmester, Kwang, & Goslin, 2011; Goodman, Cryder,
between self-efficacy and purchase intention has been
& Cheema, 2013). For example, demographics are accu-
demonstrated in various studies, including online shop-
rate (Rand, 2012), psychometric properties are valid
ping (Vijayasarathy, 2004), Internet technologies (Ajjan
(Buhrmester et al., 2011), the results are reliable (Good-
& Hartshorne, 2008), and information technology
man et al., 2013), and differences between MTurk par-
(Lam et al., 2007). Therefore, the following hypothesis
ticipants and traditional samples are not significant
was proposed and Figure 2 displays the proposed
(Goodman et al., 2013). Therefore, a total of 433 partici-
research model:
pants were recruited from Mturk. Table 1 shows all
H10: Self-efficacy is positively related to purchase demographic information.
intention.

3. Methods
3.1. Study design and procedure
A quantitative approach was used to examine the
proposed mode. A recruit letter, a consent form, smart
closets’ instructions (e.g. pictures and functions), and a
10-minute online survey were developed and submitted
to the Institutional Review Board (IRB). After receiving
permission, the survey was administered by Qualtrics
and then linked to Amazon Mechanical Turk (MTurk).
4. Results
4.1. Measurement model
The fit indices of the measurement model indicated an
acceptable fit (Bentler & Bonnet, 1980; Hair, Black,
Balin, & Anderson, 2010): The ratio of χ 2 to the degrees
of freedom (χ 2/df) was 2.76 ( = 556.82, p < .0001), com-
parative fit index (CFI) was .95, Tucker–Lewis index
(TLI) was .93, standardised root mean square residual
(SRMR) was .06, and root mean square error of approxi-
mation (RMSEA) was .06 (Table 2).
 INTERNATIONAL JOURNAL OF FASHION DESIGN, TECHNOLOGY AND EDUCATION 5

Table 1. Demographics. The measurement model also had a good convergent

Frequency (n) % validity, which refers to the degree to which the measure-
Gender Male 190 43.9 ments are related to the measured construct (Chan,
Female 243 56.1
Age 18–25 81 18.7 Cheung, Shi, & Lee, 2015) (Table 3). Factor loading,
25.1–35 199 46 composite reliability, and average variance extracted
35.1–45 79 18.2
≥45.1 74 17.1
(AVE) were used to assess convergent validity (Fornell
Income ≤25,000 163 37.6 & Larcker, 1981). The standardised factor loadings ran-
25,001–50,000 141 32.6 ged from .70 to .96, which were higher than the suggested
50,001–75,000 73 16.9
≥75,001 56 12.9 value of .50 (Hair et al., 2010). The composite reliability
Education ≤High school 86 19.9 of measurements ranged from .86 to .96, considerably
Associate 70 16.2
Bachelor 183 42.3 above the cut-off value, .7 (Hair et al., 2010). The AVE
≥Master 94 21.8 of all measurements ranged from .67 to .90, exceeding
Age Max Min Mean SD the acceptable value, .5 (Hair et al., 2010).
78 19 35.06 11.23 Each construct also had a good discriminant validity,
which refers to the degree to which the item is not related
to other constructs. Discriminant validity can be assessed
Table 2. Fit indices of the measurement and structural models. by testing whether the correlations are smaller than the
Fit index Recommended value Measurement model Structural model square roots of AVEs (Fornell & Larcker, 1981). As
Downloaded by [Tulane University] at 11:49 09 July 2016

χ²/df ⩽3.00 2.76 2.74 shown in Table 4, all constructs met this requirement.
CFI ⩾0.95 0.95 0.95
TLI ⩾0.90 0.93 0.94
SRMR ⩽0.08 0.06 0.06
RMSEA ⩽0.06 0.06 0.06
4.2. Structural model and hypothesis test
The structural model had satisfactory levels of fit indices:
The measurement model had a good internal χ 2/df = 2.74, χ² (215) = 588.27, p < .0001, CFI = .95, TLI
reliability, which was assessed by Cronbach’s α and = .94, SRMR = .06, RMSEA = .06. The standardised coef-
item-total correlation. The Cronbach’s α values of all ficients of eight proposed paths were significant and two
constructs ranged from .85 to .96, exceeding the accepted were insignificant (Figure 3).
cut-off of .7 (Hair et al., 2010). The item-total correlation Results indicated that consumers’ attitude towards
indicates whether each item is related to the overall scale. buying smart closets was insignificantly influenced
It ranged from .61 to .94, exceeding the accepted cut-off by aesthetics attributes (H1: β = .08, p = .10), rejecting
of .30 as well (Nunnally & Bernstein, 1994). hypothesis H1, while consumers’ attitude towards buying

Table 3. Internal reliability and convergent validity of the constructs.

