Está en la página 1de 7

Journal of Business Research 63 (2010) 950956

Contents lists available at ScienceDirect

Journal of Business Research

Has e-marketing come of age? Modeling historical inuences on post-adoption era


Internet consumer behaviors
David G. Taylor, David Strutton
University of North Texas, Denton, TX 76203, United States

a r t i c l e i n f o a b s t r a c t

Article history: E-marketing evolved from a comparatively isolated group of dot-com rms to a mainstream marketing
Received 1 May 2008 channel activity. This progress has also inuenced customers and their e-behavior in the process. This
Received in revised form 1 September 2008 study reviews the literature from information systems and marketing and uses meta-analysis to synthesize
Accepted 1 January 2009
key ndings. Three important insights emerged. From this exercise, a framework was developed integrating
both perspectives, Marketing and IS empirical and conceptual literatures, enhancing the understanding of
Keywords:
Internet
the customer e-behavior process by which attitudes, perceptions and evaluations interact to inuence
Consumer behavior purchasing intentions. The analysis suggests also that perceived usefulness and perceived ease of use, the
Meta-analysis primary constructs of the technology acceptance model (TAM), continue to inuence purchasing intentions
in the post-adoption online context. Thus, the multi-disciplinary nature of online purchasing behavior was
demonstrated, underscoring a need for both cross-disciplinary research and a more integrative and
collaborative strategy for decision-makers and managers.
Published by Elsevier Inc.

For all the hoopla surrounding the Internet, nding your way spread from Silicon Valley to Main Street, Internet marketing was
onto the information superhighway remains surprisingly difcult. absorbed into marketing's mainstream.
And the sluggish performance may leave you wondering what Academic study of the Internet follows a similar pattern. In the
the excitement is about. Business Week, 1995 early 1990s, information systems (IS) articles on Internet topics began
to appear in publications such as MIS Quarterly, Information Systems
Research, Information & Management, Information Technology and
Management and others. Studies examine technological aspects of the
1. Introduction Internet, including user adoption, vendor attributes, Web site design,
platform performance and stability.
Since leaping into the national consciousness in the mid-1990s, In the mid-1990s, leading marketing journals also began publishing
the Internet has walked a tightrope between hype versus reality, prescient, forward-looking articles about the Internet's transformational
frustration versus promise, and potential versus pipe-dream. In its potential as a driver of marketing activities. Among the rst are Hoffman
early years, the Internet captured the popular imagination but and Novak's (1996) hypermedia computer-mediated communication
remained solidly in the realm of those technophiles who possessed model, Alba and Lynch's (1997) work on incentives to go online and
the ability to understand and navigate this brave new world. As Bu- Burke's (1997) predictions about the future of online consumer
siness Week pointed out, in 1995 the Internet was no place for the marketing. A revolution is predicted in marketing thought and strategy
neophyte, requiring high levels of technological savvy and patience to (Sharma and Sheth, 2004), and analogies are drawn between the
navigate with slow, unstable connections (Wildstrom, 1995). potential of the Internet and the impact of the printing press and
But as technology improved Internet usability, acceptance grew democratic governance (Dickson, 2000).
exponentially. Early e-retailers blazing the way were dot-coms, For these foundational marketing theorists, the sky is the limit for
discrete entities unfettered by brick-and-mortar stores. Internet the Internet's future impact. Many fellow academicians apparently
marketers were viewed as part of the e-commerce world, distinct agree, and scores of articles address various aspects of the Internet.
from their Old Economy brethren. However, just as Internet usage While the pioneers generally proposed conceptual models for
studying the Internet and discuss its marketing potential, such topics
Corresponding author. Department of Marketing, University of North Texas,
quickly gave way to more diverse research streams addressing
Denton, TX 76203, United States. mainstream topics such as pricing, strategy, nancial/legal issues or
E-mail address: Strutton@unt.edu (D. Strutton). channels, with articles appearing routinely in leading marketing

0148-2963/$ see front matter. Published by Elsevier Inc.


doi:10.1016/j.jbusres.2009.01.018
D.G. Taylor, D. Strutton / Journal of Business Research 63 (2010) 950956 951

journals. Several journals dedicated exclusively to e-commerce and extract the constructs most relevant to Internet-based marketing
Internet marketing emerged, including The Journal of Interactive research and practice. The following questions emerge:
Marketing, International Journal of Electronic Commerce, Electronic
In the context of Internet purchasing behaviors, what consumer
Commerce Research, and the Journal of Electronic Commerce Research.
attributes exercise the strongest effect on consumer behavioral
The IS literature concurrently conducted research on issues of
intention?
interest to marketers, such as purchasing behavior, channel preferences
What marketing-related insights about Internet purchasing behaviors
or customer satisfaction. All the while, the IS literature represents a
may be gleaned from the information systems literature?
parallel stream from which marketers can and do pluck a wealth of
What is an appropriate conceptual model for studying Internet
information regarding Internet marketing.
purchasing behavior in this, a purportedly, post-Internet-adoption
era?
2. Research objectives

