Individuals’ Preference Orientations toward Facets of Internet Shopping Sites:
A Conceptual and Measurement Model

Brian F. Blake, Cleveland State University
Rhiannon L. Hamilton, Ipsos-ASI
Kimberly A. Neuendorf, Cleveland State University
Ryan Murcko, Cleveland State University

Introduction    
     Internet shopping is spreading around the world, spawning a wide range of research in information technology, communication, psychology, marketing, and related disciplines about this innovation. Many of the research questions require a viable taxonomy of site features in order to be investigated properly. Such research questions pertain to theory development, e.g., the role of site properties in the diffusion of this innovation (e.g., Rogers, 2003), or the degree to which the impact of site/features varies with product/service domains (e.g., Levin, Levin, & Weller, 2005) or with characteristics of the user (e.g., Yen, 2005). Other applications involve methodological development, such as determining how to compare site feature preferences cross-nationally (Blake & Neuendorf, 2004; Blake, Shamatta, Neuendorf, & Hamilton, 2009), and practitioner oriented investigations, such as devising web sites fitting the demands of specific subpopulations (e.g., Blake, Neuendorf, & Valdiserri, 2005).
     A review of this burgeoning body of research suggests four conclusions. First, there is no single comprehensive listing of business-to-consumer (B2C) site features that is generally accepted. Several investigators have taken steps to provide broad coverage of a wide range of features (e.g., Torkzadeh & Dhillon, 2002). Others have narrowed their focus to specific product/service realms like destination travel (Park & Gretzel, 2007) or to site features related to particular factors determining shopping behavior like perceived risk (e.g., Cases, 2002) or trust (e.g., Bart, Shankar, Sultan & Urban, 2005; Gefen & Straub, 2004). Still others have delimited analyses to specific forms of commercial web sites, such as sites allowing shoppers to consume services without direct assistance from service personnel (e.g., Yen, 2005). Second, the “features” (or attributes) have been conceptualized at various levels of abstraction running from the highly concrete (e.g., “24x7x365 user accessibility” and “a standard navigation bar, a home button, and back/forward button available on every page” in van Iwaarden, van der Wiele, Ball, & Millen, 2004) to the highly abstract (e.g., “the purpose is clear” in van Iwaarden et al., 2004). Further, the facets sometimes are “objective” aspects of a site, such as the two examples of concrete features in the van Iwaarden et al. schema; still more often they are user judgments about the site’s features, e.g., “well organized hyperlinks” and “customers enjoy visiting the website” in Liu and Arnett (2000). Third, features differ, often dramatically, in their importance in the eyes of site users (e.g., Zhang & von Dran, 2001-2002) and in their impact on shopping behaviors (e.g., Park & Kim, 2003). Fourth, many research questions require the use of a set of site features that: a) has broad coverage of feature types, b) is theoretically linked to (and, hence, is interpretable in terms of) established theoretical constructs, and c) potentially can be applied to commercial sites for most product/service categories.
     The objectives of this paper are, first, to propose a set of site features for use in investigations requiring a broad coverage of feature types, items with a theoretical grounding, and applicability to most consumer product/service classes. Second, based on an analysis of feature preferability, we suggest underlying dimensions for use when asking what features users want in a consumer shopping site.
VISA
     We propose VISA, a listing of 55 site features for use in scientific and professional research. VISA, “Variegated Inventory of Site Attributes,” is wide ranging in both abstraction and coverage.
     Abstraction is the extent to which a feature is a concrete/objective characteristic of a site or is an evaluative response either to the site as a whole or to a particular component of that site. Items are included at each of three levels of abstraction: a) “manifest characteristics,” readily observable by a user and requiring little insight into the contribution or role played by a characteristic (e.g., absence of typographical errors in the text); b) “attributional properties,” requiring user judgments about the consequences of a concrete characteristic or group of concrete characteristics (e.g., “interesting graphics,” “convenient operation,” “simple navigation”); and c) latent functions, still greater degree of abstraction requiring some degree of user insight about the goal of turning to an online shopping site (e.g., “enjoyable to visit,” “good place to find a bargain”). The majority of VISA’s facets are at the intermediate attributional properties level, consistent with the level of abstraction used in the bulk of the extant theoretical taxonomies of site features. Distinctions among the three levels are not hard and fast; user perceptions of a feature’s abstraction should vary with the experience and sophistication of those users. Not included are features outside the lexicon of most users but rather pertain to processes imputed by knowledgeable investigators, e.g., “flow” or “presence.”