Internal reliability Convergent and discriminant validity
Item Cronbach’s α Item-total correlation Loading Composite Reliability AVE
Perceived aesthetics Aes_1 .96 .90 .93 .96 .88
Aes_2 .92 .95
Aes_3 .91 .94
Perceived Compatibility Com_1 .90 .75 .79 .90 .75
Com_2 .82 .87
Com_3 .83 .93
Fear of obsolescence Fea_1 .85 .75 .86 .87 .69
Fea_2 .61 .70
Fea_3 .79 .92
Perceived performance Per_1 .87 .82 .89 .89 .73
Per_2 .80 .88
Per_3 .69 .80
Self-efficacy Sel_1 .86 .70 .84 .86 .67
Sel_2 .73 .79
Sel_3 .77 .82
Attitude Att_1 .91 .79 .86 .91 .76
Att_2 .83 .87
Att_3 .86 .92
Subjective norm Sub_1 .96 .92 .94 .96 .90
Sub_2 .94 .96
Sub_3 .91 .94
Purchase intention Int_1 .94 .82 .96 .90 .88
Int_2 .90 .94
Int_3 .89 .95
 6 A. PERRY

Table 4. Correlations between latent constructs and the square Purchase intentions were significantly affected by atti-
root of AVE. tude towards using (H8: β = .64, p < .0001) and subjective
1 2 3 4 5 6 7 8 norm (H9: β = .32, p < .0001), supporting hypotheses H8
1. Perceived aesthetics .94 and H9. However, purchase intention was insignificantly
2. Perceived .66 .87
compatibility affected by self-efficacy (H10: β = −.06, p = .24), rejecting
3. Fear of obsolescence −.07 .06 .83 hypothesis H10. Together, they accounted for 70% of var-
4. Perceived .75 .76 −.11 .85
performance iance in purchase intention.
5. Self-efficacy .58 .56 −.26 .78 .82
6. Attitude towards .68 .84 .02 .79 .58 .87
buying
7. Subjective norm .36 .67 .35 .45 .23 .58 .95 5. Discussion
8. Purchase intention .56 .83 .11 .61 .39 .77 .68 .94
Note: Bold and italic number indicates square root of AVE. The present study developed a survey to investigate fac-
tors that influence consumers’ willingness to purchase
smart closets. The theory of planned behaviour was
smart closets was significantly influenced by compatibil- used as a framework. Antecedents of attitude, subjective
ity (H2: β = .72, p < .0001) and performance (H4: β = .18, norm, self-efficacy, and purchase intention were
p = .009), supporting hypotheses H2 and H4. Together, examined.
they accounted for 86% of the variance in attitude
Downloaded by [Tulane University] at 11:49 09 July 2016

towards buying.
Subjective norm was influenced by both compatibility
(H3: β = .65, p < .0001) and fear of obsolescence (H6: β
= .31, p < .0001), supporting hypotheses H3 and H6.
Therefore, information influences, such as compatibility
and fear of technology obsolescence, were significant
social pressure sources of the subjective norm. Together,
they accounted for 54% of the variance in subjective
norm.
Self-efficacy was positively influenced by performance
(H5: β = .75, p < .0001) and negatively impacted by fear
of obsolescence (H7: β = −.18, p < .0001), supporting
hypotheses H5 and H7. Together, they accounted for
63% of variance in self-efficacy.
5.1. Antecedents of attitude
The result demonstrated that attitude towards buying
was significantly influenced by the product’s functional
variable, performance, and suitability variable, compat-
ibility. The standardised coefficient of compatibility (β
= .72) was four times more than the coefficient of per-
formance (β = .18). Therefore, although consumers
expect smart closets would perform as well as described,
being suitable to the consumers’ existing life styles,
needs, and technology are the most important require-
ments (Schierz et al., 2010). Design features (β = .08)
were insignificant when considered with suitability and
performance. The results suggest that suitability and

Figure 3. Structural model results. Note: *p = .05, **p = 0.001, *** p = .0001. Solid line indicates significant path and dashed line indi-
cates insignificant path.
