Understanding the roles of forum search aids in consumer information search and response to online product reviews
Date of Issue2013
College of Business (Nanyang Business School)
In online product review forums, consumers often encounter a vast number of reviews under a threaded discussion. As processing all the reviews is both nearly impossible and unlikely to be worth the effort and time, it gives rise to important concerns about how consumers select particular reviews to process and how they process them. Their information search activity, which indicates such selection and processing of reviews, would in turn influence their response to the review information in terms of attitude towards the product under review, evaluation confidence and purchase intention regarding the product, etc. To address this issue, I introduce and test a behavioral model of consumer online review search that incorporates consumers’ anchoring bias in information search stage and memory-based decisions during product evaluation stage. I propose that these two mechanisms can manifest themselves in the types of reviews inspected as well as depth of processing, and in the decision making process respectively. Forum search aids (summary statistics) are proposed to influence consumer information search behaviors in the selection and processing of different reviews and thereby the response to the product under review. Specifically, online reviews are classified as reviews that are consistent with an average rating (consistent reviews) and those that deviate from the average rating (deviant reviews). The search aids are expected to exert different effects on the search variables corresponding to these two types of reviews. The results from an in-depth interview and an eye-tracking experiment support many of the proposed relationships, and, in particular, confirm the mediating role of search variables. In the search stage, the results show that presence of an average rating has a positive effect on the two major components of search time, namely, the preview ratio and the average search depth of consistent reviews. Presence of an average rating also has a positive effect on the average search depth of deviant reviews, owing to the distinctiveness of these reviews. Additionally, the effect of distribution presence on the preview ratio of consistent reviews is negative, on condition that the average rating is provided. In the evaluation stage, search time spent on both consistent and deviant reviews significantly affect their product attitude. Across the two stages, provision of an average rating affects product attitude through search variables only when the product rating distribution is absent and when the average rating is low. On the other hand, the provision of a product rating distribution shifts the attitude in a direction opposite to the valence of the average rating through the search variables. The findings provide insights into the complex relationships between search aids, search variables, product attitude and purchase intention during both search and evaluation stages. Finally, managerial implications of this work are described, and potential extensions of the joint model are suggested.