Predecting Sustainable Boycott Intentions through Digital Eco-Activism:SEM & Machine Learning Approach

The present work develops a predictive empirical technique for mapping the behavioral and system architecture of digital marketplace resistance against tech providers. Based on the structural framework of the Theory of Planned Behavior (TPB) and Institutional Anomie Theory, the proposed paradigm models the spread of corporate product-sustainability failures (PSF) – from environmental misconduct to ethical exploitation and economic green washing – as demand-side consumer shocks among hyper-connected Gen Z cohorts. The objective of this study is to provide a two-step analytical approach to unveil the determinants of durable boycott intentions, forecast consumers’ boycott decisions correctly, and thus bridge the conventional intention–behavior gap. This framework combines a variance-based Structural Equation Modelling (SEM) approach to disentangle complex internal mediating mechanisms (value misalignment and eco-ethical outrage) and upper boundary moderation filters (fractured trust filters) with downstream machine learning classification models (SVM, Random Forest) to predict definitive binary anti-consumption choices. The Paper allows multi-product high-tech firms to evaluate market risks and demonstrate that long-term brand equity depends on transparency, sustainable product lifecycles, and real corporate responsibility. This study promotes SDGs 12, 16 and 9 by supporting responsible consumerism, ethical corporate governance and predictive sustainability analytics via an integrated SEM-machine learning framework.