 INTERNATIONAL JOURNAL OF FASHION DESIGN, TECHNOLOGY AND EDUCATION
function of smart closets are far more important than
aesthetics.
5.2. Antecedents of subjective norm
The results indicated that the suitability of smart closets
(compatibility) and a belief of technology (fear of obso-
lescence) were important information influences of sub-
jective norm. The standard coefficient of compatibility
(β = .65) was twice as much as the coefficient of fear of
obsolescence (β = .31). Again, compatibility had the
greatest impact. These findings extended the antece-
dents’ range of subjective norm: not only are significant
others (e.g. family and peers) and mass media (e.g. TV
and magazine) resources of subjective norm. Product
information (e.g. compatibility) and messages prompted
from mass media (e.g. fear of obsolescence) may also be
significant resources. Fear of obsolescence was positively
Downloaded by [Tulane University] at 11:49 09 July 2016
related to subjective norm. Therefore, rapid changing of
technology seems to evoke more social influences.
5.3. Antecedents of self-efficacy
Self-efficacy was positively influenced by the product
functionality variable, performance, and negatively
related to the belief of technology, fear of obsolescence.
Thus, confidence in smart closets’ performance increases
consumers’ self-confidence in using smart closets (Krue-
ger & Dickson, 1994), while uncertainty of rapidly chan-
ging technology decreases consumers’ self-confidence in
using smart closets. Performance (β = .75) was the most
important driving force of self-efficacy, while fear of
technology obsolescence (β = −.18) was a significant
barrier of self-efficacy.
5.4. Antecedents of purchase intention
Consistent with the theory of planned behaviour, pur-
chase intention was significantly influenced by attitude
(β = .64) and subjective norm (β = .32). Thus, a favour-
able attitude and social influences were sources of pur-
chase intention of smart closets. However, the present
study did not support the relationship between self-effi-
cacy (β = −.06) and purchase intention. Confidence in
using smart closets does not necessarily lead to buying
intention. A previous study also indicated an insignifi-
cant relationship between the two constructs, for
example, in a mobile phone learning study (Gunawar-
dana & Ekanayaka, 2009). A possible reason was that
nowadays participants had sufficient skills, experiences,
and confidence in using technology. Therefore, skills
and confidence in using technology do not necessarily
influence their purchase intention.
7
This study, for the first time, identified complicated
roles of fear of technology obsolescence in relation to
purchase intention. Firstly, the results suggest that
there was an indirect relationship between fear of obso-
lescence and purchase intention through mediators.
Secondly, fear of obsolescence was not always an
unfavourable condition. From the original concept to
empirical tests, fear of obsolescence was labelled as a
barrier of purchase intention (e.g. Brown et al., 2006;
Venkatesh & Brown, 2001). However, fear of technology
obsolescence negatively influenced self-efficacy and posi-
tively influenced subjective norm. Self-efficacy was not
necessarily related to willingness to purchase, while sub-
jective norm remained a significant resource of purchase
intention. Theoretically, the results from the current
study expand the roles of fear of technology obsoles-
cence. In addition, the complicated roles of fear of obso-
lescence are probably the reason why some studies
indicated a significant relationship between fear of obso-
lescence and purchase intention (e.g. Venkatesh &
Brown, 2001), but other studies reported an insignificant
result (e.g. Brown et al., 2006; Lin et al., 2011). Practi-
cally, that is probably the reason why technology has
been updating so rapidly nowadays: even fear of obsoles-
cence has downsides that hurt consumers’ confidence in
using a product, but it also has a unique advantage that
increases purchase intention of the product via subjective
norm.
6. Conclusion
The current study incorporated the theory of planned
behaviour to empirically test how four antecedents, aes-
thetic attributes, functional performance, a general belief
of technology, and an appropriate factor, influence pur-
chase intention through three mediators, attitude, social
pressure, and self-confidence. Aesthetics was not impor-
tant. Compatibility was the most important determinant
of both attitude and subjective norm. Performance
played a significant secondary role in determining atti-
tude and the most important role in self-efficacy. Fear
of obsolescence enhanced subjective norm but impeded
self-efficacy. Attitude was the primary determinant and
subjective norm was the secondary determinant of pur-
chase intention. Self-efficacy did not influence purchase
intention.