3. Methodology
Customer acceptance and usage of the Internet has grown tremen-
dously in recent years, and so has its strategic application. The Internet
3.1. Establishing a baseline of knowledge
has emerged as a critical channel across marketing arenas, offering cost
efciencies, expanded geographic scope and new opportunities for
Between 1996 and 2007, A+ level marketing journals published
targeting or understanding customers. Understanding how, when or
72 articles on various subjects related to the Internet. Of these, 25 are
why consumers choose to transact online is as important now as it has
in the domain of online consumer behavior. These issues include
ever been. And that importance will likely grow over time. Indeed,
information-seeking, decision-making, privacy and trust and even the
Internet marketing efforts may well have already come of age. For most
effects of avatars or computer-generated graphic representations of
B2C rms, Internet marketing capabilities have passed the point where
human characters. A detailed review that summarizes authors, titles,
they represent mere nice-to-haves. Instead, such capabilities now
journals and key ndings is available upon request.
represent absolute musts.
The remaining 57 top-tier marketing articles are classied into one
Consequently, this study is driven by three objectives. They are to:
of ve additional categories: (Internet-related) marketing strategy,
nancial/legal, pricing, distribution or word-of-mouth behaviors.
Provide comprehensive reviews of studies of online consumer
behaviors published in leading marketing and IS journals during the
3.2. Developing the population of relevant articles
past decade.
Quantitatively synthesize and evaluate these new historical ndings
To integrate the extant literature, a two-pronged approach was
to secure theoretical and practical insights into how post-adoption
used. First, a review of marketing literature (excluding the A+ level
online consumer behaviors might be studied more effectively in the
journals) as well as management and interdisciplinary journals
future.
was conducted. This search uncovered numerous additional studies
Combine marketing and IS insights into a unifying Internet
relevant to this meta-analysis. Seven articles were used from
marketing framework; i.e., a model capable of explaining with
publications such as Journal of Business Research, Journal of Interactive
greater clarity the inuence of various interdisciplinary concepts on
Marketing, Managing Service Quality, Organizational Science, Psychology
the behavioral intentions of consumers who shop online.
& Marketing and Total Quality Management & Business Excellence. One
unpublished study, in press at Electronic Commerce Research, was also
2.1. Practical and theoretical signicance of this research included.
Second, an exhaustive search was conducted of the IS literature,
These objectives are timely for several reasons. First, evidence identifying the remaining studies included in the meta-analysis. For
suggests Internet usage and online purchasing behaviors have each inclusion, IS articles had to meet all of the following criteria:
reached a post-adoption stage at which a new framework is appropriate.
Published between 1996 and 2007;
Online sales were expected to top $116 billion during 2007 in anyone's
Contain empirical data generated from a online survey or experiment;
ledger, a material number. But sector studies also suggest the growth
Specify online behavioral intentions (e.g. loyalty or purchasing) as a
rate for consumer Internet adoption has already peaked (Richtel and
dependent measure;
Tedeschi, 2007). As such, behavioral models used previously to predict
Include at least one marketing-oriented predictor construct.
online consumer behavior models largely based on technology
acceptance and diffusion of innovation theories may no longer
3.3. Additional renement and execution
adequately capture Internet consumer behaviors.
Secondly, with stable, easily usable and accessible Web sites merely
The literature review included 112 articles. Of these, 25 met all
the price of admission to the Internet marketing environment,
criteria and were ultimately included in the quantitative analysis.
technology-related issues may be fading into the Internet background.
Each of these 25 studies is based on one of the following theories:
Today, all but the most luddite consumer shops online. The focus of
Internet research, therefore, should shift from the Web site to the Web Theory of reasoned action (TRA). Conceptualizes behavior as being
customer. Marketing academics should become the primary drivers of driven by behavioral intentions, which are functions of the subject's
such study. The objectives driving this research are both complementary attitude toward a behavior and the subjective norms surrounding
to and consistent with these trends. that behavior. Attitude toward the behavior encompasses the actor's
positive and/or negative feelings about performing the behavior,
2.2. Description of study while subjective norms are dened as the subject's perceptions of
how others feel about the action to be performed (Fishbein and
This study synthesizes the existing marketing and IS literatures to Ajzen, 1975).
develop an empirically-based consensus about the nature and Theory of planned behavior (TPB), a variant of TRA, adds behavioral
inuence of vital consumer-related antecedents to consumers' control to attitude and subjective norms as antecedents to
Internet behaviors. Specically, this study systematically reviews behavioral intention (Ajzen, 1991). Behavioral control is dened
and analyzes the literature on Web site behavioral intentions to as the perceived difculty of performing a behavior, lying on a
952 D.G. Taylor, D. Strutton / Journal of Business Research 63 (2010) 950956

