How does content and sentiment coherence influence online discussion?

Abstract

While online discussion has been extensively studied in the previous literature, the role of information coherence along the course of discussion is less being explored. In this study, we investigate the influence of information coherence in online discussion platforms with a focus on the sequential patterns of user posts. We measure the textual information by two dimensions: content and sentiment. Using the data from two popular online automobile discussion platforms, we empirically estimate the dependency of content and sentiment coherence among user posts and evaluate the effect of information coherence on user participation in the online discussion. The empirical evidence sheds light on the heterogeneity of the coherence influence and reveal the synergetic relationships among information coherence, dependency, and topics in the online discussion platforms. Our findings provide insights into online discussion platforms and business managers on their strategic designs to keep users engaged and facilitate digital collaboration.