Journal of Services Marketing

Consumer participation in online product recommendation services: augmenting the technology acceptance
model
Xiaojing Sheng Mohammadali Zolfagharian
Article information:
To cite this document:
Xiaojing Sheng Mohammadali Zolfagharian , (2014),"Consumer participation in online product recommendation services:
augmenting the technology acceptance model", Journal of Services Marketing, Vol. 28 Iss 6 pp. 460 - 470
Permanent link to this document:
[Link]
Downloaded on: 31 January 2016, At: 19:35 (PT)
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 19:35 31 January 2016 (PT)

References: this document contains references to 43 other documents.

To copy this document: permissions@[Link]
The fulltext of this document has been downloaded 1312 times since 2014*
Users who downloaded this article also downloaded:
Indranil Roy Chowdhury, Sanjay Patro, Pingali Venugopal, D. Israel, (2014),"A study on consumer adoption of technology-
facilitated services", Journal of Services Marketing, Vol. 28 Iss 6 pp. 471-483 [Link]
Nicole Koenig-Lewis, Adrian Palmer, (2014),"The effects of anticipatory emotions on service satisfaction and behavioral
intention", Journal of Services Marketing, Vol. 28 Iss 6 pp. 437-451 [Link]
Shannon Cummins, James W. Peltier, John A. Schibrowsky, Alexander Nill, (2014),"Consumer behavior in the online context",
Journal of Research in Interactive Marketing, Vol. 8 Iss 3 pp. 169-202 [Link]

Access to this document was granted through an Emerald subscription provided by emerald-srm:272736 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service
information about how to choose which publication to write for and submission guidelines are available for all. Please visit
[Link]/authors for more information.
About Emerald [Link]
Emerald is a global publisher linking research and practice to the benefit of society. The company manages a portfolio of
more than 290 journals and over 2,350 books and book series volumes, as well as providing an extensive range of online
products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee on Publication Ethics
(COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation.

*Related content and download information correct at time of download.

 Consumer participation in online product
recommendation services: augmenting the
technology acceptance model
Xiaojing Sheng and Mohammadali Zolfagharian
Department of Marketing, University of Texas-Pan American, Edinburg, Texas, USA

Abstract
Purpose – This paper aims to examine the complex role of consumer participation by empirically testing a theoretical model within the online
context of consumer using product recommendation agents (RAs) that integrates consumer participation into the technology acceptance model
(TAM). Recent research starts questioning whether consumer participation is a double-edged sword, although its positive effects have been widely
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 19:35 31 January 2016 (PT)

documented in the services marketing literature.

Design/methodology/approach – Data were collected from 243 respondents comprising both genders with varied age groups, education levels,
household income and ethnicities. Participants interacted with an existing RA, searching product information for a simulated purchase involving
either high or low financial risk. Confirmatory factor analysis (CFA) and Structural equations modeling (SEM) were used for assessing construct
validity and testing hypotheses, respectively.
Findings – Consumer participation was found to positively affect enjoyment but negatively affect perceived ease of use of an RA, which was
exacerbated when the purchase was of high financial risk. Predictions based on TAM were all supported in this study.
Research limitations/implications – This research evidenced consumer participation’s dual-effect and identified financial risk as a moderator of
consumer participation’ negative effect. Future research could employ a different design such as using controlled lab settings and conducting
longitudinal studies to replicate the study.
Practical implications – Findings from this research provide guidance to RA design regarding the amount of consumer participation, consumer
perceptions of RA use (ease of use, usefulness and enjoyment) and financial risk involved in the purchase.
Originality/value – This research sheds new light on the role of consumer participation through uncovering its dual-effect, thus complementing the
services marketing and RA literature.
Keywords Participation, Technology, Online, Recommendation
Paper type Research paper

An executive summary for managers and executive collaborative and content filtering (Ansari et al., 2000; Ariely
readers can be found at the end of this issue. et al., 2004). Collaborative filtering methods predict an
individual’s preferences based on the preferences of other
Introduction “like-minded” people, thus mimicking word-of-mouth
recommendations. For example, we come across
Online product recommendation agents (hereafter RAs) have recommendations such as “Customers Who Bought This Item
become increasingly available to consumers as an interactive Also Bought [ . . . ]” when shopping on [Link].
decision aid, aiming to facilitate consumers’ product search
Content filtering methods make recommendations on the
and evaluation process on the Internet (Wang and Benbasat,
basis of an individual’s preferences for product attributes. For
2009). By definition, RAs are software-based technology example, when searching for product information on
designed to help consumers systematically and efficiently [Link], we are asked a set of questions regarding
screen large sets of product alternatives and provide product product attributes such as brand, price range and features
recommendations based on consumer inputs generated from specific to the product.
the preference elicitation process (Xiao and Benbasat, 2007). To some degree, an RA’s interface design, along with the
When eliciting consumer preferences, RAs utilize different type of computer algorithms it uses to make
computer algorithms such as collaborative filtering, content recommendations, determines what and how much
filtering and hybrid methods that combine the merits of information consumers can provide in the process of using the
RA. But at the same time, regardless of the design and
computer algorithms, some consumers simply choose to
The current issue and full text archive of this journal is available at
spend more time interacting with an RA and expend more
[Link]/[Link]
effort, providing detailed information about their product
interests and likes or dislikes regarding certain product

Journal of Services Marketing

28/6 (2014) 460 –470 Received 18 April 2013
© Emerald Group Publishing Limited [ISSN 0887-6045] Revised 14 December 2013
[DOI 10.1108/JSM-04-2013-0098] Accepted 30 January 2014

460
 Consumer participation in online product recommendation services Journal of Services Marketing
Xiaojing Sheng and Mohammadali Zolfagharian Volume 28 · Number 6 · 2014 · 460 –470

attributes. Other consumers, however, choose to limit their usefulness refers to the extent to which an individual believes
participation to a minimum. The question arises: When using that using a technology will provide benefits to him or her in
RAs, does it matter to what extent consumers participate? performing certain activities, perceived ease of use reflects the
Does participation influence consumers’ evaluations of their belief that using the technology will be free of effort.
RA use experiences? Enjoyment taps the fun or pleasure aspect of using a
Existing RA research offers many insights by examining the technology and refers to the extent to which the activity of
effectiveness of computer algorithms underlying the resultant using a technology is “perceived to be enjoyable in its own
recommendations and the effect of RA use on consumer right, apart from any performance consequences that may be
search effort and decision quality. For example, some studies anticipated” (Davis et al., 1992, p.1113).
showed that using RAs could help consumers reduce cognitive TAM has been used in both organizational and consumer
effort, improve decision quality and decrease information contexts to explain user acceptance of a variety of technologies
overload (Häubl and Trifts, 2000; Todd and Benbasat, 1992, such as word processing software (Davis et al., 1989), business
2000). Others explored the issue of the types of computer graphics programs (Davis et al., 1992), executive information
algorithms that make effective recommendations (Ansari et al., systems (Pijpers and van Montfort, 2005), the World Wide
2000). Still others argue that RAs present much potential for Web (Moon and Kim, 2001), virtual try-on technology for
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 19:35 31 January 2016 (PT)

