Assessment of Differential Item Functioning in Health-Related Outcomes: A Simulation and Empirical Analysis with Hierarchical Polytomous Data

Table 1

Detection power for within-group variable by type of analysis and manipulated factors.

Variable			Item discrimination
			Low (0.5–0.99)				High (1.5–2)
ρ	Cluster		DIF = 0.4		DIF = 0.8		DIF = 0.4		DIF = 0.8
			OLR	HOLR	OLR	HOLR	OLR	HOLR	OLR	HOLR
0.05	50	5	0.31	0.316	0.807	0.811	0.775	0.778	1	1
		10	0.559	0.564	0.986	0.987	0.976	0.976	1	1
		20	0.83	0.831	1	1	1	1	1	1
	100	5	0.519	0.523	0.977	0.977	0.966	0.966	1	1
		10	0.85	0.851	1	1	1	1	1	1
		20	0.984	0.984	1	1	1	1	1	1
	200	5	0.809	0.812	1	1	0.999	0.999	1	1
		10	0.988	0.989	1	1	1	1	1	1
		20	1	1	1	1	1	1	1	1
0.25	50	5	0.303	0.312	0.795	0.804	0.771	0.78	1	1
		10	0.556	0.558	0.988	0.988	0.977	0.977	1	1
		20	0.837	0.839	1	1	1	1	1	1
	100	5	0.533	0.54	0.976	0.977	0.971	0.972	1	1
		10	0.853	0.855	1	1	1	1	1	1
		20	0.981	0.981	1	1	1	1	1	1
	200	5	0.807	0.811	1	1	1	1	1	1
		10	0.99	0.99	1	1	1	1	1	1
		20	1	1	1	1	1	1	1	1
0.45	50	5	0.308	0.312	0.803	0.81	0.78	0.784	1	1
		10	0.558	0.561	0.989	0.99	0.975	0.977	1	1
		20	0.832	0.835	1	1	1	1	1	1
	100	5	0.533	0.537	0.98	0.98	0.971	0.972	1	1
		10	0.845	0.848	1	1	1	1	1	1
		20	0.984	0.984	1	1	1	1	1	1
	200	5	0.809	0.814	1	1	1	1	1	1
		10	0.991	0.991	1	1	1	1	1	1
		20	1	1	1	1	1	1	1	1