Understanding the Order Effect of Online Reviews: A Text Mining Perspective

Online reviews are aimed to help prospective buyers in their decision-making. While prior research has focused on the economic impact of ratings, review volume, helpfulness and sentiments, open research questions remain regarding the evolution of text attributes associated with online reviews. Using a large dataset, we extract sentiment intensity, along with novel attributes - product usage contexts and product features - present in each online review and analyze their pattern over the temporal order of the reviews. Results indicate that sentiment intensity as well as the number of product features and usage contexts diminish with respect to the increase in review order, suggesting that earlier reviews tend to have more information for prospective customers. However, the declining trend of sentiment intensity is less when reviews mention a higher number of product features and usage contexts. These findings contribute to the literature while having key practical implications for e-commerce websites, retailers, and customers.

Automated summary

The study reveals that sentiment intensity, number of product features, and usage contexts decrease as the review order increases, indicating that earlier reviews are more informative for potential customers.