Factor structure, reliability and criterion-related validity of the English version of the Problematic Series Watching Scale.

Abstract

Psychological research in the past decade has investigated the psychosocial implications of problematic use of on-demand online video streaming services, particularly series watching. Yet, a psychometric measure of problematic series watching in English is not available. The present study aimed to test the factor structure, reliability and criterion-related validity of the English version of the Problematic Series Watching Scale, a six-item self-report assessing problematic series watching, based on the biopsychosocial components model of addiction. Participants were recruited from two UK university student samples. Study 1 (n = 333) comprised confirmatory factor analysis, reliability tests and item response theory analyses to test the original unidimensional model and investigate each item's levels of discrimination and information. Study 2 (n = 209) comprised correlation analyses to test the criterion-related validity of the scale. There was a good fit of the theoretical model of the scale to the data (Comparative Fit Index = 0.998, Root Mean Square Error of Approximation = 0.024 [90% CI 0.000-0.093], Standardised Root Mean square Residual = 0.048), satisfactory reliability (ω = 0.79) and item levels of discrimination and information. The scale positively correlated with time spent watching series (rs = 0.26, P < 0.001) and negative affect (rs = 0.43, P < 0.001), and correlated negatively with positive affect (rs = -0.12, P > 0.05), mental well-being (rs = -0.25, P < 0.001) and sleep quality (rs = -0.14, P < 0.05). Results are discussed in relation to the ongoing debate on binge watching and series watching in the context of positive reinforcement versus problematic behaviour.