     Coverage pertains to the variety of feature groupings (categories or dimensions) tapped by the items. Features were selected to represent, first, taxonomies of features proposed to cover the major dimensions customers use to judge the effectiveness or quality of a commercial site. While several other taxonomies are relevant (e.g., Aladwani & Palvia, 2002; Liu & Arnett, 2000; Muyelle, Moenaert, & Despontin, 2004; Wolfinbarger & Grilly, 2003), three appeared most pertinent. The Torkzadeh and Dhillon (2002) and Ranganathan and Ganapathy (2002) categorizations were devised for B2C shoppers; the Chakraborty, Lala, and Warren (2003) taxonomy was proposed for business-to-business (B2B), but offers insights for B2C also. Second were two widely employed adoption of innovation models, the diffusion of innovation framework of Rogers (2003) and the Technology Acceptance Model (Davis, Bagozzi, & Warshaw, 1989), with components drawn from the latter’s extension, “TAM 2” (Venkatesh & Davis, 2000; Venkatesh, Morris, Davis, & Davis, 2003). Third, features were added that pertain to factors empirically well established to be particularly influential in consumers’ use of online B2C sites but are not specifically treated in major taxonomies (trust in vendor, perceived risk, perceived newness, and presence).
     In order to keep the inventory reasonably short, as much as possible an item prepared to represent a category or dimension in one taxonomy was written so as to also operationalize a category/dimension in another taxonomy. This objective was possible because, first, as noted previously (e.g., van Iwaarden et al., 2004), a category in one theoretical taxonomy is sometimes similar (if not identical) to a category in an alternative taxonomy, although those two categories may be labeled differently. Second, a given feature may perform multiple functions e.g., striking graphics might increase both presence and enjoyment. Except for particular Cases (2002) categories, each VISA item pertains to two or more categories or dimensions. Conversely, each category or dimension is represented by two or more items.
     Identification numbers (see Table 1) of the features pertinent to each taxonomic category/process are listed in parentheses in the several pages that follow. The items are intended to represent the conceptual definition of a given category/dimension and not necessarily to paraphrase or to reproduce specific items used by previous investigators to exemplify their conceptual categories.
     Based on Keeney’s (1999) distinction between means objectives and fundamental objectives, and later fine tuned by Chang, Torkzadeh, and Dhillon (2004), Torkzadeh and Dhillon (2002) proposed nine consumer values that serve as yardsticks against which consumers evaluate a site. Two (“Internet ecology” and “shopping travel”) were deleted because they do not pertain to consumer preferences for one site over another. The value dimensions (and the features devised in this research to represent them) are: a) shopping convenience (Nos. 1, 2, 9, 10, 20, 26, 27, 30, 33), b) customer relations (Nos. 18, 21, 30, 38), c) product value (Nos. 4, 13, 16, 22), d) product choice (Nos.  12, 15, 44, 45, 52), e) online payment (Nos. 17, 37, 47, 55), f) vendor trust (Nos. 7, 25, 31, 50), and g) shipping error (Nos. 35, 40).   
     Ranganathan and Ganapathy (2002) suggested that the aspects of B2C web sites that determine the success of a site can be grouped into four rather broad categories: a) content (Nos. 2, 5, 15, 21, 30, 33, 38, 44, 52), b) design (Nos. 4, 20, 23, 26, 39, 41, 42, 46, 48, 49, 51, 53, 54), c) security (Nos. 17, 31, 47, 55), and d) privacy (Nos. 25, 33, 47). Next, Chakraborty, Lala, and Warren (2003) presented a set of seven preference dimensions that underlie customers’ judgments of what is important in a B2B web site. These dimensions are pertinent to B2C shopping also. We do, though, combine two dimensions, ”non-transaction related interactivity” and “transaction related interactivity,” into a single dimension for B2C, “interactivity” (Nos. 23, 30, 38).  The five other dimensions are: a) organization (Nos. 1, 2, 26, 33), b) privacy/security (Nos. 17, 25, 31, 37, 47, 55), c) informativeness (Nos. 5, 15, 21, 44, 52), d) personalization (Nos. 36, 23), and e) entertainment (Nos. 6, 8, 14).
