Tweeting with the Stars: Automated Text Analysis of the Effect of Celebrity Social Media Communications on Consumer Word of Mouth

Abstract

Prior research has focused on analyzing the content and intent of celebrity social media communications. By observing that the linguistic style of such celebrity communications drives consumer word of mouth, the main goal of the current research is to broaden this limited perspective. An automated text analysis of narrative/analytical, internally/externally focused, and negative/positive emotional styles in tweets by celebrity chefs, personal trainers, and fashion bloggers was conducted to this effect. The findings are threefold. First, across celebrity categories externally focused, narrative styles are more effective in terms of word of mouth. Second, emotional styles are not effective. Third, angry outbursts are an exception; they are effective drivers of word of mouth for personal trainers. As such, this research furthers scholarly and practitioner understanding of the state-the-art of celebrity social media communication: the effect of tweets' linguistic styles on consumer word of mouth.