Measuring differential item and test functioning across academic disciplines

Language Testing in Asia

Table 4 Item score significant differences across major*

*Item*	*Majors*	*DIF Contrast*	*Joint S.E.*	*Welch t*	*d.f*	*Prob*
2	AD	-7.25	3.16	-2.3	154	0.0229
5	AB	-12.08	4.4	-2.75	32	0.0072
	BD	12.23	4.77	2.56	99	0.0118
17	AE	-11.62	5.46	-2.13	32	0.0411
	BE	-12.71	5.97	-2.13	39	0.0395
20	CE	-14.13	6.21	-2.28	39	0.0285
23	AE	19.03	6.75	2.82	29	0.0086
	BE	16.78	7.17	2.34	35	0.0251
	CE	19.17	7.20	2.66	36	0.0115
	DE	18.54	6.99	2.65	33	0.0122
29	AC	9.17	3.76	2.44	103	0.165
	DC	9.61	4.17	2.3	105	0.0232
30	AE	-12.04	5.21	-2.31	34	0.0269
32	AD	10.99	5.23	2.1	129	0.0377
34	AC	8.28	3.59	2.31	106	0.0229
35	AD	-7.54	3.61	-2.09	106	0.0393
	CA	7.54	3.62	2.09	106	0.0393
41	CD	8.81	4.04	2.18	108	0.0315
	DA	-6.80	3.24	-2.1	151	0.0372
48	AD	-6.69	3.18	-2.1	152	0.0371
	AE	-12.9	5.46	-2.36	32	0.0244
49	AB	8.69	3.75	2.31	103	0.0226

*The DIF CONTRAST measure illustrates the difference between the item measures, or the difference in difficulty of the item between the two groups (a difference of at least 0.5 logits is required for the DIF to be noticeable). A positive DIF contrast represents that the item was more difficult for the CLASS (major) that is listed on the left whereas a negative DIF contrast demonstrates that the item was easier for the major on the left. The PROB illustrates the probability of observing this amount of contrast by chance. The JOINT S.E. is the standard error of the DIF CONTRAST, the Welch t gives the DIF significance as a t-statistic.