marketers to learn and generate knowledge about their online apparel shopping (Kim and Forsythe, 2008),
consumers (Ariely et al., 2004). Notwithstanding these smartphone use (Chun et al., 2012), mobile Internet
insights, there is a lack of research on the role of consumer technology (Venkatesh et al., 2012), e-service Web sites (Xu
participation in using RAs. This is an important gap that et al., 2013), shopbots (Gentry and Calantone, 2002) and RAs
warrants further research for two reasons. First, when (Wang and Benbasat, 2005). Results from these studies
consumers spend time and efforts in interacting with RAs and provide strong empirical support for the predictive validity and
providing input to the questions raised by RAs, consumer robustness of TAM. As Lee et al. (2003) summarized in their
participation becomes an inherent part of using RA review, TAM is “the most influential and commonly
technology. And yet, consumer participation has been a employed theory for describing an individual’s acceptance of
largely ignored concept in previous RA research. Moreover, information systems” (p. 752).
the role of consumers in past RA research is limited to being Parsimony is considered TAM’s main strength and has
a passive receiver of the RA technology. Therefore, examining become a catalyst for research efforts aimed at “broadening”
the impact of consumer participation and taking a the model by introducing additional predictors of intentions
consumer-centric perspective can further our understanding and by adding moderating variables to qualify the effects of
of RA use. Second, consumer participation has long been a perceived usefulness and perceived ease of use on intentions
topic of interest to services marketing researchers (Bateson, (Bagozzi, 2007). For example, as a TAM extension,
1985; Lovelock and Young, 1979). It is important to examine Venkatesh et al.’s (2003) unified theory of acceptance and use
whether the effects of consumer participation found in other of technology (UTAUT) was built upon a synthesis of eight
research contexts can be extended to the context of RA use. prominent theoretical models that have been used to explain
As such, this research intends to examine the role of user acceptance of technology. UTAUT predicts that
consumer participation in using RAs, thus filling the gap in the performance expectancy, effort expectancy, social influence
extant RA literature, by empirically testing a proposed and facilitating conditions determine intentions to use a
theoretical model that integrates consumer participation into technology and that age, gender, experience and voluntariness
the technology acceptance model (TAM). In so doing, this of use moderate various UTAUT relationships. Performance
research hopes to provide answers to the following two expectancy refers to the degree to which an individual believes
research questions: that using a technology will help achieve job performance
RQ1. Does participation in using an RA affect consumers’ gains and is conceptually similar to existing constructs such as
perceptions of the ease of use, usefulness, and perceived usefulness, extrinsic motivation, job-fit, relative
enjoyment in using the RA? advantage and outcome expectations. Effort expectancy
relates to the degree of ease associated with using a technology
RQ2. Do the positive effects of consumer participation widely and overlaps with the concept of perceived ease of use and
documented in the services marketing literature also complexity. Social influence is comparable to subjective
apply to the context of RA use? norms and that the definition of facilitating conditions is
similar to that of perceived behavioral control and
compatibility.
Research hypotheses Venkatesh et al. (2012) further extended TAM and
TAM and related research developed UTAUT2 in which hedonic motivation, price value
The TAM (Davis, 1989; Davis et al., 1989, 1992) is a theory and habit were introduced as three additional predictors of
for explaining and predicting technology acceptance and use. intentions and that voluntariness of use was dropped as a
Central to TAM is a causal chain of effects, whereby the actual moderator to fit the consumer technology use context.
behavior of adopting and using a technology is determined by Similarly, Baron et al. (2006) proposed “a consumer TAM”
intentions to adopt and use the technology, which, in turn, are that captures the unique aspects of TAM research in the
determined by a set of salient beliefs associated with the context of consumer markets for technology-based products
consequences of using the technology, i.e. perceived ease of or services. In Baron et al.’s model, perceived usefulness,
use, perceived usefulness and enjoyment. Whereas perceived perceived ease of use, perceived enjoyment, social influence

461
 Consumer participation in online product recommendation services Journal of Services Marketing
Xiaojing Sheng and Mohammadali Zolfagharian Volume 28 · Number 6 · 2014 · 460 –470

and perceived behavioral control are theorized to impact Researchers also point out that the benefit of reduced cost
intentions and that this impact is predicted to be moderated by from involving consumers as partial employees in service
consumer traits (i.e. demographic variables and personality production and delivery might be overshadowed by risks
characteristics such as self efficacy and need for interaction) associated with consumer participation (Hilton et al., 2013).
and experience. How does consumer participation influence perceived ease of
Without a doubt, TAM extensions such as UTAUT and use, perceived usefulness and enjoyment of using an RA?
UTAUT2 have helped broaden the original TAM When a higher level of consumer participation is needed to
conceptualization with newly added variables and use an RA, consumers will be expected to put forth a greater
relationships. However, we have yet to adequately “deepen amount of time and efforts. This will have a dampening effect
TAM in the sense of explaining PU and PEU, on perceived ease of use of the RA because perceived ease of
reconceptualizing existing variables in the model, or use is the degree of ease associated with using a technology
introducing new variables explaining how the existing and reflects the belief that using a technology is free of effort.
variables produce the effects they do” (Bagozzi, 2007, p. 244) But, on the other hand, more fun and pleasure can be derived
and to determine whether “the statistical-modelling-based from spending more time and making greater effort while
TAM developments ignore the paradoxes of technology interacting with an RA. The rationale is that the more the
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 19:35 31 January 2016 (PT)

