Exploring social media determinants in fostering pro-environmental behavior: insights from social impact theory and the theory of planned behavior

Introduction This study investigates the impact of social media on pro-environmental behavior (PEB) through the lenses of the Theory of Planned Behavior (TPB) and Social Impact Theory. The research aims to elucidate how social media influences Environmental Attitude (EA) and Subjective Norms (SN), and how these factors contribute to Behavioral Intentions (BI) that ultimately affect PEB. Additionally, it examines the moderating effect of Perceived Behavioral Control (PBC) on the relationship between BI and PEB.Methods To explore these relationships, the study employs a dual methodological approach using Variance-Based Structural Equation Modeling (VBSEM) and Artificial Neural Networks (ANN). Data were collected from two distinct samples: 1200 participants from Taiwan for the SEM analysis and 602 respondents for the ANN study. SEM was utilized to explore causal relationships, while ANN was employed to enhance predictive accuracy.Results The SEM analysis reveals that social media significantly affects both EA and SN, except for Social Networking Site Involvement (SNSI), which does not significantly impact EA. Additionally, the findings indicate that BI mediates the relationship between EA and PEB. However, BI does not mediate the SN-PEB relationship, and the link between SN and BI is found to be non-significant. Empirical evidence also suggests that PBC moderates the BI-PEB relationship, with a stronger influence observed under higher levels of PBC and a weaker influence under lower levels.Discussion These results underscore the complex dynamics between social media factors and pro-environmental behavior. The study concludes that while social media plays a significant role in shaping EA and SN, its impact on EA is not mediated by SNSI. Furthermore, PBC significantly moderates the BI-PEB relationship, highlighting its critical role in PEB. The discussion addresses the implications of these findings, acknowledges the limitations encountered, and suggests potential avenues for future research.