Perceived Interactivity leading to E-loyalty: Development
of a Model for Cognitive-Affective User Responses
Running Headline:
Perceived Interactivity leading to E-loyalty
Forthcoming in International Journal of Human Computer Studies
Dianne Cyr
(corresponding author)
Faculty of Business
Simon Fraser University
15th Floor, Central City Tower
13450 102nd Avenue
Surrey, BC Canada V3T 5X3
Tel: 1.604.552.9504
Fax: 1.604.268.7485
cyr@sfu.ca
Milena Head
DeGroote School of Business
McMaster University
Hamilton, Ontario, Canada
headm@mcmaster.ca
Alex Ivanov
School of Interactive Arts and Technology
Simon Fraser University
Surrey, BC Canada
aivanov@sfu.ca
2009
2
Perceived Interactivity leading to E-loyalty: Development
of a Model for Cognitive-Affective User Responses
AB S TRA CT
1
Novel applications of website interactivity are important to attract and
retain online users. In this empirical study five designs for interactivity are examined using
different web-poll interfaces. The goal of the investigation is to examine perceived interactivity
in a model which includes most commonly tested cognitive elements such as efficiency and
effectiveness, but augments this model with the inclusion of a cognitive-affective element for
trust, and an affective element of enjoyment. More specifically, a model is created to validate
the relationship of perceived interactivity (comprised of user control, user connectedness, and
responsiveness of the web-poll application) to efficiency, effectiveness, trust and enjoyment, of
the website. In turn, efficiency, effectiveness, trust, and enjoyment are tested for their influence
on user behavioral intentions for e-loyalty. All relationships in the model are supported. In
addition, exploratory evaluation of qualitative comments is conducted to investigate additional
insights between the five web poll treatments in this investigation. The research confirms the
complexity of a model in which cognitive, cognitive-affective and affective elements are
present, and advances knowledge on the consequences of perceived interactivity. In
additional to theoretical advancements, the research has merit for web designers and online
marketers regarding how to enhance interactive online web applications.
KEY WO RDS perceived interactivity, e-loyalty, web-poll design
1. IN TRO DUC TION
Despite the potential for interactivity provided by the Internet, little attention has been paid to
how interactivity might be more fully utilized (Johnson et al., 2006). Rice (1984) defined
interactivity as the capability of a computer-enabled communication system that permits
exchange of roles between the sender and receiver in real or delayed time so that
communicators have more control over the structure, pace, and content of the communication.
Although interactivity has numerous dimensions, a common theme is that the website
successfully provides information to the user, is perceived as responsive, and allows a sense
of connection – often with other users. While previous research has aimed to conceptually
unravel contributing factors to interactivity, there is relatively little empirical work that
systematically examines the consequences of interactivity. There are some exceptions. Jiang
1
An earlier version of this paper was accepted for presentation at the Sixth Pre-ICIS HCI Research in MIS
Workshop (HCI/MIS’07), Montreal, 2007. The authors sincerely appreciate the suggestions of three anonymous
reviewers.
3
and Benbasat (2007) examined interactivity related to purchase intention and intention to
return to the website. Johnson et al. (2006) tested the relationship of perceived interactivity to
attitude toward the website and involvement. Lee (2005) examined various components of
interactivity related to trust. Chen and Yen (2004) sought to determine elements of interactivity
that result in website quality. Finally, Teo et al. (2003) tested a model in which interactivity
resulted in user satisfaction, effectiveness, and efficiency of websites.
In the current investigation we aim to test user perceived interactivity using a web-poll
design. Web-polls are meant to solicit quick input/opinions from a web user, which is typically
displayed for viewing by other visitors to the site. Simple versions of web-polls are found on
various websites with static indicators such as star-ratings, bar graphs or pie charts
representing how many people make a choice for a particular product or service. Advanced
web-poll designs can have a more interactive element. For example, Ivanov et al. (2006)
developed a platform on a website for an herbal antidepressant that allowed users to know
through an ‘at-a-glance’ X-Y plot diagram how many other users found the product useful, at
what dosages, and for how long. Users could click or ‘plot’ their current mood onto this
interactive canvas, which was essentially a collaborative visualization populated by members
of the community. The experience of mapping or externalizing one’s subjective experience into
a public space, albeit anonymously, was meant to instill a feeling of contribution, a sense of
control, and connection to others on the site. For additional information on data representation
and web-polls refer to Appendix A.
In a web-poll context, it is expected that user perceptions of interactivity will be represented
in a model that includes both cognitive and affective components. As in the example of the
mood web-poll noted above, not only will users provide cognitive attributions to the website as
to whether or not it is perceived as efficient and effective, but they will also ideally feel an
affective connection to the site and a sense of enjoyment from the visit. Related to this
cognitive-affective duality of experience, much previous research on IT adoption has primarily
a cognitive orientation (Venkatesh et al., 2003). With specific reference to the popular
technology acceptance model (Davis, 1989), emphasis has been on constructs such as
perceived usefulness or perceived ease of use. More recently, research into website design
has recognized the importance of the inclusion of affective elements such as enjoyment (Cyr
et al. 2007; Cyr and Head, 2008; Kim et al., 2007; Sun and Zhang, 2006; Tractinsky, 2004;
Zhang and Li, 2004). If either or both cognitive or affective components are present then users
are more likely to return to the site or to visit it in the future, termed e-loyalty (Cyr at al. 2007;
Flavián et al., 2006; Lam et al., 2004).
