Assessment of Differential Item Functioning in Health-Related Outcomes: A Simulation and Empirical Analysis with Hierarchical Polytomous Data
Table 3
Summary results from the HOLR method to identify differential item functioning on PedsQL TM 4.0 questionnaire.
Gender
Type of school
β
Chi-square
value
β
Chi-square
value
Physical health
0.463
3.111
0.08
0.247
1.07
0.3
0.476
8.02
0.004
0.025
0.021
0.9
0.656
10.234
0.001
−0.165
0.478
0.5
0.817
16.309
<0.001
0.327
1.707
0.2
0.811
2.422
0.1
1.115
7.488
0.006
0.237
2.179
0.1
−0.08
0.26
0.6
0.255
1.423
0.2
−0.33
2.49
0.1
0.164
0.767
0.4
−0.177
0.889
0.3
Emotional functioning
0.468
7.508
0.006
0.188
1.075
0.3
0.688
10.183
0.001
−0.452
3.67
0.055
0.092
0.164
0.7
−0.298
2.598
0.1
−0.421
3.547
0.06
−0.587
8.273
0.004
0.029
0.002
0.97
−0.415
2.977
0.08
Social functioning
−0.286
1.572
0.2
−0.208
0.819
0.4
−0.892
10.795
0.001
0.312
1.047
0.3
−1.307
21.365
<0.001
0.298
0.628
0.4
−0.449
5.836
0.01
0.07
0.144
0.7
−0.567
4.735
0.02
0.54
4.09
0.04
School functioning
−0.518
4.748
0.02
−0.282
1.221
0.3
−0.137
0.691
0.4
−0.517
10.264
0.001
−0.939
11.538
<0.001
−0.482
2.35
0.1
−0.004
0.004
0.9
0.517
4.089
0.04
−0.194
0.755
0.4
0.334
2.14
0.14
Bold numbers represent the value of value for items showing uniform DIF; Chi-square is the value of the difference in −2 log-likelihood of models 1 and 2, for testing uniform DIF.