acceptance faced by individuals” (Baron et al., 2006, p. 118). consumers use an RA, the more familiar they become with the
Researchers have made considerable inroad toward addressing RA and the better they understand how the RA works, thus
these important questions. For example, using a grounded leading to more favorable affective reactions to the use of the
theory approach, Mick and Fournier (1998) investigated RA. This reasoning is supported by the finding that consumer
consumers’ perspectives, meanings and experiences in relation participation in service production and delivery leads to
to a range of technological products from computers to greater customer satisfaction and service quality evaluations.
washing machines and developed a conceptual framework drawn We, therefore, propose that:
upon eight central paradoxes of technological products (e.g.
control/chaos, freedom/enslavement and engaging/disengaging) H1. The level of consumer participation in using an RA will
uncovered in the study and the paradoxes’ influences on negatively affect perceived ease of use of the RA.
consumers’ emotional reactions as well as behavioral coping
H2. The level of consumer participation in using an RA will
strategies. Baron et al. (2006) conducted two qualitative
positively affect enjoyment of using the RA.
studies (diary writing and personal interviews) in the UK to
examine consumers’ text message usage behavior and Different from perceived ease of use and enjoyment, perceived
unearthed paradoxical and counter-intuitive behaviors as well usefulness relates to consumer perceptions that using a
as the intense social and emotional elements in consumers’ technology will improve their task performance and bestow
actual text message usage. Our research contributes to this line them with benefits in carrying out certain activities. Consumer
of inquiry by proposing consumer participation as an participation in using an RA is not expected to influence
antecedent to perceived ease of use and enjoyment, thus perceived usefulness of the RA because performance
taking another step toward answering Bagozzi’s (2007) call to expectations of an RA are largely determined by the technical
deepen TAM through explaining perceived ease of use and design of that RA, not by how much consumers participate.
enjoyment of using RAs. Therefore, we do not propose a hypothesis regarding the
relationship between consumer participation in using an RA
Consumer participation, perceived ease of use and and perceived usefulness of the RA, but will observe this
enjoyment relationship post hoc.
Consumer participation in using an RA refers to the extent to
which consumers are involved in using the RA and manifests Perceived ease of use, perceived usefulness, enjoyment
in different ways. For example, consumers can participate and intention
through spending time interacting with the RA. Participation Specific to the RA use context, Gentry and Calantone (2002)
can also take the form of answering the questions raised by the adopted a SEM approach in comparing the predictive ability
RA and providing information regarding one’s product of three models, i.e. the theory of reasoned action, the theory
interests and preferences. of planned behavior and TAM. Results from Gentry and
The services marketing literature has accumulated a rich Calantone’s (2002) study showed that TAM outperformed
body of research that demonstrates the positive impact of the other two models in terms of the model fit and the amount
consumer participation on satisfaction and service quality of variance explained in behavioral intentions. Wang and
evaluations from consumers’ perspective (Bitner et al., 1997; Benbasat (2005) integrated trust into TAM and tested the
Cermak et al., 1994; Dellande et al., 2004; Chang et al., 2009) relationships between trust, perceived ease of use, perceived
and on reduced cost and improved productivity from firms’ usefulness and RA adoption intentions. Their study once
perspective (Bowers et al., 1990; Lovelock and Young, 1979; again confirmed the predictive validity of TAM and showed
Mills et al., 1983). But recent research is starting to question that perceived ease of use of an RA positively affected
whether consumer participation is a “double-edged sword” perceived usefulness of the RA which, in turn, positively
because consumer participation represents a source of affected adoption intentions. The authors also found that
uncertainty to employees and has been found to lead to perceived ease of use increased consumer trust in the RA and
employee feelings of role ambiguity and decreased job that trust positively impacted on perceived usefulness and
satisfaction (Chan et al., 2010; Hsieh et al., 2004). adoption intentions. Our research complements past RA

462
 Consumer participation in online product recommendation services Journal of Services Marketing
Xiaojing Sheng and Mohammadali Zolfagharian Volume 28 · Number 6 · 2014 · 460 –470

research by adding enjoyment, a construct that has yet to be different product alternatives and arrive at the optimal choice.
studied in the RA use context, to a set of predictions regarding This suggests that when financial risk is high, the positive
consumer use of RAs derived from TAM. We propose the effect of consumer participation on enjoyment will become
following hypotheses: stronger. Based on this line of reasoning, we propose that:

H3. Perceived ease of use of an RA will positively affect H8. The level of financial risk involved in a purchase will
enjoyment of using the RA. weaken the negative effect of consumer participation on
perceived ease of use of the RA.
H4. Perceived ease of use of an RA will positively affect
perceived usefulness of the RA. H9. The level of financial risk involved in a purchase will
strengthen the positive effect of consumer participation
H5. Enjoyment of using an RA will positively affect on enjoyment in using the RA.
perceived usefulness of the RA.
H1-9 are depicted in Figure 1.
H6. Perceived usefulness of an RA will positively affect
intention to reuse the RA. Methodology
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 19:35 31 January 2016 (PT)

H7. Enjoyment of using an RA will positively affect Procedure

intention to reuse the RA. Data for this study were collected through an independent
Internet research company. The study was set up fully online.
The moderating effect of financial risk Participants were given monetary incentives and participation
Past research suggests that consumer search behaviors and was completely voluntary. Participants had the freedom of
decision outcomes change as the financial risk involved in a completing the study at their own pace, at any time and from
purchase changes (Beatty and Smith, 1987; Dowling and anywhere they would like to as long as they had Internet
Staelin, 1994). As such, we incorporate the moderating effect access. First, participants were instructed to read a scenario
of financial risk on the relationships between consumer that asked them to search for product information and get
participation and perceived ease of use and enjoyment of using recommendations for a simulated purchase situation. The
an RA. purchase situation involved either high financial risk (buying a
When the level of financial risk involved in a purchase is laptop computer priced at $1000-1200) or low financial risk
high, consumers will likely find the time and effort that they (buying a digital camera priced at $80-120). Participants were
need to spend in interacting with an RA and providing randomly assigned to one purchase scenario and then
information to the RA to be worthwhile, reasoning that such instructed to use the RA on the Web site of
inputs will reduce the risk associated with the purchase. This [Link] to search for information and get
suggests that the negative impact of consumer participation on recommendations for the product in the scenario. On
perceived ease of use proposed in H6 will become weaker completion of the search task, participants were directed to fill
when the purchase is of high financial risk. But at the same out the online survey.
time, consumers will be more engaged in the decision making Using an existing RA and Web site makes the study setting
process as a result of the high financial risk involved in the a realistic online environment where participants conduct
purchase, which may, in turn, lead to greater feelings of fun information search and comparison tasks as they would
and pleasure during the use of an RA that helps evaluate normally do in their daily lives. Participants were also

Figure 1 Research model

Financial Risk Involved

in a Purchase

H8 (+) H9 (+)

Perceived Ease of
Use of the RA

H3 (+)

H1 (–)
Consumer Perceived
Participation in Using Usefulness of the H6 (+) Intention to Reuse
H4 (+)
an RA RA the RA

H2 (+) H7 (+)

H5 (+)

Enjoyment of Using
the RA

463
 Consumer participation in online product recommendation services Journal of Services Marketing
Xiaojing Sheng and Mohammadali Zolfagharian Volume 28 · Number 6 · 2014 · 460 –470

prescreened by asking them whether they had previously used that these items cross-loaded on unintended constructs. As a
the RA on [Link] to ensure that they had result, these eight items were dropped and a second run of
no prior experience with the Web site or the RA. CFA was conducted on the revised model. The results showed
significant improvement and an acceptable model fit: ␹2 ⫽
Sample characteristics 303.78, df ⫽ 160, ␹2/df ⫽ 1.89, GFI ⫽ 0.90, CFI ⫽ 0.97 and
The survey comprised 243 responses of which, 11 contained RMSEA ⫽ 0.06.
large amounts of missing data and were, therefore, removed To assess the reliability and validity of the constructs, factor
from further data analysis. The sample consisted of 45.7 per loadings of the measurement items, composite reliability of
cent female participants and 54.3 per cent male participants. each construct and average variance extracted (AVE) were
In terms of age distribution, 16.7 per cent were younger than examined. As Table I shows, all measurement items in the
24 years of age, 38.6 per cent were between the age of 25 and revised model correctly loaded on their intended constructs,
44 years, 33.1 per cent were between the age of 45 and 64 supporting convergent and discriminant validity of the
years and 11.6 per cent were aged 65 years and older; 68.2 per constructs. Moreover, item factor loadings ranged from 0.76
cent of the participants are Caucasian, with the rest being to 0.97, all ⬎ 0.4 as recommended by Hair et al. (1998) and
Asian (7.7 per cent), African-American (9.4 per cent), all significant at the 0.001 level, indicating high individual
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 19:35 31 January 2016 (PT)