continuum between easily performed behaviors and those requiring same marketing stimuli. The recursive nature of such relationships is
more effort and resources. reected in Fig. 1.
Technology acceptance model (TAM), also based on TRA, is specic to
the IS eld. In this model, actual technology usage is a function of
3.5. Pre-purchase consumer perception attributes
behavioral intention to use the technology, which in turn is
determined by perceived usefulness and perceived ease of use
(Davis et al., 1989).
3.5.1. Perceived usefulness and ease of use
Flow theory entails a state in which a subject becomes immersed in
It is unsurprising that perceived usefulness and perceived ease
an activity to the point of excluding external stimuli, i.e., in the
of use are both widely used in IS literature, since they are key
ow (Csikszentmihalyi, 1988). With ow theory, the process of
constructs in the technology acceptance model (Davis, 1986).
Internet shopping is a reward in itself (Hoffman and Novak, 1996).
Perceived usefulness is dened as the degree to which a person
In the articles included in the meta-analysis, dependent variables believes that using a particular system would enhance his or her job
were behavioral intentions such as purchase, repurchase or loyalty performance. Perceived ease of use captures the degree to which a
behaviors, or intention to spread positive word-of-mouth. Each person believes that using a particular system would be relatively free
outcome would be uniformly viewed as favorable and important by of effort (Davis, 1989).
Internet marketers. Thus, a single outcome variable of behavioral TAM has been applied to a broad variety of software, network
intentions encompassing all of these intended actions was designs and information systems research and can be appropriately
employed as the outcome variable for the meta-analysis. Independent applied to studies of Internet purchasing as the acceptance or
variables and constructs were then compiled from these 25 studies. adoption of a new technology. According to TAM, all else being
Each construct included in the meta-analysis was cited in at least two equal, customers will choose a system that provides the most benet
studies. to the performance of a task and the system that is easiest to use.
In studying consumer acceptance of online shopping (e.g. Van den
3.4. Analysis Poel and Leunis, 1999; Doherty et al., 2003), TAM provides a useful
framework. However, it can be argued that in a post-adoption
Following accepted practice, wherever possible, the constructs from environment, TAM no longer represents an appropriate vehicle
each study were coded into common categories. This was done to through which to study online purchasing behavior. But perceived
create the fewest possible constructs with the maximum number of ease of use and perceived usefulness each remain key predictors of
observations within each construct. Coefcients were then summated behavioral intention. A regression analysis of the studies using TAM
and a mean calculated. As an alternative measurement, the coefcients found that the age of study was not statistically signicant in
were combined using weighted averages to account for differences in predicting the size of the coefcients. This suggests perceived
sample sizes. The equations for both are given below. usefulness or perceived ease of use continue to exert a consistent
and signicant effect size, despite the already widespread acceptance
ri Ni ri i of the Internet as a marketing and purchasing channel. In this meta-
Simple average r = Weighted average r w = analysis, perceived ease of use had a mean effect size of .335.
n Ni
Perceived usefulness, in turn, featured a slightly smaller mean effect
size of .317.
Finally, using criteria specied by Cohen (1988), effect sizes were The marketing discipline tends not to be intrinsically concerned with
categorized as small (.10), medium (.30) or large ( .50). issues such as Web site design, stability or ease of navigation. This
analysis, however, implies such issues ought not to remain the exclusive
3.4.1. Results of the quantitative analysis domain of IS researchers. In a post-adoption context, technical attributes
From the meta-analysis, eight constructs emerged that appeared such as visual design, user interfaces and navigation continue to inuence
to inuence online consumers' behavioral intentions. These eight the purchase intentions of e-consumers. Whereas IS researchers
attributes, in turn, were classied in three distinct categories. These generally account for these interactions through the lens of technology
categories were: acceptance, marketing academics logically might consider these same
issues from other perspectives, i.e. points of view more dear to their
Pre-purchase user perception attributes: key attributes of the Web hearts.
site that consumers consider in their evaluation of the site. These Some may view the emerging study of virtual retail spaces as being
include perceived risk, value, ease of use and usefulness. analogous to the evolution of retailing atmospherics (e.g. Donovan
Pre-purchase user attitudes: pre-existing user attitudes toward and Rossiter, 1982). Just as the design of a retail shopping center
online-purchases in general or toward the e-retailer. These include coalesces the domains of marketing, architecture and interior design
personal Internet interest/afnity, privacy concern and trust. in the physical world, Web site architecture and design surely should
Post-purchase user attitude: the evaluation of the purchasing straddle both the marketing and IS disciplines. While a handful of
experience, conceptualized as satisfaction/dissatisfaction. studies (e.g. Rosen and Purinton, 2004; Hausman and Siekpe, 2009)
have broached the ISmarketing divide in this area, it is a rich
Effect sizes for all constructs fell into Cohen's (1988) medium
untapped area of research.
range (between .10 and .50). However, personal interest in the
Internet exercised a substantially smaller effect size than the other
attributes, while perceived risk and satisfaction featured large 3.5.2. Perceived value
negative and positive effects, respectively, relative to the other In its simplest form, perceived value entails consumers' overall
attributes. The effect sizes are reported in Table 1. assessment of a product's utility based on perceptions of what is given
The conceptual framework for depicting each user attribute and the and received (Zeithaml, 1988). This utilitarian denition is widely seen
relationships tested in this analysis are shown in Fig. 1. As conceived, the as overly simplistic in most current business environments. However,
framework appears consistent with Bagozzi's (1992) appraisal, within e-commerce domains, the intangibility and lack of physical clues
attitude, behavior model, based on the premise that prior experience probably makes the evaluation of value even more complex. Holbrook
with any marketing stimuli, e-based or otherwise, should inuence (1999) denes consumer value as an interactive relativistic preference
subsequent evaluations of the prospect of future interactions with the experience (p. 5), with perceived value encompassing all of the
D.G. Taylor, D. Strutton / Journal of Business Research 63 (2010) 950956 953

Table 1
Summary of meta-analysis and effects.

Emergent construct categories Pre-purchase perceptions Pre-purchase attitudes Post-purchase attitude

Usefulness Ease of use Value Risk Trust Privacy concern Internet interest/afnity Satisfaction

Mean effect sizea


Weighted .32 .34 .31 .46 .33 .39 .26 .46
(unweighted) (.32) (.36) (.33) ( .56) (.34) (.40) (.27) (.40)

Studies
Author(s) Journalb N
Agarwal and Karahanna (2000) MISQ 288 .475c .307c
Anderson and Srinivasan (2003) P&M 1211 .335 .414 .28
Bhattacherjee (2002) JMIS 122 .36
Chellappa and Sin (2005) IT&M 50 .16c .34c