Related to the preceding, a primary aim of this investigation is to create a model for
perceived interactivity that includes both cognitive as well as affective outcomes, with
4
subsequent influence on loyalty. In alignment with Komiak and Benbasat (2006) our
overarching theoretical model is derived from the theory of reasoned action (Fishbein and
Ajzen, 1975) in which attitudes influence behavioral intention. More specifically, in the
development of our research model we draw on work by Lee (2005), who examined
interactivity as comprised of various components such as user control, user connectedness,
and responsiveness to the user of an application – now applied in a web-poll context.
Perceived interactivity is expected to affect cognitive constructs such as efficiency and
effectiveness (as previously considered by Teo et al., 2003). In addition, Komiak and Benbasat
(2007) examined adoption of recommendation agents and found that trust in this context has
both cognitive and affective elements. In our model we include trust since this represents a
mid-range position between purely cognitive and purely affective constructs. Finally, we are
interested to examine perceived interactivity related to enjoyment which has been considered
an affective component in numerous studies (Childers et al., 2001; Cyr et al., 2007; van der
Heijden, 2003).
In sum, the primary goal of this investigation is to validate a model in which both user
cognitive and affective responses are examined with respect to interactivity. Specifically,
perceived interactivity (modeled as a formative construct that includes user control,
connectedness and responsiveness) is proposed to influence cognitive perceptions (for
efficiency and effectiveness), cognitive-affective perceptions (for trust), and affective
perceptions (for enjoyment). In turn, we validate whether efficiency, effectiveness, enjoyment
and trust influence e-loyalty in a web-poll context. A secondary goal is an exploration of five
treatments of web-poll designs that vary in complexity and design. The purpose of
incorporating different formats of web-poll design is to gain understanding of design elements
and any additional insights as they complement our quantitative analysis. Of interest is user
experience of each treatment, not only as it contributes to validation of our model, but also
concerning design applications and their utility in practice. The examination of various web-poll
designs constitutes an investigation of how the IT artifact influences salient beliefs about a
website. Benbasat and Barki (2007) note the importance of considering how IT artifacts
influence such beliefs, and the need to consider other salient beliefs besides perceived
usefulness and perceived ease of use as offered in the current research.
This paper begins with an outline of our theoretical framework and research model.
Hypotheses are next developed as derived from previous work in the area of perceived
interactivity and related fields, and results of this investigation are presented. The paper
concludes with a discussion of the findings for both academics and practitioners.
5
2. T HEO RET I CAL FOU NDA TION S AN D RE SEA RCH MO DEL
The theoretical framework for this investigation draws from the theory of reasoned action
(TRA) in which attitudes influence behavioral intention (Fishbein and Ajzen, 1975). TRA has
been extensively used by IS and other researchers to explain IT adoption (i.e. Davis et al.
1989; McKnight et al. 2002; Venkatesh et al. 2003). Based on TRA, an individual’s behavior is
predicted by his or her intention to perform this behavior. The theory specifies that intention is
influenced by attitudes toward the behavior as well as subjective norms as to whether others
who are valued believe the behavior should be performed. As argued by Komiak and
Benbasat (2006), when a behavior is voluntary (versus mandatory) and when first-hand
experience of a website is available, then it is sufficient to focus on attitude and to omit
subjective norms. This is consistent with other research (i.e. Gefen et al. 2003), and the
technology acceptance model (TAM) in which intention to accept or use a new technology is
determined by perceived usefulness and perceived ease of use of the technology. In the
context of the current investigation, perceived interactivity results in user reactions toward the
website (such as efficiency, effectiveness, enjoyment, or trust), which in turn influence
behavioral intention (e-loyalty).
Further, Ajzen (2001) suggests “a multi-component view of attitude and assumes that
evaluations are influenced by cognition as well as affect” (p. 34). In alignment with this view,
the original TAM model with a utilitarian emphasis has been augmented to include a “hedonic”
component (Childers et al., 2001; Cyr et al. 2007; Cyr and Head, 2008). In previous research
this affective or hedonic element is often referred to as enjoyment (van der Heijden, 2003).
Further, beliefs such as trust have both cognitive as well as affective characteristics (Komiak
and Benbasat (2003).
To investigate perceived user interactivity in the specific context of a web-poll, a model for
E-loyalty is presented in Figure 1. In a mobile commerce context, Lee (2005) proposed and
tested that user control, connectedness, responsiveness, and personalization are elements of
interactivity that result in trust and ultimately in intention to use a technology. The first three
elements were significant, while personalization was not. In the current study, we build on this
earlier work to examine user control, connectedness and responsiveness as components of
interactivity, now tested in a stationary commerce web-poll setting.
Relationships of perceived interactivity to efficiency and effectiveness are based on the
work of Teo et al. (2003) and as noted above represent cognitive elements of the model. We
add to these cognitively-based concepts the cognitive-affective element of trust (as per
Komiak and Benbasat, 2006), and the affective construct of enjoyment (Cyr et al. 2007) as
outcomes of perceived interactivity. The exogenous variable in our model is loyalty, which is
