Flow in Individual Web Sites: Model Estimation and Cross-Cultural Validation

EXTENDED ABSTRACT - The Internet has become a medium used by consumers worldwide to search for information, to make purchases, or merely to seek entertainment. However, research on what makes a web site effective and appealing to customers is still in its infancy. Few empirical investigations have examined the factors that may lead to an optimal navigation experience for site visitors. Little is known about whether and when consumers’ navigation experiences when visiting a web site will translate into positive marketing outcomes.

Citation:

David Luna, Laura A. Peracchio, and Maria D. de Juan (2003) ,"Flow in Individual Web Sites: Model Estimation and Cross-Cultural Validation", in NA - Advances in Consumer Research Volume 30, eds. Punam Anand Keller and Dennis W. Rook, Valdosta, GA : Association for Consumer Research, Pages: 280-281.

Advances in Consumer Research Volume 30, 2003     Pages 280-281

FLOW IN INDIVIDUAL WEB SITES: MODEL ESTIMATION AND CROSS-CULTURAL VALIDATION

David Luna, University of Wisconsin-Whitewater

Laura A. Peracchio, University of Wisconsin-Milwaukee

Maria D. de Juan, Universidad de Alicante, Spain

EXTENDED ABSTRACT -

The Internet has become a medium used by consumers worldwide to search for information, to make purchases, or merely to seek entertainment. However, research on what makes a web site effective and appealing to customers is still in its infancy. Few empirical investigations have examined the factors that may lead to an optimal navigation experience for site visitors. Little is known about whether and when consumers’ navigation experiences when visiting a web site will translate into positive marketing outcomes.

In the present research, we operationalize and extend previous frameworks (Hoffman and Novak 1996; Novak, Hoffman and Yung 2000) modeling consumers’ optimal network navigation experience or flow on the world wide web. Our research is unique in that we create a model of flow applicable to the navigation of a specific web site. We hypothesize that a web site which facilitates achieving an optimal navigation experience may result in desired consequences for e-marketers.

Our model of flow for a specific web site incorporates constructs examined in the existing literature and includes others previously unexplored in the context of web site navigation. In particular, flow is modeled as a function of consumers’ perceptions of two site characteristics, interactivity and challenge, and two psychological constructs, focused attention and attitude toward the site. Interactivity, challenge, and focused attention have been components in previous web navigation models of flow (Hoffman and Novak 1996; Novak, Hoffman and Yung 2000). Our model introduces attitude toward the site as an antecedent of flow that mediates the impact of those three components on flow for a specific web site. In our model, the experience of flow is expected to result in greater intention to purchase products from the site and to revisit the site.

One of the characteristics of the internet that make it a unique medium is its global reach. Individuals all over the world can access web sites regardless of where they are hosted. It is of considerable importance, therefore, to examine whether similar models of web site navigation and effectiveness can be applied to consumers from different cultures. Consequently, our model of flow will be tested with three independent samples to examine its cross-cultural validityBtwo from Europe and one from the U.S. Also, to examine our model’s validity, we will apply it to different web sites. We begin our investigation by defining the flow construct. Then, we estimate and validate our model of flow for a specific web site.

In this research we construct a model of flow which includes the two types of antecedent components of flow reported in the literature: cognitive and affective. We hypothesize that attitude toward the site mediates the relationship between cognitive factors and flow, which is itself an experiential, or behavioral, construct (Hoffman and Novak 1996; Privette and Bundrick 1987). Our research explores two consequences of flow that are important to assessing marketing effectiveness: intentions to revisit and to purchase from a site. Our hypothesized model is depicted in Figure 1.

The model fits the data well (c²⁼178.27, d.f.=155, RMSEA=.037, CFI=.98, GFI=.86). Because the GFI and AGFI statistics tend to underestimate the fit of models with smaller samples (Durvasula et al. 1993), we will focus on the RMSEA, CFI and Chi-square statistics in the rest of this paper. Parameter estimates are reported on Table 3.

To validate our model, we collected data from two additional samples. Sample 2 validates our model with the same population as sample 1 but a different web site. Sample 3 validates the model cross-culturally with a different population but the same site as sample 1. The methodology used to validate the model was adapted from Byrne (1998), Durvasula et al. (1993) and Joreskog and Sorbom (1996). Hence, the model we first estimated with sample 1 was estimated again using multiple group procedures. First, we compared the model as it fitted sample 1 and sample 2. Then, we compared the model fit between sample 1 and sample 3. If the results of our comparisons indicate that the model fits the three samples equally well, we will accomplish a validation of our model across cultures and web site designs. In particular, we tested whether (a) the factor structure fits the estimation and validation samples well when estimated conjointly; (b) the covariance structures of each of the two validation samples are equal to the estimation sample, and (c) the estimated structural parameters in the estimation sample are equal to the parameters in each of the validation samples.

The results of our model validation indicate that our structural model applies to two populations and two different sites, since constraining (a), (b), and (c) above did not result in a significantly worse fit that the unconstrained models.

CONCLUSION

Our model of flow highlights the important role of attitude towards the site. We find that A_site mediates a great deal of the influence of interactivity, challenge and focused attention on flow. This finding is of particular relevance considering that prior research on flow in e-commerce had not explored this effect. It is also significant that we found flow to mediate the effect of A_site on intentions variables. Altogether, the successful integration of flow with more traditional measures of marketing effectiveness in one single model may constitute an important step forward in our understanding of the flow construct and its influence on consumer behavior.

FIGURE 1

HYPOTHESIZED CONCEPTUAL MODEL

TABLE 3

ESTIMATION RESULTS

REFERENCES

Byrne, Barbara M. (1998), Structural Equation Modeling with Lisrel, Prelis, and Simplis. Mahwah, NJ: Lawrence Erlbaum Associates.

Durvasula, Srinivas, J. Craig Andrews, Steven Lysonski, and Richard G. Netemeyer (1993), "Assessing the Cross-national Applicability of Consumer Behavior Models: A Model of Attitude toward Advertising in General," Journal of Consumer Research, 19 (March), 626-636.

Hoffman, Donna and Thomas Novak (1996), "Marketing in Hypermedia Computer-Mediated Environments: Conceptual Foundations," Journal of Marketing, 60 (July), 50-68.

Joreskog, Karl and Dag Sorbom (1996), Lisrel 8: Users’ Reference Guide. Chicago: Scientific Software International.

Novak, Thomas, Donna Hoffman and Yiu-Fai Yung (2000), "Measuring the Customer Experience in Online Environments: A Structural Modeling Approach," Marketing Science, 19 (1), 22-42.

Privette, Gayle and Charles Bundrick (1987), "Measurement of Experience: Construct and Content Validity of the Experience Questionnaire," Perceptual and Motor Skills, 65, 315-332.

----------------------------------------