Chen and Barnes (2007) IMDS 103 .211d


Chiou (2004) I&M 209 .27 .37
Devaraj et al. (2002) ISR 171 .253d .74d
Dinev and Hart (2005) IJEC 422 .26c
Dinev and Hart (2006) ESJ 369 .59c .38c .48c
Eastlick et al. (2006) JBR 477 .23
Flavian, et al. (2006) I&M 351 .27 .31
Gefen and Straub (2003) ESJ 161 .32c .25c .42c
Gefen et al. (2003) MISQ 213 .40c .26c
Grazioli and Jarvenpaa (2000) ITSMC 80 .325c
Jarvenpaa et al., (2000) JCMC 120 .29 .59
Koufaris (2002) ISR 280 .415d .119
Lee and Lee (2005) JCIS 163 .71d
Lin and Wang (2006) I&M 255 . .36 .29 .45
Luarn and Lin (2003, 2006) JECR 180 .23d .163c .219
d
Malhotra et al. (2004) ISR 449 .63 .23d
Ribbink, et al., (2004) MSQ 184 .14 .68
Schaupp and Belanger (2005) JECR 188 .181c .439c
Stafford and Stern (2002) IJEC 329 .174 .665 .05
Taylor et al. (2009) ECR 394 .13d .62d
Thatcher and George (2004) JOCEC 441 .64d
a
All effects signicant at .05 unless otherwise noted.
b
Key to abbreviations: ECR = Electronic Commerce Research; ESJ = E-Services Journal; I&M = Information & Management; IJEC = International Journal of Electronic Commerce;
IMDS = Information Management & Data Systems; ISR = Information Systems Research; IT&M = Information Technology and Management; ITSMC = IEEE Transactions on
Systems, Man & Cybernetics; JCIS = Journal of Computer Information Systems; JCMC = Journal of Computer-Mediated Communication; JOCEC = Journal of Organizational
Computing And Electronic Commerce; MISQ = MIS Quarterly; MSQ = Managing Service Quality; P&M = Psychology & Marketing.
c
Signicant at .01.
d
Signicant at <.0001.

subjective, affective and hedonistic dimensions of the consumer's the product or service) (Parasuraman and Grewal, 2000). In Internet
experience. exchanges, the give component may entail degrees of privacy or
Perceived value is thus the evaluation of all get component security, close geographic proximity to and/or physical location of the
(benets buyers receive from the seller) and a give component seller and the ability to physically examine a product being
(costs monetary or otherwise consumer must expand to acquire surrendered, as well as monetary costs. The get component, in
turn, may consist of time savings, home delivery, lower prices or
increased availability, among other benets. Product performance and
service quality are often identied as antecedents of value. But, as
Holbrook (1999) discusses, the perception of value is far more
subjective and relativistic than merely an evaluation of service and
quality, particularly in online transactions with virtual retailers.
Perceived value increases purchase and loyalty intentions by
reducing consumers' propensity to seek information on alternatives.
Even when customers are satised, they may engage in search
behavior if they fail to perceive they are receiving best value
(Anderson and Srinivasan, 2003).
The previously suggested positive relationship with both purchase
intentions and customer loyalty was supported in the meta-analysis,
with perceived value exercising a moderate effect size (mean
r = .308).

3.5.3. Perceived risk


Perceived risk refers to a consumer's belief about the likelihood
of gains and losses being associated with a given consumption
decision (Bradach and Eccles, 1989). Consumers routinely encounter
more uncertainty online than in face-to-face transactions, as sellers
Fig. 1. An integrative model of online purchasing behavior. may be smaller, geographically dispersed or completely unknown
954 D.G. Taylor, D. Strutton / Journal of Business Research 63 (2010) 950956