     The first innovation adoption framework, Rogers (2003), proposes that five aspects of innovation determine the rate of adoption/diffusion of that innovation: a) comparative advantage (Nos. 4, 13, 16, 22, 25, 45, 52), b) compatibility with one’s social environment (Nos. 6, 10, 11, 32), c) complexity of use (Nos. 1, 2, 21, 26, 52), d) trialability (i.e., informs users of the innovation’s characteristics without requiring full adoption; Nos. 1, 2, 21, 26, 52), and e) observability (of the innovation’s use by others; Nos. 19, 34). The other adoption approach is the Technology Acceptance Model of Davis (Davis, Bagozzi, & Warshaw, 1989), inspired by the theory of reasoned action by Fishbein and Ajzen (1975). Originally proposed for adoption of new technologies within an organizational setting, the theory was subsequently augmented (Venkatesh & Davis, 2000; Venkatesh, Morris, Davis, & Davis, 2003) with additional cognitive instrumental and social influence processes consistent with its grounding in the theory of reasoned action. The two basic (Davis et al.,1989) sets of features determining acceptance of a new technology are: a) perceived usefulness (Nos. 5, 12, 15, 16, 22, 44, 49) and b) perceived ease of use (Nos. 1, 2, 20, 21, 26, 30, 33, 38, 52). While many of the components of the later augmentations are of less relevance to online consumer shopping (e.g., voluntariness of adoption decision), a helpful addition to the base model is that of social norms, the opinions or pressure from important other persons in one’s social milieu (Nos. 6, 11, 32, 34).
     The first additional factor important to the nature of consumers’ online shopping is consumer trust in a site’s vendors(s) (e.g., Wang & Emurian, 2005). The detailed analysis of Bart et al.(2005) indicated that eight categories of site characteristics drive shoppers’ trust in B2C sites: a) privacy (Nos. 25, 33, 37, 55), b) security (Nos. 17,31,47,50), c) navigation and presentation (Nos. 2, 14, 20, 23, 26, 39,41, 42, 46, 51, 53), d) brand strength (Nos. 7, 34, 45), e) advice (Nos. 15, 44, 52), f) order fulfillment (Nos. 1, 9, 13, 18, 26, 35, 40), g) community (Nos. 5, 6, 11), and h) absence of errors (Nos. 27, 29).
     The next factor is perceived risk, related to but not the same as trust (Das & Teng, 2004); it has been established as an impediment to online shopping (e.g., Cases, 2002). Cases, first, proposes that there are four sources of risk: product, remote transaction, Internet, and (web) site. For each risk source there are particular risk relievers: a) product - price information (No. 4), b) product - information from merchandising (Nos. 15, 23), c) product - comparison of products (Nos. 44, 52), d) product - seeing the product in advance (Nos. 23, 46, 53), e) product - well known brand(s) (Nos. 7, 45), f) remote transaction – money back guarantee (No. 50), g) remote transaction – can exchange product (No. 21), h) remote transaction – speak with salesperson (Nos. 30, 38), i) Internet – payment security (Nos. 17, 25, 31, 37, 47, 55), j) Internet – word of mouth (Nos. 5, 32), k) Internet – remote contact (Nos. 38, 18), l) web site – site reputation (Nos. 6, 34), m) web site – familiarity/experience with site (Nos. 3, 19, 24, 28, 43). Further, Cases differentiated among eight types of risk associated with B2C e-tail shopping: a) product performance (Nos. 5, 15, 21, 45), b) financial (Nos. 4, 13, 16, 22, 50), c) time (Nos. 1, 2, 9, 18, 35, 40, 52), d) delivery (Nos. 9, 35), e) social (Nos. 6, 19, 32), f) privacy (Nos. 25, 31, 33, 37), g) payment (Nos. 17, 47, 55), h) sources (i.e., concern about reliability of the site itself; Nos. 3, 7, 24, 28, 34, 44).
     Being new and/or different from other sites (Nos. 3, 24, 28, 43) has been proposed as an attractant to online shopping (e.g., Blythe, 1999), but has also been shown to be a disincentive to shopping at a site (Blake, Valdiserri, Neuendorf, & Valdiserri, 2007). Finally, “presence” has been defined as an information receiver’s sense that there is no barrier or a medium interposed between the receiver and the information source (Lombard & Ditton, 1997). Presence has been noted to impact the magnitude and nature of online B2C shopping (Gefen & Straub, 2004; Jahng, Jain, & Ramamurthy, 2000). Potentially presence enhancing features are Nos. 36, 39, 41, 42, 46, 48, 49, 51, 53, and 54.
Method
Data Collection
     Respondents were recruited by a snowball procedure attempting to obtain a geodemographically diverse sample. Research team members and their colleagues contacted church, business, social, and their organizations as well as persons in their own social networks throughout the US.
     Participants were invited to a web site hosted by the researchers’ university. There a cover letter described the study as an exploration of shoppers’ perspectives on online shopping and noted the anonymity of their response. Attached was a questionnaire containing the VISA items and demographic questions, among others.
Sample
     Of the 534 Internet users responding, 489 had at least some online shopping experience and had provided useful data. Respondents lived principally in the Midwest (76%) and Northeast (13%) states. They tended to be fairly young (median age of 29), female (65%), unmarried (58%), college educated (52% graduated college), employed full time (52%), with a median household income of $45,000. The majority had used the Internet for over six years and varied in frequency of online purchasing (53% purchased at least monthly).
Measures
