Table 4.
Measurement invariance of the Gaming Disorder Test (GDT) and Gaming Disorder Scale for Adolescents (GADIS-A) across gender (males vs. females) and time spent on gaming (< 2 h daily vs. ≥ 2 h daily)
| GDT | GADIS-A | |||||
|---|---|---|---|---|---|---|
| Across gender | M0 | M1vs.M0 | M2vs.M1 | M0 | M1vs.M0 | M2 vs.M1 |
| χ2(df)/ p-value | 2.170 (4)/0.704 | – | – | 51.023 (54)/0.59 | – | – |
| Δ χ2(Δdf)/ p-value | – | 9.462 (3)/0.024 | 1.152 (3)/0.765 | – | 8.217 (8)/0.413 | 10.17 (8)/0.253 |
| CFI | 1.000 | – | – | 1.000 | – | – |
| ΔCFI | – | -0.008 | 0.003 | – | 0.000 | 0.000 |
| RMSEA | 0.000 | – | – | 0 | – | – |
| ΔRMSEA | – | 0.047 | -0.017 | – | 0.000 | 0.000 |
| SRMR | 0.003 | – | – | 0.056 | – | – |
| ΔSRMR | – | 0.038 | -0.008 | – | 0.005 | -0.002 |
| Across time spent gaming | ||||||
| χ2(df)/ p-value | 1.094 (2)/0.579 | – | – | 31.498 (44)/0.921 | – | – |
| Δ χ2(Δdf) / p-value | – | 6.36 (3)/0.095 | 2.824 (3)/0.420 | – | 5.458 (7)/0.604 | 17.505 (7)/0.014 |
| CFI | 1.000 | – | – | 1.000 | – | – |
| ΔCFI | – | -0.005 | 0.000 | – | 0.000 | 0.000 |
| RMSEA | 0.000 | – | – | 0.000 | – | – |
| ΔRMSEA | – | 0.040 | -0.009 | – | 0.000 | 0.000 |
| SRMR | 0.021 | – | – | 0.042 | – | – |
| ΔSRMR | – | 0.028 | -0.006 | – | 0.004 | 0.002 |
Bold values indicate invariance: ΔCFI > -0.01; ΔRMSEA < 0.015; ΔSRMR < 0.03 (for factor loading) or < 0.01 (for item threshold)
M0 configural model, M1 Model with loadings constrained equal, M2 Model with loadings and thresholds constrained equal
CFI Comparative fit index, RMSEA Root mean square error of approximation, SRMR Standardized root mean square residual