The findings indicated that attitude and social influ-
ence were the key determinants of the purchase intention
of smart closets, while confidence in using smart closets
was not related to purchase intention. To create a posi-
tive attitude, smart closets’ function and compatibility
with lifestyles should be improved. To enhance social
influence (subjective norm), compatibility and rapid
 8 A. PERRY
updating of technology played key roles. Consumers’
confidence in using smart closets was increased as pro-
duct performance was higher, while it decreased as tech-
nology updating was too fast. The current study, for the
first time, suggests that fear of technology obsolescence is
not always an unfavourable condition. Although rapid
technology obsolescence impeded self-confidence in
using smart closets, it also increased subjective norm,
which in turn enhanced purchase intention.
6.1. Contribution
This study provided insights into the intention of pur-
chasing smart closets. It is the starting point of under-
standing how to develop smart closets that consumers
are willing to buy. Businesses may create more effective
technology and marketing strategies based on the cur-
rent study. In addition, this study expanded the theory
Downloaded by [Tulane University] at 11:49 09 July 2016
of planned behaviour application on the technology
field and identified several new antecedents of attitude,
subjective norm, and self-efficacy. The proposed model
presented more comprehensive understanding of consu-
mers’ purchase intention of smart closets.
6.2. Limitation and future studies
The current study investigated fear of obsolescence and
perceived compatibility as antecedents of subjective
norm. Numerous researchers have demonstrated that
family, peers, superiors, and mass media influence sub-
jective norm (Bhattacherjee, 2000). It is unsure how
fear of obsolescence, perceived compatibility, family,
peers, superiors, and mass media together influence sub-
jective norm. Future studies should investigate them in
one study. Future studies also should explore and ident-
ify more possible antecedents of subjective norm. In
addition, since the participants read the instructional
materials that showed pictures and functions of the
smart closets, their responses may have been influenced
from the smart closets described in the materials.
References
Ajjan, H., & Hartshorne, R. (2008). Investigating faculty
decisions to adopt Web 2.0 technologies: Theory and
empirical tests. Internet and Higher Education, 11(2), 71–80.
Ajzen, I. (1991). The theory of planned behavior.
Organizational Behavior and Human Decision Processes,
50(2), 179–211.
Ajzen, I. (2002). Perceived behavioral control, self-efficacy,
locus of control, and the theory of planned behavior.
Journal of Applied Social Psychology, 32(4), 665–683.
Akturan, U., & Tezcan, N. (2002). Mobile banking adoption of
the youth market: Perceptions and intentions. Marketing
Intelligence & Planning, 30(4), 444–459.
Al-Omar, N. N., Al-Rashed, N. M., Al-Fantoukh, H. I., al-
Osaimi, R. M., Al-Dayel, Al-H. A., & Mostefai, S. (2013).
The design and development of a web-based virtual closet:
The smart closet project. Journal of Advanced
Management Science, 1(1), 124–128.
Baker, M. J., & Churchill, G. A. Jr. (1977). The impact of phys-
ically attractive models on advertising evaluations. Journal
of Marketing Research, 14(4), 538–555.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of
behavioral change. Psychological Review, 84(2), 191–215.
Bar, M., & Neta, M. (2006). Humans prefer curved visual
objects. Psychological Science August, 17(8), 645–648.
Bentler, P. M., & Bonnet, D. C. (1980). Significance tests and
goodness of fit in the analysis of covariance structures.
Psychological Bulletin, 88(3), 588–606.
Bhattacherjee, A. (2000). Acceptance of e-commerce services:
The case of electronic brokerages. IEEE Transactions on
Systems, Man, and Cybernetics, Part A, 30(4), 411–419.
Brown, S., & Venkatesh, V. (2005). Model of adoption of tech-
nology in households: A baseline model test and extension
incorporating household life cycle. MIS Quarterly, 29(3),
399–426.
Brown, S. A., Venkatesh, V., & Bala, H. (2006). Household
technology use: Integrating household life cycle and the
model of adoption of technology in households. The
Information Society, 22(4), 205–218.
