| HBS Working Paper Series
The Benefits of Product Quality Disagreement - When Divergent Reviews Drive Online Word of Mouth
A consumer's decision to leave an online product review may not only depend on satisfaction with the product but may depend on other product characteristics and on social effects, including the disagreement among previously posted reviews. We develop and test a theoretical model that extends and advances the disparate literature about online word of mouth (WOM) by accounting for social effects (particularly disagreement), product effects, and their interaction, on consumers' propensity to write online product reviews. To quantify disagreement, we introduce a new measure based on an expectation-maximization algorithm for finite mixture models as a summary statistic for the mostly non-normally distributed review ratings. In addition to numerical star-ratings, we use sentiment analysis and automated nonparametric content analysis of review texts to measure disagreement in consumer opinions. We test our model on nearly 280,000 reviews for 433 movies and find the propensity to review is higher when there is high disagreement about the movie's quality. We also find that while high disagreement generates a 35% increase in WOM for niche products, it leads to an 8% decrease in WOM for hit products. Finally, we find that the positive effect of disagreement is amplified by the length of the textual reviews but dissipates over time.
Keywords: online word of mouth;
online product reviews;