(Jarvenpaa et al., 2000). Because it is negatively related to online and the resulting behaviors. Milne (2000) conceptualizes consumer
purchasing, perceived risk is important in online consumer behavior privacy as a second-order construct with two dimensions: consumer
(Forsythe and Shi, 2003). Consumers are more likely to purchase knowledge and control. Sheehan and Hoy (2000) identify ve
online when they perceive risk as being low (Lee and Tan, 2003). dimensions: awareness of information collection, usage of informa-
Perceived risk should be positively correlated with privacy tion, information sensitivity, familiarity and compensation. Emerging
concern, and negatively correlated with trust (Jarvenpaa et al., e-commerce research, however, appears to challenge prior concep-
2000; Malhotra et al., 2004). This meta-analysis supports the assertion tualizations of privacy concern as a second-order construct. The
that perceived risk has a strong negative effect on behavioral concept of privacy and as a result, the concept of privacy concern
intentions (mean r = .46). This conrms prior notions that in e-commerce may be a rst-order construct in its own right.
consumers associate risk with the Internet, often outweighing the Numerous studies of online privacy concerns have established the
benets they perceive in the product or service, and implies negative correlation between privacy concern and behavioral inten-
practitioners should consider this risk perception dimension and tions. The results of this meta-analysis support the relationship,
pursue appropriate compensatory measures. revealing a median coefcient of .393, in turn, reecting a medium
effect size.
3.6. Pre-purchase consumer attitudinal attributes
3.6.3. Internet literacy and afnity
3.6.1. Trust Two articles included in this study showed an effect by consumers'
Milne and Boza (1999) introduce trust as a key antecedent to level of Internet literacy and their afnity for transacting online (Stafford
consumers' intention to purchase online, spawning a series of studies and Stern, 2002; Dinev and Hart, 2005). The effect was slightly smaller
(Jarvenpaa et al., 2000; Lee and Turban, 2001; Bhattacherjee, 2002; compared to the other attributes studied (mean r = .263). And, as the
Eastlick et al., 2006) solidifying trust as an integral part of online diffusion of Internet technology continues, consumer literacy levels may
purchasing. Marketing studies generally treat trust as an attitude or well grow more homogenous. But the attribute still merits attention in
belief toward another entity. Gundlach and Murphy (1993) dene trust this study.
as faith or condence that another party in an exchange will fulll Computer literacy is dened by Dinev and Hart (2005) as the ability
obligations. Moorman et al. (1992) dene trust as the willingness to rely to use an Internet-connected computer and Internet applications to
on an exchange partner in whom one has condence. Morgan and Hunt accomplish practical tasks. Higher levels of Internet literacy allows
(1994) dene trust as condence in the reliability and integrity of an consumers to better protect their privacy and information, and this
exchange partner, while Garbarino and Johnson (1999) dene trust as a higher level of control over information may interact with and reduce
customer's condence in the quality and reliability of service offered. privacy concern, given the previously-stated connection between
Jarvenpaa and Tractinsky (1999) conceptualize trust as a consumer's information control and privacy concern.
willingness to rely on the seller and engage in behaviors that make the Computer afnity is related to the level of computer literacy. In
consumer vulnerable. Other studies echo this idea of vulnerability, addition to the level of competence with the computer and the Internet,
adding the elements of uncertainty (Milne and Boza, 1999) and control consumers with high levels of positive feelings about computers and
(Lee and Turban, 2001). The literature also addresses the construct of online shopping have higher levels of computer afnity than consumers
initial trust, in which consumers form trusting behaviors with who can do without their computer for several days and would not
unfamiliar entities (McKnight et al., 2002; Wakeeld et al., 2004; Lee miss them if they were broken (Stafford and Stern, 2002, p. 139).
and Lee, 2005).
Some marketing authors distinguish between trust, trusting beliefs 3.7. Post-purchase consumer attitudes
and trusting behaviors. Some argue that trusting beliefs are a necessary
but not sufcient condition for the existence of trust, given that trusting 3.7.1. Satisfaction
beliefs do not always lead to trusting intentions (Schlosser et al., 2006). The concept of satisfaction is built around customers' post-purchase
However, Morgan and Hunt (1994) state that trusting beliefs are valid evaluation of their experience. Expectations create a frame of reference
measures of trust. For simplicity's sake, this study treats trust, trusting against which consumers make comparative judgments, rating the
beliefs and initial trust were treated as a single construct. outcome as better (positive disconrmation) or worse than expected
In this study, trust was the most widely-included attribute by a large (negative disconrmation) (Oliver, 1980). Anderson and Sullivan
margin. Sixteen of the 25 studies, or 64%, included trust as a signicant (1993) conceptualize satisfaction as a function of not only expectations
predictor in their models. Not only does trust provide a medium effect but also perceived quality and the ease with which consumers are able
size (mean r = .332) toward behavioral intentions, but it is also well- to evaluate quality. Satisfaction has been described as an active, dynamic
documented as a key mediator in relationship marketing efforts evaluative process, integrating meaning, emotion, social processes and
(Morgan and Hunt, 1994; Palmatier et al., 2006). contextual factors (Fournier and Mick, 1999). It captures an ongoing
evaluation of the surprise inherent in the acquisition or consumption of
3.6.2. Privacy concern a product or service (Oliver, 1997).
Privacy concern is an important concept in the study of Internet The positive relationship between satisfaction and repurchase
marketing. In addition to its role in capturing purchasing history and intentions is well-documented (Oliver, 1980, 1981, 1997; Oliver and
other transactional data, technology allows each click, page view or Linda, 1981; Anderson and Sullivan, 1993). This meta-analysis
sequence of pages to be monitored, recorded and analyzed. The supports previous ndings. Anderson and Srinivasan (2003) and
unprecedented ability to collect information threatens to overwhelm Flavian et al. (2006) nd that satisfaction is positively related to
existing mechanisms for regulating privacy (Langenderfer and Cook, loyalty, with the effect moderated by inertia, convenience motivation
2004). The impersonal nature of Web sites and the knowledge or and purchase size. Lin and Wang (2006) report similar results in the
awareness that personal information is owing over a network of mobile environment, with satisfaction exerting not only a direct effect
interconnected computers has been shown in numerous studies to on loyalty behavior, but also a signicant interaction with trust and
increase the level of apprehension about the sharing of personal data. perceived value. Some studies conceptualize commitment as a
Privacy concerns entail consumers' level of apprehension or dis- mediator between satisfaction and behavioral intention (Thatcher
comfort about the potential breach of privacy. Phelps et al. (2001) and George, 2004), or as having direct relationships with both
dene privacy concern as consumers' perceptions regarding exchange commitment and loyalty behaviors (Luarn and Lin, 2003). Another
relationships with marketers that gather and use personal information suggests direct relationships between satisfaction and both trust and
D.G. Taylor, D. Strutton / Journal of Business Research 63 (2010) 950956 955