Buhrmester, M., Kwang, T., & Goslin, S. D. (2011). Amazon’s
Mechanical Turk: A new source of inexpensive, yet high-
quality, data? Perspectives on Psychological Science, 6(1),
3–5.
Chan, T. K. H., Cheung, C. M. K., Shi, N., & Lee, M. K. O.
(2015). Gender differences in satisfaction with Facebook
users. Industrial Management & Data Systems, 115(1),
182–206.
Clow, K. E., James, K. E., Kranenburg, K. E., & Berry, C. T.
(2006). The relationship of the visual element of an adver-
tisement to service quality expectations and source credi-
bility. Journal of Services Marketing, 20(6), 404–411.
Compeau, D. R., & Higgins, C. A. (1995). Application of social
cognitive theory to training for computer skills. Information
Systems Research, 6(2), 118–143.
Fornell, C., & Larcker, D. F. (1981). Evaluating structural
equation models with unobservable variables and measure-
ment error. Journal of Marketing Research, 18(1), 39–50.
Goodman, J. K., Cryder, C. E., & Cheema, A. (2013). Data col-
lection in a flat world: The strengths and weaknesses of
Mechanical Turk samples. Journal of Behavioral Decision
Making, 26(13), 213–224.
Grewal, D., Gotlieb, J., & Marmorstein, H. (1994). The moder-
ating effects of message framing and source credibility on
the price-perceived risk relationship. Journal of Consumer
Research, 21(1), 145–153.
Gunawardana, K. D., & Ekanayaka, S. (2009). An empirical
study of the factors that impact medical representatives’
attitude toward the intention to use M-Learning for career
development. Sasin Journal of Management, 15(1), 1–26.
Hair, J. F., Black, W. C., Balin, B. J., & Anderson, R. E. (2010).
Multivariate data analysis: Maxwell Macmillan
International Editions. Upper Saddle River, NJ: Prentice
Hall.
Hardgrave, B. C., Davis, F. D., & Riemenschneider, C. K.
(2003). Investigating determinants of software developers’
 INTERNATIONAL JOURNAL OF FASHION DESIGN, TECHNOLOGY AND EDUCATION
intentions to follow methodologies. Journal of Management
Information Systems, 20(1), 123–151.
Hwang, C. (2014). Consumers’ acceptance of wearable technol-
ogy: Examining solar-powered clothing (Master’s thesis).
Retrieved from http://lib.dr.iastate.edu/cgi/viewcontent.
cgi?article = 4957&context = etd
Khasawneh, M., & Ibrahim, H. (2012). A model for adoption
of ICT in Jordanian higher education institutions: An
empirical study. Journal of e-Learning & Higher
Education, 2012, 1–10.
Ko, E., Sung, H., & Yun, H. (2009). Comparative analysis of
purchase intentions toward smart clothing between
Korean and US consumers. Clothing and Textiles Research
Journal, 27(4), 259–273.
Krueger, N. Jr., & Dickson, P. R. (1994). How believing in our-
selves increases risk taking: Perceived self-efficacy and
opportunity recognition. Decision Sciences, 25(3), 385–400.
Lam, T., Cho, V., & Qu, H. (2007). A study of hotel employee
behavioral intentions towards adoption of information
technology. Hospitality Management, 26(1), 49–65.
Lamb, J. M., & Kallal, M. J. (1992). A conceptual framework for
Downloaded by [Tulane University] at 11:49 09 July 2016
apparel design. Clothing and Textiles Research Journal, 10
(2), 42–47.
Levinger, G., & Rands, M. (1985). Compatibility in marriage
and other close relationships. In W. Ickes (Ed.),
Compatible and incompatible relationships (pp. 309–331).
New York, NY: Springer.
Lin, F.-R., Lu, I.-Y., & Hsieh, P.-S. (2011, May 25–27).
Understanding the adoption of wireless sensor network ser-
vice in households. 2011 International Joint Conference on
Service Sciences, Taipei, Taiwan, pp. 218–222.
MacKenzie, S. B., Lutz, R. J., & Belch, G. E. (1986). The role of
attitude toward the ad as a mediator of advertising effective-
ness: A test of competing explanations. Journal of Marketing
Research, 23(2), 130–143.
Manaktola, K., & Jauhari, V. (2007). Exploring consumer atti-
tude and behavior towards green practices in the lodging
industry in India. International Journal of Contemporary
Hospitality Management, 19(5), 364–377.