Hispanic (6.9 per cent), Native American (2.6 per cent) and item reliability. Cronbach’s alpha value was 0.93 for perceived
unspecified ethnicities (5.2 per cent). Over half of the sample usefulness, 0.88 for perceived ease of use, 0.92 for perceived
(59.2 per cent) reported a combined household income enjoyment, 0.97 for intention, 0.91 for perceived participation
of ⬍$40,000, with 32.7 per cent between $50,000 and and 0.88 for perceived financial risk. These values were all
$110,000 and 8.1 per cent ⬎$110,000. In terms of education above the recommended threshold of 0.7 (Nunnally, 1978),
level, 41.6 per cent of the participants received at least college verifying high internal consistency of the measures.
degrees, 32.6 per cent had some college education and 23.2 Composite reliability was also calculated for each construct:
per cent graduated from high school with GED. The 0.94 for perceived usefulness, 0.91 for perceived ease of use,
participants are Internet savvy, with 97.8 per cent having used 0.94 for perceived enjoyment, 0.97 for intention, 0.91 for
the Internet for at least 10 years. perceived participation and 0.89 for perceived financial risk.
These values were all ⬎ 0.70, as recommended by Hair et al.
Measures (1998), providing further evidence of construct reliability.
Perceived usefulness of an RA was measured with four items AVE measures the variance captured by a construct in
adapted from Gentry and Calantone (2002) and from Wang relation to the variance due to random measurement error
and Benbasat (2005) to tap participants’ perceptions of the (Fornell and Larcker, 1981). AVE was.84 for perceived
extent to which the RA was effective and useful in helping usefulness, 0.78 for perceived ease of use, 0.79 for perceived
them accomplish the information search task and make enjoyment, 0.90 for intention, 0.78 for perceived participation
purchase decisions. Perceived ease of use was assessed with and 0.73 for perceived financial risk. These values were all
five items adapted from Gentry and Calantone (2002) and well above the commonly recommended threshold of 0.5
Wang and Benbasat (2005) to reflect participants’ perceptions (Bagozzi and Yi, 1988; Fornell and Larcker, 1981).
of the degree of ease or difficulty of using an RA. Five items Furthermore, the square root of each construct’s AVE shown
were used for measuring perceived enjoyment derived from in Table I was greater than the correlations of that construct
using the RA. These five items were based on the work by with other constructs, indicating that each construct shared
Davis et al. (1992) and Moon and Kim (2001). The level of more variance with its own measures than it shared with other
consumer participation was captured by four items measuring constructs. As such, all of the constructs demonstrate
participants’ perceptions of the level of effort and the amount discriminant validity.
of information, work and time that they contributed when
using the RA. These four items were adapted from Bendapudi
Hypothesis testing
and Leone (2003) and Fang (2008). The level of financial risk
SEM with AMOS 21 was used to test the hypotheses. Results
involved in a purchase was also captured by measuring
were summarized in Table II. The structural model had an
participants’ perceptions. The five items used for measuring
acceptable fit with the data: ␹2 ⫽ 232.67, df ⫽ 111, ␹2/df ⫽
perceived financial risk were adapted from Biswas and Biswas
2.09, GFI ⫽ 0.90, CFI ⫽ 0.97 and RMSEA ⫽ 0.07.
(2004). A 5-point Likert or Likert-type scale was used for all
As predicted, participation in using an RA was found to
the measures except the level of consumer participation, which
have a negative effect on perceived ease of use of the RA and
was measured on a 7-point scale with endpoints “very
a positive effect on enjoyment of using the RA. The
minimal” (1) and “quite a lot” (7).
standardized ␤ coefficient was ⫺0.35 (p ⬍ 0.001) for
perceived ease of use and 0.17 (p ⫽ 0.014) for enjoyment.
Analysis and results
Thus, both H1 and H2 were supported. The results also
Measure validity supported H3 and H4, indicating that perceived ease of use of
Confirmatory factor analysis (CFA) was conducted to assess an RA positively affected perceived usefulness (␤ ⫽ 0.15,
the validity of the measures using AMOS 21. Results from the p ⫽ 0.008) and enjoyment of using the RA (␤ ⫽ 0.44, p ⬍
initial run of CFA showed poor model fit indices: ␹2 ⫽ 0.001). Enjoyment was also shown to have a positive effect on
1456.99, df ⫽ 341, ␹2/df ⫽ 4.27, GFI ⫽ 0.61, CFI ⫽ 0.81 perceived usefulness (␤ ⫽ 0.72, p ⬍ 0.001), supporting H5.
and RMSEA ⫽ 0.12. An examination of the results revealed Finally, consistent with H6 and H7, intention to reuse an RA
that eight items had extremely high modification indices and was found to be positively affected by perceived usefulness of

464
 Consumer participation in online product recommendation services Journal of Services Marketing
Xiaojing Sheng and Mohammadali Zolfagharian Volume 28 · Number 6 · 2014 · 460 –470

Table I Construct attributes

Factor Composite Cronbach’s
Construct and items loadings reliability alpha 1 2 3 4 5 6
a
1. Intention to reuse the RA 0.97 0.97 0.95
I would use this agent to help with my future purchase
decisions 0.94
I would never use this agent againb –
I would recommend this agent to my friends 0.93
I would let this agent assist me in searching for product
information 0.95
I would use this agent as a guide for my product
purchases in the future 0.96

2. Perceived usefulness of the RA 0.94 0.93 0.81ⴱⴱ 0.92

Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 19:35 31 January 2016 (PT)

The agent enhanced my effectiveness in searching for

product information 0.90
The agent was not useful to me at all b –
I found the agent very helpful in accomplishing my task 0.92
I think this agent would be very useful in the future to
help me make purchase decisions 0.93

3. Enjoyment of using the RA 0.94 0.92 0.76ⴱⴱ 0.72ⴱⴱ 0.89

I found it interesting to use this agent 0.91
I had a lot of fun interacting with the agent 0.89
It was boring to use the agent b –
I very much enjoyed using the agent 0.94
The way the agent works was entertaining to me 0.80

4. Perceived ease of use of the RA 0.91 0.88 0.37ⴱⴱ 0.41ⴱⴱ 0.33ⴱⴱ 0.88
Using the agent was very difficult for me 0.84
The way the agent worked was hard to understand 0.94
I found it very easy to use the agent b –
Learning to use the agent was confusing 0.86
The working of the agent was simple and straightforwardb –

