Understanding and predicting individual retweeting behavior: Receiver perspectives

Abstract

Individual retweeting behavior plays a fundamental role in information propagation on social media. While researchers have demonstrated the benefits of individual retweeting behavior to online retailers, a largely uninvestigated issue is the mechanism of individual retweeting behavior. Drawing on the social communication theory, we develop an integrated conceptual framework to investigate this issue. An analysis of a panel dataset, which consists of 440 valid users from Twitter.com across six years, indicates that all the major components of a social communication process have significant impacts on individual retweeting decision: (1) information source (e.g., source trustworthiness), (2) information richness of the stimuli (3) information receiver's topical preference, (4) relationship between the source and the receiver (e.g., social tie strength), and (5) contextual factors (e.g., bandwagon effect). Among them, social tie strength and topical relevance with the receiver are the most important factors, followed by information richness of the stimuli and bandwagon effect. However, information source-related factors have trivial impacts on individual retweeting decision. We reevaluate the ranking using multiple feature selection methods and confirm the ranking result. Prediction performance is carried out between the salient features and the full feature set. Finally, we discuss important implications for practice and future research directions.