The impact of user personality traits on word of mouth: Text-mining social media platforms

摘要

Online consumer choices can be vastly influenced by word of mouth (WOM) through online forums and communities, and post-purchase social shares. Social media also provides opportunity to gain deeper insights into user characteristics that impact effectiveness of WOM from the observable (e.g., sex, age and relationship status) to the latent space of characteristics (e.g., personality). Based on theories in psychology and social sciences, the article examines whether personality traits of social media users attenuate or accentuate WOM effectiveness. Using big data and machine-learning techniques to extract information from unstructured textual content, the study examines whether and how latent personality characteristics of a user affect purchases of actual products made by social media peers after exposure to WOM messages. A quasi-experimental setting is used where exposure to such WOM instances is independent of characteristics of the content or source and recipient of WOM messages. Results show that exposure to WOM messages from similar users in terms of personality, rather than dissimilar users, increases the likelihood of a post purchase by 47.58%. WOM originating from users who exhibit high levels of agreeableness, conscientiousness and openness is more likely to be effective, whereas for users with low levels of conscientiousness or agreeableness, the opposite effect is more likely. Introverted users were found to be more susceptible to WOM, in contrast to extroverted users. WOM originating from users with low levels of emotional range affects similar users, whereas for high levels of emotional range, increased similarity usually has the opposite effect. These results were found to be robust to a wide variety of alternative econometric specifications and additional controls. In terms of economic importance, it was found that a WOM message from an extroverted user to an introverted peer increases purchase likelihood by 71.28%.