5. Consumer participation in using an RA 0.91 0.91 ⫺0.11 ⫺0.09 0.05 ⫺0.37ⴱⴱ 0.88
When using the agent, the amount of information I
provided was. . .b –
When using the agent, the level of effort I put in was. . . 0.88
When using the agent, the amount of work I did was. . . 0.97
The amount of time I spent in using the agent was. . . 0.79

6. Financial risk involved in a purchase 0.89 0.88 ⫺0.05 ⫺0.05 0.01 ⫺0.30ⴱⴱ 0.11 0.85
It would involve a great deal of financial risk 0.76
I would not be concerned about the money for this
purchase b –
This is something I can easily afford b –
It would be risky for me to buy the product at that price 0.91
This would certainly be a high-risk purchase for me 0.89
Notes: a Diagonal elements are square roots of the AVE; and off-diagonal elements are inter-construct correlations. b
Indicate that the items are
dropped due to cross-loadings. ** Indicate that the correlations are significant at the 0.01 level (two-tailed)

the RA (␤ ⫽ 0.59, p ⬍ 0.001) and by enjoyment of using the enjoyment. To test these two hypotheses, financial risk was
RA (␤ ⫽ 0.35, p ⬍ 0.001). As expected, participation in using divided into high and low groups using median split. First,
an RA did not have affect perceived usefulness of the RA participants in the high-financial risk group were compared to
(␤ ⫽ ⫺0.05, p ⫽ 0.111). those in the low-financial risk group to ascertain whether these
Regarding the moderating effect of financial risk, H8 and two groups differ in terms of age, income and education levels.
H9 predict that financial risk weakens the negative effect of Results from the independent samples t-test show that there is
perceived participation on perceived ease of use and no significant difference of age (t ⫽ 1.09, p ⫽ 0.28), income
strengthens the positive effect of perceived participation on (t ⫽ 0.06, p ⫽ 0.96) or education (t ⫽ 0.70, p ⫽ 0.48) between

465
 Consumer participation in online product recommendation services Journal of Services Marketing
Xiaojing Sheng and Mohammadali Zolfagharian Volume 28 · Number 6 · 2014 · 460 –470

Table II Summary of hypothesis testing results

Supported or not
Hypothesis Results supported
H1: Consumer participation in using an RA ¡ (ⴚ) Perceived ease of
use of the RA ␤ ⫽ ⫺0.35, p ⬍ 0.001 Supported
H2: Consumer participation in using an RA ¡ (ⴙ) Enjoyment of
using the RA ␤ ⫽ 0.17, p ⫽ 0.014 Supported
H3: Perceived ease of use of the RA ¡ (ⴙ) Perceived usefulness of
the RA ␤ ⫽ 0.15, p ⫽ 0.008 Supported
H4: Perceived ease of use of the RA ¡ (ⴙ) Enjoyment of using the
RA ␤ ⫽ 0.44, p ⬍ 0.001 Supported
H5: Enjoyment of using the RA ¡ (ⴙ) Perceived usefulness of the
RA ␤ ⫽ 0.72, p ⬍ 0.001 Supported
H6: Perceived usefulness of the RA ¡ (ⴙ) Intention to reuse ␤ ⫽ 0.59, p ⬍ 0.001 Supported
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 19:35 31 January 2016 (PT)

H7: Enjoyment of using the RA ¡ (ⴙ) Intention to reuse ␤ ⫽ 0.35, p ⬍ 0.001 Supported
H8: Financial risk involved in a purchase weakens the negative Consumer participation had a stronger negative Not supported
effect of consumer participation on perceived ease of use effect on perceived ease of use for the high-
financial risk group (␤ ⫽ ⫺0.43, p ⬍ 0.001)
than that for the low-financial risk group
(␤ ⫽ ⫺0.20, p ⫽ 0.047), indicating the
strengthening effect of financial risk
H9: Financial risk involved in a purchase strengthens the positive Consumer participation had a positive effect on Not supported
effect of consumer participation on enjoyment enjoyment for the high-financial risk group
(␤ ⫽ 0.14, p ⫽ 0.004) but had no effect on
enjoyment for the low-financial risk group
(␤ ⫽ 0.04, p ⫽ 0.42)

the high- and low-financial risk groups. Then the structural however, the difference was not significant (⌬␹2/⌬df ⫽ 1.68,
model with free parameter estimate (unconstrained model) p ⫽ 0.195). Therefore, H9 is not supported. Further
and the model with an equality constraint imposed on the path examination of the ␤ weights revealed that perceived
between perceived participation and perceived ease of use participation positively affected enjoyment only when financial
(constrained model) were tested simultaneously. Results show risk was high (␤ ⫽ 0.14, p ⫽ 0.004). This positive effect
a slight, yet significant (⌬␹2/⌬df ⫽ 7.37, p ⫽ 0.007) fit became insignificant for the low-financial risk group
superiority for the unconstrained model (␹2 ⫽ 507.09, df ⫽ (␤ ⫽ 0.04, p ⫽ 0.42).
230, ␹2/df ⫽ 2.20, GFI ⫽ 0.807, CFI ⫽ 0.932, RMSEA ⫽
0.072) relative to the constrained model (␹2 ⫽ 514.45, Post hoc analysis
df ⫽ 231, ␹2/df ⫽ 2.23, GFI ⫽ 0.805, CFI ⫽ 0.930, RMSEA Although there is no extant literature on the moderating
⫽ 0.073). This result underscores the significant difference effects of demographic variables and Internet experience on
between the high- and low-financial risk groups and provides the relationship between consumer participation and
support for the moderating effect of financial risk on the perceived ease of use as well as enjoyment, these variables have
relationship between perceived participation and perceived been tested or proposed to be included as moderators in
ease of use. The estimated standardized ␤ coefficient of the TAM-based models such as UTAUT (Venkatesh et al.,
effect of perceived participation on perceived ease of use 2003), UTAUT2 (Venkatesh et al., 2012) and consumer
was ⫺0.20 (p ⫽ 0.047) for the low-financial risk group and TAM (Baron et al., 2006). As such, additional analysis was
⫺0.43 (p ⬍ 0.001) for the high-financial risk group. The conducted to examine whether gender, age, income,
increased ␤ weight indicates that the negative effect of education and the Internet experience had any moderating
perceived participation on perceived ease of use strengthens effect on the relationship between consumer participation and
when the purchase decision involves a higher level of financial perceived ease of use and enjoyment of using an RA.
risk. Therefore, contrary to what H8 had predicted, the results The analysis followed the same procedure used in testing
showed that financial risk strengthens, rather than weakens, H8 and H9. Age, education, income and the Internet
the negative impact of perceived participation on perceived experience were first recoded into categorical variables.
ease of use of the RA. Specifically, for age, participants younger than 45 years were
The procedure described above was used in testing H9 grouped into one set (128 participants) and the remainder in
which proposes that financial risk strengthens the relationship the other set. Participants with high school or some college
between perceived participation and enjoyment of using the education were in one group (135 participants) and the
RA. The unconstrained model, in which the coefficient of the remainder in the other group. Likewise, participants with an
relationship between perceived participation and enjoyment annual household income ⬍$40,000 were in one group (111
was freely estimated, showed slightly better model fit than the participants) and the rest in the other group. As for the
constrained model with the equality constraint imposed; Internet experience, those with ⬎ 10 years of Internet usage

