Plenty of attitude: evaluating measures of attitude toward the site

Purpose - Researchers often attempt to assess how different features and content will improve the experience of web site users. One assessment technique is to measure the attitude toward the site. A common version of this measure is the Chen and Wells attitude toward the site scale. The purpose of this paper is to determine if there is a difference in performance between that scale and the less used Bruner and Kumar scale so that researchers might use the better of the two related, but different, published scales. Design/methodology/approach - Analysis is done on survey data from an experiment utilizing three different experimental groups that all completed surveys with both the Chen and Wells and the Bruner and Kumar attitude toward the site scales. Scales are assessed for loading and reliability, as well as measures compared for equivalence within groups and used within partial least squares (PLS) models to compare overall model fit. Findings - In all tests, the Bruner and Kumar scale is better than, or equivalent to, the Chen and Wells scale in each comparison. Research limitations/implications - The research implication is that the Bruner and Kumar scale would be a better choice when selecting scales for future research projects. Originality/value - Although Bruner and Kumar had previously performed comparisons of the two scales, in a follow-up article, this is the first paper to compare the two scales between three different groups and demonstrate how the two different scales would perform within the same conceptual model using PLS structural equation modeling. It will help researchers select the best scale for attitude toward the site.