Informace o publikaci
Latent vs. Network Psychometric Validation : The Example of Cross-National Evaluation of the Internet Gaming Disorder Scale in European Adolescents
| Autoři | |
|---|---|
| Rok publikování | 2025 |
| Druh | Další prezentace na konferencích |
| Fakulta / Pracoviště MU | |
| Citace | |
| Přiložené soubory | |
| Popis | Effective measurement practices are critical in addiction-like disorder research to avoid overpathologization or underdetection of health risks. While factor analytic (FA) approaches dominate questionnaire validation, their causal assumptions are rarely discussed. Additionally, model fit evaluations often rely on Hu and Bentler's (1999) guidelines, which may not generalize beyond their original simulation conditions. This study applied both factor and network psychometric approaches to validate the 9-item Internet Gaming Disorder Scale (Lemmens et al., 2015) using data from the Health Behavior in School-aged Children project (N = 44,126; adolescents aged 11–15 from 12 European regions). The approaches were compared for utility, theoretical soundness, and empirical outcomes. Findings revealed that: (1) conventional fit index cut-offs could yield divergent validity interpretations; (2) the network approach provided nuanced insights into invariance patterns across data; and (3) dynamic fit index cut-offs (e.g., McNeish & Wolf, 2023) offered a more robust assessment of model fit. Additionally, exploratory graph analysis and network trees were critically evaluated for their benefits and limitations. Validation studies should align causal theoretical assumptions with appropriate statistical methods. Advances in psychometrics provide refined tools that improve measurement practices. These findings contribute to ongoing debates on the conceptualization and measurement of gaming and other addiction-related disorders. |
| Související projekty: |