466
 Consumer participation in online product recommendation services Journal of Services Marketing
Xiaojing Sheng and Mohammadali Zolfagharian Volume 28 · Number 6 · 2014 · 460 –470

experience were in one group (129 participants) and the well supported demonstrates the nomological validity of the
remainder in the other group. Nested model comparisons TAM expanded here to incorporate consumer participation as
were then performed to examine whether the effects of an antecedent of its key constructs, i.e. perceived ease of use
consumer participation on perceived ease of use and and enjoyment. The finding that consumer participation
enjoyment differ by gender, age, education level, income level negatively affected perceived ease of use and positively
and Internet experience. The results show that of all the model affected enjoyment contributes to the TAM literature and
comparisons, two had significant differences between the answers the call to deepen TAM conceptualization through
unconstrained and constrained models. Specifically, the effect uncovering consumer participation as a construct that explains
of consumer participation on enjoyment differed by education the two salient beliefs associated with using RAs. Moreover,
levels and income levels. For those participants with high the findings that reuse intention were positively affected by
school or some college education, the effect was significant perceived usefulness and enjoyment and that perceived
(␤ ⫽ 0.15, p ⫽ 0.002). However, this effect became usefulness was positively affected by perceived ease of use and
insignificant for those who had a relatively higher education enjoyment which was, in turn, positively affected by perceived
level (␤ ⫽ 0.02, p ⫽ 0.758). The difference between the ease of use supports the causal chain of effects as predicted by
unconstrained and constrained model was significant (⌬␹2/ TAM, thus providing further evidence of its applicability in
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 19:35 31 January 2016 (PT)

⌬df ⫽ 3.88, p ⫽ 0.049). Likewise, the effect of consumer the RA use context. Although Gentry and Calantone (2002)
participation on enjoyment was significant for those and Wang and Benbasat (2005) utilized TAM in studying
participants in a relatively lower income bracket (␤ ⫽ 0.18, p consumer adoption of RAs, enjoyment of using RAs was
⬍ 0.001) but insignificant for those in a higher income bracket absent from their works. Therefore, this research also
(␤ ⫽ 0.02, p ⫽ 0.708). The difference between the contributes to the extant literature on RAs by studying the
unconstrained and constrained model was also significant effect of enjoyment as a third salient belief in addition to
(⌬␹2/⌬df ⫽ 5.48, p ⫽ 0.019). perceived usefulness and perceived ease of use.
Second, the dual-effect of consumer participation found in
Discussion this research (i.e. consumer participation had a positive effect
on enjoyment but a negative effect on perceived ease of use)
Online product RAs have become increasingly available on the
demonstrates that the positive effect of consumer participation
Internet and are gaining greater importance as an interactive
can be overshadowed by its negative effect. This finding
decision aid that aims to provide value-added services, and
contributes to the extant RA literature by shedding light on the
thus help consumers with their product information search
role of consumers as active participants in using RAs.
and purchase decision-making by making product
Moreover, the result that consumer participation had both
recommendations that match their preferences and interests.
positive and negative effects is consistent with previous
Recommendations made by an RA depend on not only the
research where consumer participation was shown to add
type of computer algorithms that the RA uses but also the
value through reducing the firms’ operational cost and
information that consumers provide to the RA in response to
enhancing customer satisfaction (Bowers et al., 1990;
its questions or inquiries during the preference elicitation
Dellande et al., 2004) and, at the same time, decreasing value
process. Although previous studies have examined the
through increasing employee job stress and role ambiguity
effectiveness of the different types of computer algorithms that
(Chan et al., 2010; Hsieh et al., 2004). This finding provides
underlie recommendations (Ansari et al., 2000; Ariely et al.,
additional empirical evidence for the notion that consumer
2004), little effort has been made to explore the consumer role
participation can be a “double-edged sword”. This finding
as an active participant in using RAs and in co-creating
also reveals the complex nature of consumer participation in
services with RAs. The current research fills this gap by
value co-creation and points out that overemphasizing the
empirically testing a theoretical model that integrates
merits of consumer participation without acknowledging its
consumer participation into the TAM to shed further light on
potential damaging effect is not warranted.
consumer beliefs and behavioral intentions with respect to
Finally, the finding that financial risk strengthens the
RAs. This research shows that consumer participation in using
negative effect of consumer participation on perceived ease of
an RA negatively affects perceived ease of use of the RA but
use and that the positive effect of consumer participation on
positively affects enjoyment of using the RA. Furthermore, the
enjoyment only exists for those participants with relatively
negative impact of consumer participation on perceived ease
lower education and income supports the moderating role that
of use was shown to become even stronger when the purchase
financial risk, education and income play, and qualifies as
decision involves a higher level of financial risk. Finally, this
another contribution by setting boundary conditions on the
research provides further support for the multiple predictions
main effects of consumer participation on perceived ease of
of TAM regarding the relationships between intention to reuse
use and enjoyment.
an RA and perceived usefulness, perceived ease of use and
enjoyment of using the RA as well as those among perceived
usefulness, perceived ease of use and enjoyment. Managerial implications
The findings that perceived usefulness and enjoyment
Theoretical contributions positively affected intention and that enjoyment had both a
This research makes multiple theoretical contributions. First, direct and an indirect positive effect through perceived
it extends the concept of consumer participation to the online usefulness on intention speaks to the importance of perceived
context of using RAs and tests its effect within the TAM usefulness and enjoyment as two determinants of consumer
framework. The fact that the proposed theoretical model was intentions to reuse an RA. Perceived ease of use did not

467
 Consumer participation in online product recommendation services Journal of Services Marketing
Xiaojing Sheng and Mohammadali Zolfagharian Volume 28 · Number 6 · 2014 · 460 –470

impact intentions directly. Instead, the effect of perceived ease the study could have influenced the results. To overcome this
of use was channeled through perceived usefulness and limitation, future research can replicate the study within a
enjoyment, which further attests to the criticality of perceived controlled lab setting so that the influence of extraneous
usefulness and enjoyment in determining whether a consumer variables can be reduced to minimum.
will return and reuse an RA. Taken together, these findings Another potential limitation is the simulated purchase
suggest that the design of an RA needs to focus on enhancing situation. Because participants did not carry out the purchase,
the usefulness and enjoyment that consumers attribute to their perceptions of the financial risk might not be as vivid as
using the RA. This by no means suggests that perceived ease when they actually pay and complete the purchase. This, in
of use can be ignored in the design of an RA. Rather, making fact, might be a possible explanation as to why the moderating
sure that an RA is easy to use and free of effort should be the effect of financial risk on the relationship between consumer
starting point because perceived ease of use was shown to participation and enjoyment only found support in the
improve perceived usefulness and enjoyment of using the RA. high-risk group. Therefore, future research could use a study
The positive effect of consumer participation on enjoyment design in which participants are given credits that allow them
suggests that consumers enjoy interacting with and using an to actually complete the transaction to see whether the
RA more so when they perceive to have highly participated in proposed model still holds.
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 19:35 31 January 2016 (PT)