Mumcu, Y., & Kimzan, H. S. (2015). The effect of visual pro-
duct aesthetics on consumers’ price sensitivity. Procedia
Economics and Finance, 26, 528–534.
Nunnally, J., & Bernstein, I. (1994). Psychometric theory.
New York, NY: McGraw-Hill.
Page, C., & Herr, P. M. (2002). An investigation of the pro-
cesses by which product design and brand strength interact
to determine initial affect and quality judgments. Journal of
Consumer Psychology, 12(2), 133–147.
Rand, D. G. (2012). The promise of Mechanical Turk: How
online labor markets can help theorists run behavioral
experiments. Journal of Theoretical Biology, 299, 172–179.
Schierz, P. G., Schilke, O., & Wirtz, B. W. (2010).
Understanding consumer acceptance of mobile payment
services: An empirical analysis. Electronic Commerce
Research and Applications, 9(3), 209–216.
Sun, Y., Wang, N., Guo, X., & Peng, Z. (2013). Understanding
the acceptance of mobile health services: A comparison and
integration of alternative models. Journal of Electronic
Commerce Research, 14(2), 184–200.
Taylor, S., & Todd, P. A. (1995). Understanding information
technology usage: A test of competing models.
Information Systems Research, 6(2), 144–176.
9
Venkatesh, V., & Brown, S. A. A. (2001). Longitudinal investi-
gation of personal computers in homes: Adoption determi-
nants and emerging challenges. MIS Quarterly, 25(1),
71–102.
Vijayasarathy, L. R. (2004). Predicting consumer intentions to
use on-line shopping: The case for an augmented technol-
ogy acceptance model. Information & Management, 41(6),
747–762.
Wang, Y. J., Hong, S., & Lou, H. (2010). Beautiful
beyond useful? The role of web aesthetics. Journal of
Computer Information Systems, 50(3), 121–129.
Appendix
Measures and items
Perceived aesthetics attributes (Hwang, 2014; Cronbach’s α
= .94):
Aes_1: The appearance of the smart closet is aesthetically
appealing to me.
Aes_2: The designs of the smart closet are aesthetically
appealing to me.
Aes_3: The overall style of smart closet is appealing to me.
Perceived compatibility (Ko et al., 2009; Cronbach’s α = .83):
Com_1: This smart closet would coordinate well with the
other technology I own.
Com_2: The smart closet would be more compatible
with my current need than the wardrobe I cur-
rently use.
Com_3: The smart closet would be appropriate for my
lifestyle.
Perceived performance (Grewal et al., 1994; Cronbach’s α
= .90):
Per_1: How confident are you that the smart closet will per-
form as described?
Per_2: How certain are you that the smart closet will work
satisfactorily?
Per_3: Do you feel that the smart closet will perform the
functions that were described in the information
page?
Fear of obsolescence (Brown & Venkatesh, 2005; Cronbach’s α
= .75):
Fea_1: The trends in technological advancement are worri-
some to me.
Fea_2: I fear that today’s smart closets will be obsolete fairly
soon.
Fea_3: I am worried about the rapid advances in high
technology.
Self-efficacy (Khasawneh & Ibrahim, 2012; Cronbach’s α
= .90):
 10 A. PERRY
Sel_1: I feel comfortable using a smart closet.
Sel_2: If I wanted to, I could easily operate a smart closet on
my own.
Sel_3: I can use a smart closet if no one is around to help me.
Attitude towards buying (MacKenzie, Lutz, & Belch, 1986;
Cronbach’s α = .92):
Att_1: Buying the smart closet is a good idea.
Att_2: Buying the smart closet is fun.
Att_3: I like buying the smart closet.
Subjective norm (Taylor & Todd, 1995; Cronbach’s α = .90):
Downloaded by [Tulane University] at 11:49 09 July 2016
Sub_1: People who are important to me think that I should
buy the smart closet.
Sub_2: Those in my social circle think I should buy the
smart closet.
Sub_3: My peers think I should buy the smart closet.
Purchase intention (Baker & Churchill, 1977; Cronbach’s α
= .81):
Int_1: It is likely that I will buy this smart closet when it
becomes available.
Int_2: I would purchase smart closets.
Int_3: I am willing to buy smart closets.