Why Do People Rate? Theory and Evidence on Online Ratings

“The average review score can vary considerably for products that have otherwise similar characteristics, and may be the only insight consumers have into a product's unobservable qualities before they buy. As Chevalier and Mayzlin (2006) demonstrate, ratings can significantly influence buyers' behavior and have a substantial impact on the success or failure of a product. But why are ratings given in the first place? Are people taking time to give these ratings in order to help their anonymous fellow shoppers, or are they writing out of gratitude or anger that they feel towards online merchants? Are raters equally likely to evaluate all products, or do they speak up only if they have a strong opinion? This paper examines possible motivations for the provision of numerical ratings in a theoretical framework and then isolates those motivations in an experimental setting.”

“To preview the results, I find evidence that consumers are motivated by concern for both buyers and sellers when they decide to rate products. Making rating less attractive through the introduction of a small cost has a large effect on the volume and distribution of ratings. Ratings in the presence of a cost take on a U-shaped distribution, which can lead to average ratings that are not representative of true quality. A possible solution to this problem is to provide small discounts to consumers who provide ratings, thereby compensating for any inconveniences or opportunity costs associated with rating products.”