using the RA. On the other hand, the greater a consumer Because only self-reported behavioral intentions were
participates in using an RA, the lower the perceived ease of the captured in this study, future research could use a longitudinal
use of the RA because more participation requires more time design and collect actual usage data to verify whether
and effort from the consumer. The dual-effect of consumer intentions lead to actual behaviors. Future research could also
participation indicates the need to balance the level of study other possible moderating variables such as product type
participation required when using an RA. Allowing more (i.e. search product, experience product and credence
room for consumers to participate and to be more involved in product) to see whether those variables have any impact on the
using an RA by asking more relevant yet interesting questions relationships between consumer participation and perceived
about their product preferences can promote the fun and ease of use as well as enjoyment of using RAs. Finally, future
enjoyment aspect of using the RA. At the same time, research could include other perspectives such as consumer
consumers should not be overburdened with too many resource integration and utilize interpretive research methods
questions or too much interaction, especially when such such as in-depth interviews and diary writing in examining
questions and interactions lie beyond consumer expectations. consumers’ RA usage behavior. Such research could delineate
The strengthening effect of financial risk on the negative the knowledge and skills that would enable consumers to
impact of consumer participation on perceived ease of use create a positive experience with RA and how those knowledge
suggests that flexibility and adaptability need to be and skills can be obtained.
incorporated in the design of RAs. In other words, RAs should
be designed in such a way that the user interface of an RA can References
change in response to the level of financial risk involved in a
Ansari, A., Essegaier, S. and Kohli, R. (2000), “Internet
purchase. When financial risk is high, use of the RA should
recommendation systems”, Journal of Marketing Research,
require even less effort from consumers to avoid escalating the
Vol. 37 No. 3, pp. 363-375.
negative impact of consumer participation on perceived ease.
Ariely, D., Lynch, J.G. and Aparicio, M. (2004), “Learning by
Furthermore, the finding that consumer participation
collaborative and individual-based recommendation
positively affected enjoyment only for those participants with
agents”, Journal of Consumer Psychology, Vol. 14 Nos 1/2,
relatively lower education and income highlights the
pp. 81-95.
importance of using education and income as two
Bagozzi, R.P. (2007), “The legacy of the technology
demographic variables to segment RA users and consumers.
acceptance model and a proposal for a paradigm shift”,
The implication is that marketers and RA developers need to
Journal of the Association for Information Systems, Vol. 8
promote the fun and enjoyment aspects associated with using
No. 4, pp. 244-254.
an RA for those consumers who have relatively lower
Bagozzi, R.P. and Yi, Y. (1988), “On the evaluation of
education and income levels to enhance the positive effect of
structural equation models”, Journal of the Academy of
consumer participation on enjoyment, which will ultimately
Marketing Science, Vol. 16 No. 1, pp. 74-94.
lead to greater intentions to reuse the RA.
Baron, S., Patterson, A. and Harris, K. (2006), “Beyond
technology acceptance: understanding consumer practice”,
Limitation and future research International Journal of Service Industry Management, Vol. 17
One potential limitation is the uncontrolled environment No. 2, pp. 111-135.
within which the study was conducted. As discussed in the Bateson, J.E.G. (1985), “Self-service consumer: an
Methodology section, participants had the freedom of exploratory study”, Journal of Retailing, Vol. 61 No. 3,
choosing when and where to complete the study. Although pp. 49-76.
this provides the benefit of making the study environment Beatty, S.E. and Smith, S.M. (1987), “External search effort:
realistic and staying true to what consumers would normally an investigation across several product categories”, Journal
do in their day-to-day lives, it is possible that, due to the of Consumer Research, Vol. 14 No. 1, pp. 83-95.
uncontrolled nature of the study environment, variables Bendapudi, N. and Leone, R.P. (2003), “Psychological
outside of the current study such as Internet connection speed implications of customer participation in co-production”,
and the type of the computer that participants use to complete Journal of Marketing, Vol. 67 No. 1, pp. 14-28.

468
 Consumer participation in online product recommendation services Journal of Services Marketing
Xiaojing Sheng and Mohammadali Zolfagharian Volume 28 · Number 6 · 2014 · 460 –470

Biswas, D. and Biswas, A. (2004), “The diagnostic role of Häubl, G. and Trifts, V. (2000), “Consumer decision making
signals in the context of perceived risks in online shopping: in online shopping environment: the effects of interactive
do signals matter more on the web?”, Journal of Interactive decision aids”, Marketing Science, Vol. 19 No. 1, pp. 4-21.
Marketing, Vol. 18 No. 3, pp. 30-45. Hilton, T., Hughes, T., Little, E. and Marandi, E. (2013),
Bitner, M.J., Faranda, W.T., Hubbert, A.R. and “Adopting self-service technology to do more with less”,
Zeithaml, V.A. (1997), “Customer contributions and roles Journal of Services Marketing, Vol. 27 No. 1, pp. 3-12.
in service delivery”, International Journal of Service Industry Hsieh, A.-T., Yen, C.-H. and Chin, K.-C. (2004),
Management, Vol. 8 No. 3, pp. 193-205. “Participative customers as partial employees and service
Bowers, M.R., Martin, C.L. and Luker, A. (1990), “Trading provider workload”, International Journal of Service Industry
places: employees as customers, customers as employees”, Management, Vol. 15 No. 2, pp. 187-199.
Journal of Services Marketing, Vol. 4 No. 2, pp. 55-69. Kim, J. and Forsythe, S. (2008), “Adoption of virtual try-on
Cermak, D.S.P., File, K.M. and Prince, R.A. (1994), technology for online apparel shopping”, Journal of
“Customer participation in service specification and Interactive Marketing, Vol. 22 No. 2, pp. 45-59.
delivery”, Journal of Applied Business Research, Vol. 10 Lee, Y., Kozar, K.A. and Larsen, K.R.T. (2003), “The
No. 2, pp. 90-100. technology acceptance model: past, present, and future”,
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 19:35 31 January 2016 (PT)

