2022 Volume 49 Issue 1 Pages 29-42
Although past studies have demonstrated the significant impact of emotional expression on the helpfulness of online reviews, the explicit effects of emotion require further exploration. This study fills a gap in the existing literature by uncovering, in detail, how emotions affect review helpfulness. We considered whether the influence of emotion is dependent on varying review content and customer uncertainties about a product. We used Latent Dirichlet Allocation, one of natural language processing (NLP) techniques, to extract review content and we used product pricing plans to represent product uncertainties. We applied a regression method to analyze 268,293 (English only) online reviews of 2,832 products to explore further the relationship between emotions and review helpfulness. Our findings suggest that 1) the impact of emotions on perceptions of helpful reviews depends strongly on the review content that emotions interact with, and 2) similar emotional dimensions have different impacts when uncertainty about a product differs. We present the effects of each emotion in detail.