loyalty behaviors (Ribbink et al., 2004). It has also been conceptualized 4.1. An integrative framework
as an outcome of the TAM constructs of perceived usability and ease of
use (Devaraj et al., 2002). The meta-analysis suggested an overall taxonomy of three categories
Most of these studies examine the effects of satisfaction on loyalty for antecedents pre-purchase consumer perceptions, pre-purchase
(or e-loyalty) rather than behavioral intention. However, the consumer attitudes and post-purchase consumer attitudes. First, pre-
denition of loyalty generally a customer's favorable attitude purchase consumer perceptions include perceived usefulness, perceived
toward a business resulting in repeat buying behavior is primarily ease of use, perceive value and perceived risk. Second, pre-purchase
an attitudinal dimension that itself is closely related to repurchase consumer attitudes entail trust, privacy concern and Internet literacy/
intentions. Thus, this meta-analysis treated loyalty as measured in afnity. Finally, post-purchase level of satisfaction is an extremely
these studies as an acceptable proxy for behavioral intentions. important antecedent to future purchases and loyalty behavior.
Unsurprisingly, this study revealed a large effect size (mean The collection of consumer perceptions and attitudes that emerged
r = .461) for satisfaction, suggesting a need for continued study in the from this analysis summarized in Fig. 1 is innately intertwined.
Internet environment. Examinations of the antecedents to online Considered jointly, these predictive concepts appear to form a complex
satisfaction should still hold promise for future research. series of individual judgments and decision-making processes that
ultimately may inuence much of the online customer's purchasing
4. Discussion of contributions behavior. The model proposed is a simplied representation, drawn from
10 years of research of Internet mediated consumer behavior. The model
As Internet purchasing becomes more widespread and diffused, may provide as a rich starting point for the study of Internet purchasing
marketers are faced not only with a changing environment, but also a and online consumer behavior in what is described as a post-adoption era.
changing customer prole. No longer are Internet users mostly tech- What is the proper framework for studying Internet purchasing?
savvy early-adopters. Instead, Internet users encompass every age Should the technical aspects of Web sites play a prominent role in the
group, socioeconomic class and level of technical experience. Further- marketing academy's investigations of Internet consumer behavior, or
more, as Internet marketing (and Internet consumer behaviors) move should such considerations reside primarily in the IS domain? These
from higher- to atter-growth stages, competition for these online results suggest that the most frequently cited IS constructs perceived
consumers is apt to become even more intense. It is important, then, for usefulness and perceived ease of use remain strong predictors of
practitioners to understand more about the processes behind their behavioral intentions. Indeed, this study suggests the need exists for even
customers' or prospects' buying behavior and loyalty. more collaboration between marketing and IS academics in order to
This meta-analysis examined the extant literature in two disciplines properly capture the breadth of attributes that shape online transactions.
to provide a comprehensive framework for the study of online The same premise appears applicable to attitudinal dimensions.
consumer behavior in a post-Internet-adoption age. A review of the While trust, privacy concern and satisfaction are relevant (and well-
marketing literature reveals a solid foundation of seminal articles studied) topics in any context, they are of particular importance in the
produced during the early years of the Internet's emergence as a force in virtual environment. Thus, a collaborative approach apparently is
the retailing landscape, followed by several dozen articles on various needed to develop a comprehensive understanding of these con-
aspects of Internet strategy, pricing, buyer behavior, legal/nancial structs in both marketing and IS disciplines.
issues and channels of distribution. Meanwhile, the IS discipline
produced a rich but (for the most part) distinct research stream 4.2. Limitations and directions for future research
addressing not just the technical aspects of the Web, but also, at times,
the behaviors of consumers using the Internet. A surprising proportion Through meta-analysis, this study attempts to form a comprehen-
of the IS discipline's literature addressed marketing-related issues. This sive and exhaustive review of the relevant extant literature regarding
study attempted to transfer this knowledge to the marketing literature, online consumer behavior and distill their results into a quantitative
and hopefully it has done so successfully. analysis. Meta-analysis is a powerful tool, but as numerous authors have
Several noteworthy observations emerged when the IS and pointed out, it is not without its limitations. First, many critics make the
marketing literature were empirically compared. Marketing academics apples and oranges argument that results from incommensurable
who are unfamiliar with the IS literature on e-commerce may be goals and procedures cannot be aggregated. However, apples and
surprised by the volume of marketing-related studies in the IS and oranges can be compared on price, weight, nutritional value, and other
management literature, but several interesting distinctions were dimensions (Franke, 2001). Every attempt has been made to ensure the
apparent. First, e-commerce research in the IS discipline frequently fruits of this study are commensurable, but certain differences are
studies online purchasing intentions as a response variable, while a inevitable within comparisons.
search of the leading marketing journals found few empirical studies Other criticisms have included the argument that meta-analyses
with that dependent variable. are inuenced by the biases, errors and methodological aws of its
While the extant marketing literature is rich with examinations of component studies (Eysenck, 1978), as well as researcher selection
moderating inuences, the e-commerce IS literature features very few bias. To mitigate biases, a thorough literature was conducted to ensure
moderators in its models, a deciency future IS researchers may wish the study's integrity.
to address. Indeed, among the studies in this analysis, only two The conceptual framework poses interesting directions for future
articles mentioned moderating effects. Anderson and Srinivasan research. The proposed model is built upon consumer perceptions and
(2003) identify ve moderators of satisfaction's inuence on loyalty attitudes toward online purchasing. As e-commerce matures and gains in
intentions, and Chen and Barnes (2007) study the moderating role of acceptance, how will these attitudes change? Already, the landscape of
familiarity between trust and purchase intentions. retailers has changed dramatically with dot-com rms giving way to
Finally, while the extant marketing literature approaches online a virtual world largely comprised of ofine click-and-mortar rms. How
consumer behavior from a number of theoretical perspectives will this affect risk and value perceptions, and or perceptions of attitudinal
(including TAM), the IS literature overwhelmingly utilizes TAM. Nine constructs such as trust and privacy concern? If the Internet-connected
of the studies were based on the TAM model, with the vast majority not home computer becomes as ubiquitous and unremarkable as a refrig-
explicitly based upon any explicit theoretical foundation. Thus, while a erator or television, will interest and afnity for the Internet disappear as
collaborative ISmarketing research partnership suggests promising a predictor? Similarly, as stability and functionality in Web sites become
directions in future research, the marketing discipline appears to be standard and undifferentiated, will technology acceptance attributes such
better positioned to lead the study of the next era in online purchasing. as ease of use and/or usefulness lose their predictive ability?
956 D.G. Taylor, D. Strutton / Journal of Business Research 63 (2010) 950956