Chan, K.W., Yim, C.K. and Lam, S.S.K. (2010), “Is Communications of the Association for Information Systems,
customer participation in value creation a double-edged Vol. 12 No. 1, pp. 752-780.
sword?, evidence from professional financial services across Lovelock, C.H. and Young, R.F. (1979), “Look to consumers
cultures”, Journal of Marketing, Vol. 74 No. 3, pp. 48-64. to increase productivity”, Harvard Business Review, Vol. 57
Chang, C.-C., Chen, H.-Y. and Huang, I.-C. (2009), “The No. 3, pp. 168-178.
interplay between customer participation and difficulty of Mick, D.G. and Fournier, S. (1998), “Paradoxes of
design examples in the online designing process and its technology: consumer cognizance, emotions, and coping
effect on customer satisfaction: mediational analyses”, strategies”, Journal of Consumer Research, Vol. 25 No. 2,
Cyberpsychology and Behavior, Vol. 12 No. 2, pp. 147-154. pp. 123-143.
Chun, H., Lee, H. and Kim, D. (2012), “The integrated Mills, P.K., Chase, R.B. and Margulies, N. (1983),
model of smartphone adoption: hedonic and utilitarian “Motivating the client/employee system as a service
value perceptions of smartphones among Korean college production strategy”, The Academy of Management Review,
students”, Cyberpsychology, Behavior, and Social Networking, Vol. 8 No. 2, pp. 301-310.
Vol. 15 No. 9, pp. 473-479. Moon, J.-W. and Kim, Y.-G. (2001), “Extending the TAM
Davis, F.D. (1989), “Perceived usefulness, perceived ease of for a world-wide-web context”, Information and
use, and user acceptance of information technology”, MIS Management, Vol. 38 No. 4, pp. 217-230.
Quarterly, Vol. 13 No. 3, pp. 319-340. Nunnally, J.C. (1978), Psychometric Theory, 2nd ed.,
Davis, F.D., Bagozzi, R.P. and Warshaw, P.R. (1989), “User McGraw-Hill, New York, NY.
acceptance of computer technology: a comparison of two Pijpers, G.G.M. and van Montfort, K. (2005), “An
theoretical models”, Management Science, Vol. 35 No. 8, investigation of factors that influence senior executives to
pp. 982-1003. accept innovations in information technology”, International
Davis, F.D., Bagozzi, R.P. and Warshaw, P.R. (1992), Journal of Management, Vol. 22 No. 4, pp. 542-555.
“Extrinsic and intrinsic motivation to use computers in the Todd, P. and Benbasat, I. (1992), “The use of information in
workplace”, Journal of Applied Social Psychology, Vol. 22 decision making: an experimental investigation of the
No. 14, pp. 1111-1132. impact of computer-based decision aids”, MIS Quarterly,
Dellande, S., Gilly, M.C. and Graham, J.L. (2004), “Gaining Vol. 16 No. 3, pp. 373-393.
compliance and losing weight: the role of the service Todd, P. and Benbasat, I. (2000), “Inducing compensatory
provider in health care services”, Journal of Marketing, information processing through decision aids that facilitate
Vol. 68 No. 3, pp. 78-91. effort reduction: an experimental assessment”, Journal of
Dowling, G.R. and Staelin, R. (1994), “A model of perceived Behavioral Decision Making, Vol. 13 No. 1, pp. 91-106.
risk and intended risk-handling activity”, Journal of Venkatesh, V., Morris, M.G., Davis, G.B. and Davis, F.D.
Consumer Research, Vol. 21 No. 1, pp. 119-134. (2003), “User acceptance of information technology:
Fang, E. (2008), “Customer participation and the trade-off toward a unified view”, MIS Quarterly, Vol. 27 No. 3,
between new product innovativeness and speed to market”, pp. 425-478.
Journal of Marketing, Vol. 72 No. 4, pp. 90-104. Venkatesh, V., Thong, J.Y.L. and Xu, X. (2012), “Consumer
Fornell, C. and Larcker, D.F. (1981), “Evaluating structural acceptance and use of information technology: extending
equation models with unobservable variables and the unified theory of acceptance and use of technology”,
measurement error”, Journal of Marketing Research, Vol. 18 MIS Quarterly, Vol. 36 No. 1, pp. 157-178.
No. 1, pp. 39-50. Wang, W. and Benbasat, I. (2005), “Trust in and adoption of
Gentry, L. and Calantone, R. (2002), “A comparison of three online recommendation agents”, Journal of the Association
models to explain shop-bot use on the web”, Psychology and for Information Systems, Vol. 6 No. 3, pp. 72-101.
Marketing, Vol. 19 No. 11, pp. 945-956. Wang, W. and Benbasat, I. (2009), “Interactive decision aids
Hair, J.F., Anderson, R.E., Tatham, R.L. and Black, W.C. for consumer decision making in e-commerce: the influence
(1998), Multivariate Data Analysis, 4th ed., Prentice-Hall, of perceived strategy restrictiveness”, MIS Quarterly,
Englewood Cliffs, NJ. Vol. 33 No. 2, pp. 293-320.

469
 Consumer participation in online product recommendation services Journal of Services Marketing
Xiaojing Sheng and Mohammadali Zolfagharian Volume 28 · Number 6 · 2014 · 460 –470

Xiao, B. and Benbasat, I. (2007), “E-commerce product Quarterly, Journal of Consumer Satisfaction, Dissatisfaction and
recommendation agents: use, characteristics, and impact”, Complaining Behavior, Health Marketing Quarterly and
MIS Quarterly, Vol. 31 No. 1, pp. 137-209. Cyberpsychology, Behavior, and Social Networking.
Xu, J.J., Benbasat, I. and Cenfetelli, R.T. (2013), “Integrating
Mohammadali Zolfagharian is the Chair and Assistant
service quality with system and information quality: an
Professor of Marketing and an Applied Anthropologist at the
empirical test in the e-service context”, MIS Quarterly,
Department of Marketing, University of Texas-Pan American,
Vol. 37 No. 3, pp. 777-794.
Edinburg, Texas. His research interests include consumer
(culture) behavior, services and social/macro marketing using
About the authors
both quantitative and qualitative (i.e. ethnographic)
Xiaojing Sheng is an Assistant Professor of Marketing at the approaches. His research is published in several books and
University of Texas-Pan American. Her research interests journals such as Decision Sciences Journal, Journal of Services
include consumer behavior within the e-commerce context, Marketing, Journal of Business & Industrial Marketing, Journal
consumer adoption of technology-based products and services, of Consumer Marketing and Services Marketing Quarterly. He
attitude formation and change, services marketing, Internet has collaborated with or provided consulting to several
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 19:35 31 January 2016 (PT)

marketing, health marketing and cross-cultural research. Her organizations including Motorola, Hospice, The
research has been published or accepted for publication in Entrepreneur Authority and City of McAllen. Mohammadali
Journal of Business Research, Journal of Marketing Theory and Zolfagharian is a corresponding author and can be contacted
Practice, the Service Industries Journal, Services Marketing at: zolfagharian@[Link]

To purchase reprints of this article please e-mail: reprints@[Link]

Or visit our web site for further details: [Link]/reprints

470
 This article has been cited by:

1. Francisco Liébana-Cabanillas, Iviane Ramos de Luna, Francisco J. Montoro-Ríos. 2015. User behaviour in QR mobile payment
system: the QR Payment Acceptance Model. Technology Analysis & Strategic Management 27, 1031-1049. [CrossRef]
2. Arun Kumar Kaushik, Zillur Rahman. 2015. An alternative model of self-service retail technology adoption. Journal of Services
Marketing 29:5, 406-420. [Abstract] [Full Text] [PDF]
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 19:35 31 January 2016 (PT)