Preference instability, consumption and online rating behavior

Abstract

In this paper we examine how individual level heterogeneities in terms of preference instability influence various online decisions (product choice, rating incidence and rating evaluation). Moreover, we investigate how individuals with different levels of preference instability differ in the way their online decisions are affected by the different aggregate measures of rating environment (volume, valance and variance). We empirically examine their dynamic three stage decision model in a controlled laboratory setting and find that individuals with greater (lower) preference instability depend more (less) on volume (valence) as a quality indicator while making their choice decisions. This finding offers a potential explanation for some of the apparent anomalies in terms of the effect of different online rating characteristics on product demand that have been reported in earlier literature. We also show that individuals with different level of preference instability differ in terms of their susceptibility toward disconfirmation. This difference in turn results in differences in online post-consumption opinion contribution. The results also indicate that choice propensity is strongly related to individual level rating incidence and evaluation propensities. Finally they find an evidence of strong temporal persistence for both choice and incidence decisions.