     Demographics. The questions (and their response categories) were: a) “What is your gender?” (male/female); b) “How old are you (in years)?” (open); c) “What is your marital status?” (single, married, separated/divorced, widowed); d) “In what state is your permanent address at this present time?” (50 state drop down box); e) “What was the last year of education you completed?” (some high school, high school, technical school, some college, college graduate/professional school); f) “What is your current employment?” (check all applicable: full time, part time, self employed, temporarily unemployed, full time student, homemaker, retired); g) “How many people live in your household, including yourself?” (number); and h) “Please indicate which of the following best represents your annual household income before taxes.” ($10,000 or less; $10,001 – 20,000; $20,001 - 30,000; $30,001 – 40,000; $40,001 – 50,000; $50,001 – 75,000; $75,000 – 100,000; more than $100,000).
     Feature preference.  Respondents were asked, “How strongly, if at all, do the following aspects of a web site encourage you to shop at a particular site?” Items were rated on a 7-point scale from (1) “Does not encourage me at all” to (7) “strongly encourages me.” Listed were the 55 VISA items.
Results
Feature Preference
     Table 1 displays the means and standard deviations of the item ratings. There is considerable spread in appeal; most items (53%) were fairly popular (M> 5.00), while at least some (20%) were relatively unattractive (M< 3.00).
Preference Dimensions
     Ratings were factor analyzed via principal components with Varimax rotation, yielding 11 factors with eigenvalues above 1.00; together they explained 61.93% of the variance. Table 1 displays the eigenvalue (e) and percent of variance explained by each factor, and the loading of an item on the factor on which its loading is highest. Review of loadings suggested the following interpretations of the emergent factors.  The features are organized by factor in Table 1.
Factor 1: “Security Transactions and Privacy.” A nine-item dimension indicating desire for features providing security of personal, financial, and transactional information.
Factor 2: “Near Ideal.” An eight-feature set appealing to many shoppers. Individuals scoring high want features that yield good and inexpensive products quickly, easily, and reliably.
Factor 3: “Visual and Auditory Richness.” Persons scoring high on this six-attribute set desire sensory experience with visual and auditory stimulation and personalized recognition.
Factor 4: “Web Site Functionality.” These six features pertain to a site’s operating clearly and efficiently, without errors in text or operation.
Factor 5: “Product Comparison.” These five attributes provide the shopper the opportunity to compare and evaluate products.
Factor 6: “New and Different.” Persons scoring high on these four items are interested in recently introduced and original sites.
Factor 7: “Uniquely Entertaining.” These four features indicate attraction to sites that are distinctive, entertaining, fun to discuss with others.
Factor 8: “True to Its Word.” These five items – including receipt of a best site award, prominent display of its privacy policy, and assurance that products dependably arrive when promised – indicate features of a credible, trustworthy site.
Factor 9: “Human Touch.” Persons scoring high on this three item dimension opt to see real people in real settings; even animated animals are anthropomorphized.
Factor 10: “Product Information.” High scores on this three-item set indicate greater interest in sites that describe the product, indicate what other people think about it, and show them how to order it.
Factor 11: “Others’ Recommendation.” Two features reflect the desire to use sites recommended by others, whether gleaned from media sources or from one’s circle of friends and family.
Demographic Correlates of Preference Dimensions
     Responses to the demographic questions were rescored to fit distributional and frequency requirements of multiple regression. Dummy variables were composed: a) gender: male = 1, female = 0; b) education: college graduates = 1, non-graduates = 0; c) marital status: married = 1, others = 0; and d) employment: full time = 1, others = 0. Income categories were re-coded using the category midpoint and scores were treated as a continuous variable. Age and household size, scored as the number of years and persons reported, were treated as continuous scores. Tolerance scores of predictor variables ranged from .58 to .97, indicating an absence of multicollinearity among these seven demographic variables.
