Predictive modeling of audience satisfaction in Chinese web dramas : A big data analytics approach

Chinese web dramas have grown rapidly as a media sector, but some shows still face low audience satisfaction. This challenge is compounded by the underutilization of big data for audience satisfaction, forecasting among Chinese web drama stakeholders, exacerbating operational uncertainties and market risks. Despite this, prior studies on Chinese web drama audience satisfaction have not developed data driven, predictive models. This study aims to develop an inclusive predictive model of audience satisfaction in Chinese web dramas, incorporating five constructs: viewing motivation, audience satisfaction, web drama exposure, audience social media engagement, and social media community size. The model is grounded in Uses and Gratifications Theory (UGT) and Channel Complementarity Theory (CCT) and operationalized through big data analytics. Eleven hypotheses were proposed for validation. Stratified systematic sampling was employed to select 288 Chinese web dramas, from which 16,057,255 unstructured social media entries and 3,744 structured records were extracted via big data analytics. Sentiment analysis was applied to convert the unstructured comments into structured datasets, and all hypotheses were tested using 6,048 final structured records via SPSS 29.0 and Smart PLS 4.1. The study confirmed six positive direct relationships among the studied variables. Web drama exposure and social media engagement partially mediated the link between viewing motivation and satisfaction, while social media community size positively moderated the relationships between viewing motivation and satisfaction, and between web drama exposure and social media engagement. This study contributes to understanding web drama audience satisfaction and its antecedents. Theoretically, it enriches UGT and CCT by validating and extending their application to develop the predictive modelling in the context of Chinese web dramas. Methodologically, it employs a novel big data analytics approach integrating sentiment analysis. Practically, it provides actionable insights for web drama stakeholders and governmental regulatory agencies