As technological aspects associated with Internet marketing lose Gundlach GT, Murphy PE. Ethical and legal foundations of relational marketing exchanges.
J Mark 1993;57(4):35.
their novelty, a new post-adoption era begs attention. While much of Hausman AV, Siekpe JS. The effect of web interface features on consumer online purchase
the research to date in both marketing and IS has studied the diffusion intentions. J Bus Res 2009;62(1):5-13.
and acceptance of e-commerce, these results suggest future research Hoffman DL, Novak TP. Marketing in hypermedia computer-mediated environments:
conceptual foundations. J Mark 1996;60(3):50.
should focus on the interaction of technology and retailing. Analogous Holbrook MB. Introduction to value. New York: Routledge; 1999.
to the study of atmospherics in the brick-and-mortar world, this Jarvenpaa SL, Tractinsky N. Consumer trust in an Internet store: a cross-cultural
research should continue to strive for understanding the behaviors validation. J Comput Mediated Commun 1999;5(2):1-35.
Jarvenpaa SL, Tractinsky N, Vitale M. Consumer trust in an Internet store. Inf Technol Manag
and attitudes of the next generation of online shoppers. Future 2000;1(12):4571.
researchers might also examine the framework developed above Koufaris M. Applying the technology acceptance model and ow theory to online consumer
through a longitudinal approach to determine whether a recursive behavior. Inf Syst Res 2002;13(2):20523.
Langenderfer J, Cook DL. Oh, what a tangled web we weave: the state of privacy
relationship exists between e-behaviors and e-attitudes, or vice-versa.
protection in the information economy and recommendations for governance. J Bus
Res 2004;57(7):73447.
References Lee SM, Lee SJ. Consumers' initial trust toward second-hand products in the electronic
market. J Comput Info Sys 2005;46(2):8598.
Agarwal R, Karahanna E. Time ies when you're having fun: cognitive absorption and Lee KS, Tan SJ. E-retailing versus physical retailing: a theoretical model and empirical
beliefs about information technology usage. MIS Quarterly 2000;24(4):66594. test of consumer choice. J Bus Res 2003;56(11):87785.
Ajzen I. The theory of planned behavior. Org Behav Hum Decis Process 1991;50(2):179211. Lee MKO, Turban E. A trust model for consumer Internet shopping. Int J Electron
Alba J, Lynch J. Interactive home shopping: consumer, retailer, and manufacturer Commer 2001;6(1):7591.
incentives to participate in electronic marketplaces. J Mark 1997;61(3):3853. Lin H-H, Wang Y-S. An examination of the determinants of customer loyalty in mobile
Anderson RE, Srinivasan SS. E-satisfaction and e-loyalty: a contingency framework. commerce contexts. Inf Manage 2006;43(3):27182.
Psychol Mark 2003;20(2):12338. Luarn P, Lin H-H. A customer loyalty model for e-service context. J Electron Commer Res
Anderson EW, Sullivan MW. The antecedents and consequences of customer 2003;4(4):15667.
satisfaction for rms. Mark Sci 1993;12(2):12543. Malhotra NK, Kim SS, Agarwal J. Internet users' information privacy concerns (iuipc):
Bagozzi RP. The self-regulation of attitudes, intentions, and behavior. Soc Psychol Q the construct, the scale and a causal model. Info Sys Res 2004;15(4):33655.
1992;55(2):178204. McKnight DH, Choudhury V, Kacmar C. Developing and validating trust measures for e-
Bhattacherjee A. Individual trust in online rms: scale development and initial test. J Manag commerce: an integrative typology. Info Sys Res 2002;13(3):33461.
Info Sys 2002;19(1):21141. Milne GR. Privacy and ethical issues in database/interactive marketing and public
Bradach JL, Eccles RG. Price, authority, and trust: from ideal types to plural forms. Annu Rev policy: a research framework and overview of the special issue. J Public Policy Mark
Sociology 1989;15:97. 2000;19(1):16.
Burke RR. Do you see what I see? The future of virtual shopping. Acad Mark Sci J 1997;25 Milne GR, Boza M-E. Trust and concern in consumers' perceptions of marketing
(4):352. information management practices. J Interact Market 1999;13(1):5-24.
Chellappa RK, Sin RG. Personalization versus privacy: an empirical examination of the Moorman C, Zaltman G, Deshpande R. Relationships between providers and users of
online consumer's dilemma. Information Technology and Management 2005;6(23): market research: the dynamics of trust within and between organizations. J Mark
181202. Res 1992;29(3):314.
Chen Y-H, Barnes S. Initial trust and online buyer behavior. Ind Manage Data Sys 2007;107 Morgan RM, Hunt SD. The commitmenttrust theory of relationship marketing. J Mark
(1):2136. 1994;58(3):2038.
Chiou J-S. The antecedents of consumers' loyalty toward internet service providers. Oliver RL. A cognitive model of the antecedents and consequences of satisfaction
Information & Management 2004;41(6):68595. decisions. J Mark Res 1980;17(4):4609.
Cohen J. Statistical power analysis for the behavioral sciences. Hillsdale, NJ: L. Erlbaum; Oliver RL. Measurement and evaluation of satisfaction processes in retail settings.
1988. J Retail 1981;57(3):25.
Csikszentmihalyi M. Optimal experience: psychological studies of ow in conscious- Oliver RL. Satisfaction: a behavioral perspective on the consumer. New York: McGraw-