     The seven demographic scores were entered as a single block into linear regressions predicting respondents’ factor scores on each of the dimensions. Respondents’ demographic characteristics were found modestly related to their preference for five of the 11 dimensions. Those with stronger preferences for the Factor 1: Secure Transaction and Privacy dimension (R_{7, 349} = .24, R² = .06, R²adj = .04, F =2.88, p<.01) tended to be female (beta = -.14, p<.01) and to have higher income (beta = .18, p<.01). Next, features on the Factor 2: Near Ideal dimension (R_{7, 349} = .23, R² = .05, R²adj = .03, F = 2.67, p<.05) appealed more to younger persons (beta =-.21, p<.01). Turning to Factor 4: Web Site Functionality features (R_{7, 349} = .26, R² = .07, R²adj = .05, F = 3.61, p<.01), we see that the more functional sites more strongly attract females (beta =-.17, p<.01) and those from smaller households (beta =-.15, p<.01). Next, those particularly drawn to sites with Factor 7: Uniquely Entertaining features (R_{7, 349} = .23, R² = .05, R2adj = .03, F = 2.69, p<.05) are more often female (beta =-.10, p<.05) and from larger households (beta = .14, p<.01). Finally, Factor 9: Human Touch (R_{7, 349} = .20, R² = .04, R²adj = .02, F = 2.10, p<.05) features attracted younger users (beta =-.16, p<.05).
     Demographics were not significantly (p>.05) related to: Factor 7: Visual and Auditory Richness, Factor 5: Product Comparison, Factor 6: New and Different, Factor 8: True to Its Word, Factor 10: Product Information, or Factor 11: Others’ Recommendation.
Summary and Discussion
     Proposed was VISA, an inventory of 55 B2C site features for use in research requiring a diversified listing of feature types that has a conceptual grounding and applicability to a wide range of consumer products/services. The survey evidence demonstrated, first, that the inventory successfully incorporates features of varying importance to users. Second, the study revealed that 11 dimensions underlie preferences for these wide ranging items, suggesting that these 11 serve as major evaluative yardsticks consumers use to decide what they want to see in a B2C site. Third, it was found that preferences for six of the 11 dimensions were unrelated to the demographic variables studied. For those five dimensions to which demographics were related, the relationships were of only modest magnitude. Thus, the relevance of the VISA items and dimensions might well hold across demographic subpopulations, at least within the US Internet environment.
     Fourth, it was discovered that none of the emergent factors are a precise match with theoretical dimensions that motivated the development of the roster of the 55 features.  That is, factors did not quite match Torkzadeh and Dhillon’s (2002) proposed nine dimensions, Rangathan and Ganapathy’s (2001) suggested four categories, nor the seven preference dimensions of Charkraborty, Lala, and Warren (2003).  Nor did the 11 factors coincide with innovativeness-related dimensions forwarded by Rogers (2003) and Davis et al. (1989), or the trust-related dimensions of Bart et al. (2005), or the risk-related dimensions of Cases (2002).  Several aspects seem to be responsible for the lack of isomorphism between this study’s 11 factors and past taxonomies. 
     Certainly, preference dimensions, as measured here, do not necessarily correspond to dimensions underlying perceptions of certain aspects of performance of web sites, as measured in some past studies.  Dimensions of the latter should reflect the co-occurrence of features in web sites as well as other peculiarities of the sites.  Further, dimensions of preference may cross functional boundaries.  Correspondingly, features serve multiple functions, e.g., “product price” may be seen diversely as an indicator of Rogers’ comparative advantage, Davis et al.’s perceived usefulness, or Ranganathan and Ganapathy’s content dimension. Another reason is that a factor structure will always reflect the range and variety of constructs presented to respondents.  The range in this study is much broader than utilized in the past.  For example, Torkzadeh and Dhillon (2002) include no aspects of playfulness, fun, flow, or presence in their work. Thus, the present investigation finds a novel, and rich, structure of web user preferences that transcends functionality and effectively “shuffles” past organizing structures.  This mixed structure clearly confirms the “variegated” nature of our approach—VISA (Variegated Inventory of Site Attributes).
     A limitation of the study is that the sample may not be strictly representative of the US online shopping population. Future research should assess the replicability of the results using a large representative sample. The observed low to negligible relationship of the preference dimensions to demographic characteristics, though, suggests optimism about the replicability of these results in sectors with different demographic profiles. A second limitation is that the study, like so many other analyses of feature importance to users (e.g., Zhang & von Dran, 2001-2002), relies on self report measures. In future studies statistical estimates of how strongly features determine purchase decisions (e.g., Levin, Levin, & Weller, 2005) should be meshed with self report indicators for a more complete understanding of the drawing power of site features.
     Finally, given the encouraging results with the VISA framework, future research should plumb differences between product/service domains in feature appeal and explore also the differential appeal of a feature for browsing and purchasing behaviors.

