The role of entropy of review text sentiments on online WOM and movie box office sales

Sentiments from online word-of-mouth (WOM) are often controversial, since individuals have different preferences toward the same products. Past studies have focused on online WOM effects by measuring WOM volume and valence. However, few studies have investigated how the entropy of the review text sentiment influences the relationship between online WOM and product sales. As WOM valence and volume are usually provided at an aggregated level, consumers often do not have enough information to make a decision. In this case, reading online review text has become an important process for consumers to make purchasing decisions. However, when consumers encounter too many positive review texts, they may doubt the credibility of online WOM itself. Thus, we analyzed the entropy of the review text sentiments by conducting text-mining techniques. We classified review text sentiment into positive, negative, and neutral categories and created an entropy variable. A high level of entropy in review texts indicates that sentiment from review texts are equally distributed, but not biased, towards positive or negative sentiment. We estimated our research model with the entropy variable in a panel dataset for 204 movies over a half-year period. The results suggest that the entropy level in the review texts has a positive moderating impact on the relationship between WOM (e.g., valence and volume) and movie box office sales. The findings imply that deleting negative reviews to enhance product sales may not help online retailers or related parties.