The heterogeneous impact of sentiment embedded in review on review helpfulness

Abstract

Recently, understanding the context of the review system is getting attention because of the indecisive role of a sentiment of Review on Review helpfulness. We suggest that consumers form their initial belief based on Prospect theory and find Review helpful if the sentiment of that Review supports their initial belief building on confirmation bias. We used Amazon book data and used Tobit regression and Zero-inflated negative binomial regression to support our hypotheses. The findings of this study show that in the product with a high average rating and low rating variance / low average rating and high rating variance, positive sentiment from review increases votes from other consumers. Our study is the first to examine the role of rating statistics (average rating/rating variance) with a sentiment of Review how those affects Review helpfulness based on incorporating Prospect theory and Confirmation bias.