ness. U.K.: Cambridge; 1988. Hill; 1997.
Davis FD. A technology acceptance model for empirically testing new end-user information Oliver RL, Linda G. Effect of satisfaction and its antecedents on consumer preference and
systems. Sloan School of Management, Massachussetts Institute of Technology; 1986. intention. Adv Consum Res 1981;8(1):8893.
Davis FD. Perceived usefulness, perceived ease of use, and user acceptance of Palmatier RW, Dant RP, Grewal D, Evans KR. Factors inuencing the effectiveness of
information technology. MIS Quar 1989;13(3):31840. relationship marketing. J Mark 2006;70(4):13753.
Davis FD, Bagozzi RP, Warshaw PR. User acceptance of computer technology: a Parasuraman A, Grewal D. Serving customers and consumers effectively in the twenty-
comparison of two. Manage Sci 1989;35(8):982. rst century: a conceptual framework and overview. J Acad Mark Sci 2000;28(1):9.
Devaraj S, Fan M, Kohli R. Antecedents of B2C channel satisfaction and preference: Phelps JE, D'Souza G, Nowak GJ. Antecedents and consequences of consumer privacy
validating e-commerce metrics. Info Sys Res 2002;13(3):31633. concerns: an empirical investigation. J Interact Market 2001;15(4):2-17.
Dickson PR. Understanding the trade winds: the global evolution of production, Ribbink D, Riel ACRv, Liljander V, Streukens S. Comfort your online customer: quality,
consumption, and the Internet. J Consum Res 2000;27(1):115. trust and loyalty on the Internet. Manag Serv Qual 2004;14(6):446.
Dinev T, Hart P. Internet privacy concerns and social awareness as determinants of Richtel M, Tedeschi B. Some buyers grow web-weary, and online sales lose steam. The
intention to transact. Int J Electron Commer 2005;10(2):7-29. New York Times 2007;(June 17):A1.
Dinev T, Hart P. An extended privacy calculus model for e-commerce transactions. Rosen DE, Purinton E. Website design: viewing the web as a cognitive landscape. J Bus
Information Systems Research 2006;17(1):6180. Res 2004;57(7):78794.
Doherty N, Ellis-Chadwick F, Hart C. An analysis of the factors affecting the adoption of Schaupp LC, Belanger F. A conjoint analysis of online consumer satisfaction. Journal of
the Internet in the UK retail sector. J Bus Res 2003;56(11):88797. Electronic Commerce Research 2005;6(2):95-111.
Donovan RJ, Rossiter JR. Store atmosphere: an environmental psychology approach. Schlosser AE, White TB, Lloyd SM. Converting web site visitors into buyers: how web
J Retail 1982;58(1):34. site investment increases consumer trusting beliefs and online purchase intentions.
Eastlick MA, Lotz SL, Warrington P. Understanding online b-to-c relationships: an integrated J Mark 2006;70(2):13348.
model of privacy concerns, trust, and commitment. J Bus Res 2006;59(8):87786. Sharma A, Sheth JN. Web-based marketing: the coming revolution in marketing
Eysenck HJ. An exercise in mega-silliness. Am Psychol 1978;33(5):517-517. thought and strategy. J Bus Res 2004;57(7):696702.
Fishbein M, Ajzen I. Belief, attitude, intention, and behavior: an introduction to theory Sheehan KB, Hoy MG. Dimensions of privacy concern among online consumers. J Public
and research. Reading, MA: Addison-Wesley; 1975. Policy Mark 2000;19(1):6273.
Flavian C, Guinaliu M, Gurrea R. The role played by perceived usability, satisfaction and Stafford MR, Stern B. Consumer bidding behavior on Internet auction sites. Int J Electron
consumer trust on website loyalty. Inf Manage 2006;43(2006):1-14. Commer 2002;7(1):135.
Forsythe SM, Shi B. Consumer patronage and risk perceptions in Internet shopping. Taylor DG, Davis DF, Jillapalli R. Privacy concern and online personalization: the
J Bus Res 2003;56(11):86775. moderating effects of information control and compensation. Electronic Commerce
Fournier S, Mick DG. Rediscovering satisfaction. J Mark 1999;63(4):5-23. Research 2009;9(3):20323.
Franke GR. Applications of meta-analysis for marketing and public policy: a review. Thatcher JB, George JF. Commitment, trust, and social involvement: an exploratory study of
J Public Policy Mark 2001;20(2):186. antecedents to web shopper loyalty. J Organ Comput Electron Commer 2004;14
Garbarino E, Johnson MS. The different roles of satisfaction, trust, and commitment in (4):24368.
customer relationships. J Mark 1999;63(2):7087. Van den Poel D, Leunis J. Consumer acceptance of the Internet as a channel of distribution.
Gefen D, Straub D. Managing user trust in B2C e-services. E - Service Journal 2003; J Bus Res 1999;45(3):24956.
2(2):7-24. Wakeeld RL, Stocks MH, Wilder WM. The role of web site characteristics in initial trust
Gefen D, Karahanna E, Straub D. Trust and TA in online shopping: an integrated model. formation. J Comput Info Sys 2004;45(1):94.
MIS Quarterly 2003;27(1):5190. Wildstrom, SH. Easing your way onto the net. Business Week 1995;(July 24):17.
Grazioli S, Jarvenpaa SL. Perils of internet fraud: an empirical investigation of deception Zeithaml VA. Consumer perceptions of price, quality, and value: a means-end model and
and trust with experienced internet consumers. IEEE Transactions on Systems, synthesis of evidence. J Mark 1988;52(3):2.
Man, and Cybernetics 2000;30(4):395410.