Understanding Generalized Trust in News Media in China

Abstract: Trust in news media (TNM) has continued declining in some countries (e.g., US) while remaining stable or even increasing in others (e.g., China). Scholars are interested in understanding what factors contribute to these differences. A reliable and equivalent measurement is essential for cross-national comparisons of TNM. However, there is currently no instrument available to measure generalized TNM in China. This study examined the psychometric properties of the Generalized Trust in News Media Scale (GTNMS) in China. The study sample comprised 578 participants, including 347 females (60.0%) and 231 males (40.0%). Confirmatory Factor Analysis (CFA), bifactor-CFA, and bifactor Exploratory Structural Equation Modelling (bifactor-ESEM) were adopted to examine the construct validity of the Chinese version of GTNMS. Measurement invariance (MI) and predictive validity of the Chinese version of GTNMS were also examined. The results revealed that the Chinese adaptation of GTNMS demonstrated optimal representation through a bifactor-ESEM, comprising a global factor (G-factor) and four specific factors (S-factors). Both the G-factor and S-factors showed acceptable internal consistency reliability. However, the S-factors retained a lower level of specificity after controlling for the G-factor, which indicated that the GTNMS was essentially unidimensional. The results from the Structural Equation Modelling analysis also revealed that only the G-factor could predict individuals’ news avoidance. In addition, MI of the GTNMS was established across participants using official and commercial media. Overall, the GTNMS is a reliable tool for measuring generalized TNM in China.