Trust, Bias, and Dependency: Predicting Millennial News Consumption in Malaysia

Abstract

Rising scepticism toward mainstream news, combined with an apparent shift toward brand-level trust in specific media platforms, raises an important directionality question for Media System Dependency (MSD) research: does confidence in a particular news brand feed back into audiences' generalised trust in the media as an institution, or does institutional trust operate independently of brand-level experience? This study identified Brand Trust (BT) as the strongest predictor of media dependency among Malaysian millennials during the COVID-19 pandemic, this analysis explores whether Brand Trust, Perceived Media Bias (PMB), and the Degree of Individual Media Dependency (DIMD) function as antecedents of generalised Trust in Media (TM). Using the cross-sectional dataset of 421 Malaysian millennials from the Klang Valley, recruited through purposive and cluster sampling of Facebook and Instagram followers of three major Malaysian news outlets, the structural model was estimated using Partial Least Squares Structural Equation Modelling (PLS-SEM. The structural results showed that Perceived Media Bias significantly predicted both Brand Trust (β = 0.230, p < .001) and Trust in Media (β = 0.150, p = .019), and that media dependency significantly predicted Trust in Media (β = 0.117, p = .024). However, Brand Trust did not significantly predict Trust in Media (β = 0.073, p = .222), and the indirect path from Perceived Media Bias to Trust in Media through Brand Trust was non-significant, indicating an absence of mediation. The reversed model's explanatory and predictive power was modest (R² = .053-.059; f² ≤ 0.056; Q²predict = .035-.043), indicating that, although statistically detectable, these relationships are substantively small. These findings suggest that, within this cross-sectional sample, trust does not flow upward from brand-level experience to generalised institutional trust, reinforcing the directionality of the original model and cautioning researchers against assuming reciprocal trust transfer in crisis-era media consumption; longitudinal or experimental designs are needed to confirm the causal direction of this relationship.