References

Aladwani, A. M., & Palvia, P. C. (2002). Developing and validating an instrument for measuring user-perceived web quality. Information & Management, 39, 467-476.

Bart, Y., Shankar, V., Sultan, F., & Urban, G. (2005). Are the drivers and role of online trust the same for all web sites and consumers? A large-scale exploratory empirical study. Journal of Marketing, 69(4), 133-152.

Blake, B. F., & Neuendorf, K. A. (2004). Cross national differences in website appeal: A framework for assessment. Journal of Computer Mediated Communication, 9(4), 1-34.

Blake, B. F., Neuendorf, K. A., & Valdiserri, C. M. (2005). Tailoring new websites to appeal to those most likely to shop online. Technovation, 25, 1205-1214.

Blake, B. F., Shamatta, C., Neuendorf, K. A., & Hamilton, R. (2009, in press). The cross-national comparison of website feature preferences: A practical approach. International Journal of Internet Marketing and Advertising. 

Blake, B. F., Valdiserri, C. M., Neuendorf, K. A., & Valdiserri, J. N. (2007). The online shopping profile in the cross-national context: The role of innovativeness and perceived innovation newness. International Journal of Consumer Marketing, 19(3), 23-51.

Blythe, J. (1999). Innovativeness and newness in high-tech durables. Journal of Product & Brand Management, 8(5), 415-429.

Cases, A. (2002). Perceived risk and risk-reduction strategies in Internet shopping. International Review of Retail, Distribution & Consumer Research, 12(4), 375-394.

