Table 2. Conditional likelihood ratio test of overall model fit and global DIF.
Sample 1 | Sample 2 | ||||||
---|---|---|---|---|---|---|---|
Model | Homogeneity | DIF by sex | DIF by education | DIF by employment status | Homogeneity | DIF by education | DIF by employment status |
RM | χ2(15) = 69.6, p<0.001 | χ2(15) = 15.8, p = 0.392 | χ2(15) = 13.6, p = 0.554 | χ2(15) = 31.2, p = 0.008 | χ2(15) = 25.1, p = 0.049 | χ2(15) = 19.3, p = 0.200 | χ2(15) = 12.8, p = 0.621 |
GLLRM | χ2(38) = 52.7, p = 0.057 | χ2(38) = 50.4, p = 0.086 | χ2(38) = 39.5, p = 0.397 | χ2(30) = 43.9, p = 0.048 | χ2(25) = 22.2, p = 0.625 | χ2(25) = 10.3, p = 0.996 | χ2(25) = 16.4, p = 0.903 |
Note. The subgroups were defined according to lower and higher scores (i.e. homogeneity) to evaluate overall model fit; and according to sex (Men; Women), education (Up to high school; TAFE or University) and employment status (Not employed; Employed) to evaluate overall DIF.