Analyzing Online Reviews Through the Lens of Speech Act Theory: Implications for Consumer Sentiment Analysis

Sentiment Analysis, the process of automatically distilling sentiment from text, is often used in consumer research to assess online consumer evaluations by counting positive and negative words. However, more granular sentiment expressions—such as activation levels, implicit meanings, and patterns of sentiment across sentences (e.g., in an online review)—are relatively poorly understood. Drawing on Speech Act Theory, this study goes beyond positive and negative word counts to examine the effects of finer-grained explicit and implicit sentiment expressions, within and across sentences. We demonstrate the significance of sentiment force levels, implicit sentiment expressions, and discourse patterns on overall consumer sentiment (i.e., star ratings) in an empirical study using online consumer reviews. Two follow-up studies enhance the relevance and generalizability of the findings. As this study confirms, both implicit and explicit expressions as well as discourse patterns allude to consumers’ sentiments. These expressions also drive actual purchasing behavior; and are generalizable to other social media contexts such as Twitter and Facebook. These findings contribute to research on consumer sentiment analysis by offering an in-depth understanding of how the unique speech act features constitute consumers’ sentiment expressions and their implications.