Enhancing the Helpfulness of Online Consumer Reviews: The Role of Latent (Content) Factors

Past empirical studies have analysed the influence of manifest online review content factors and the reviewer-related factors on online review helpfulness. However, the influence of latent content factors, which are implied from the text and that result in the differential helpfulness perceptions of review receivers, have been ignored. Hence, using the lens of the Elaboration Likelihood Model (ELM), we develop a comprehensive model to study the influence of content- and reviewer-related factors on review helpfulness. This study not only includes the manifest content-related and reviewer-related factors, but also the latent content factors consisting of argument quality (comprehensiveness, clarity, readability and relevance) and message valence. The study initially employs a manual content analysis to analyse the argument quality of ~500 TripAdvisor reviews (Study 1). Subsequently, model testing techniques are used to study the holistic and relative influence of these different factors on review helpfulness. In the validation study (Study 2), Machine Learning and Natural Language Processing techniques are used to extract latent content information and test the above model with 50,000 reviews from Yelp.com. The results show that latent review content variables like argument quality and valence influence the helpfulness of the reviews better and beyond the previously studied, manifest review content- and reviewer-related factors.