Chakraborty, G., Lala, V., & Warren, D. (2003). What do customers consider important in B2B websites?. Journal of Advertising Research, 43(1), 50-61.

Chang, J.-C., Torkzadeh, G., & Dhillon, G. (2004). Re-examining the measurement models of success for Internet commerce. Information & Management, 41(5), 577-584.

Das, T. K., & Teng, B.-S. (2004). The risk-based view of trust: A conceptual framework. Journal of Business and Psychology, 19(1), 85-116.

Davis, F., Bagozzi, R., & Warshaw, P. (1989). User acceptance of computer technology: A comparison of two theoretical models. Management Science, 35(8), 982-1003.

Fishbein, M. A., & Ajzen, I. (1975). Belief, attitude, intention and behavior: An introduction to theory and research. Reading, MA: Addison Wesley.

Gefen, D., & Straub, D. W. (2004). Consumer trust in B2C e-commerce and the importance of social presence: Experiments in e-products and e-service. Omega, 32, 407-424.

Jahng, J., Jain, H., & Ramamurthy, K. (2000). Effective design of electronic commerce environments: A proposed theory of congruence and an illustration. IEEE Transactions on Systems, Man and Cybernetics Part A – Systems and Humans, 30(4), 456-471.

Keeney, R. L. (1999). The value of Internet commerce to the customer. Management Science, 45, 533-542.

Levin, A., Levin, I., & Weller, J. (2005). A multi-attribute analysis of preferences for online and offline shopping: Differences across products, consumers, and shopping stages. Journal of Electronic Commerce Research, 6(4), 281-290.

Liu, C., & Arnett, K. P. (2000). Exploring the factors associated with website success in the context of electronic commerce. Information & Management, 38, 23-33.

Lombard, M., & Ditton, T.  (1997). At the heart of it all:  The concept of presence. Journal of Computer-Mediated Communication, 3(2).

Muyelle, S., Moenaert, R., & Despontin, M. (2004). The conceptualization and empirical validation of web site user satisfaction. Information & Management, 41, 543-560.

Park, C.-H., & Kim, Y.-G. (2003). Identifying key factors affecting consumer purchase behavior in an online shopping context. International Journal of Retail & Distribution Management, 31(1), 16-29.

Park, Y., & Gretzel, U. (2007). Success factors for destination marketing websites: A qualitative meta-analysis. Journal of Travel Research, 46(1), 46-63.

Ranganathan, C., & Ganapathy, S. (2002). Key dimensions of business-to-consumer web sites. Information & Management, 39(6), 457.

Rogers, E. (2003). Diffusion of innovations (5th ed.).  New York, NY: Free Press.

Torkzadeh, G., & Dhillon, G. (2002). Measuring factors that influence the success of Internet commerce. Information System Research, 13(2), 187-2

van Iwaarden, J., van der Wiele, T., Ball, L., & Millen, K. (2004). Perceptions about the quality of websites: A survey amongst students at Northeastern University and Erasmus University. Information & Management, 41, 947-959.

Venkatesh, V., & Davis, F. D. (2000). A theoretical extension of the Technology Acceptance Model: Four longitudinal field studies. Management Science, 46(2), 186-204.

Venkatesh, V., Morris, M., Davis, G., & Davis, F. (2003). User acceptance of information technology: Toward a unified view. MIS Quarterly, 27(3), 425-478.

Wang, Y. D., & Emurian, H. H. (2005). An overview of online trust: Concepts, elements, and implications. Computers in Human Behavior, 21, 105-125.

Wolfinbarger, M., & Grilly, C. (2003). eTail Q: Dimensionalizing, measuring and predicting etail quality. Journal of Retailing, 79(3), 183-198.

Yen, H. R. (2005). An attribute-based model of quality satisfaction for Internet self-service technology. The Service Industries Journal, 25(5), 641-659.

Zhang, P., & von Dran, G. (2001-2002). User expectations and ranking of quality factors in different website domains. International Journal of Electronic Commerce, 6(3), 9-33.

Table 1
VISA Features Organized by Factor

*e = eigenvalue; % var. = percent of total variance