Table 2.

Average mean-squared errors of the estimated parameters in the simulation studies.

			Small DIF			Large DIF
			High	Medium	Low	High	Medium	Low
$N=500$	Small $d_j$	$\mathit{a}$	0.0482	0.0485	0.0485	0.0502	0.0490	0.0486
		$\mathit{d}$	0.0316	0.0317	0.0318	0.0317	0.0317	0.0316
		$\mathit{\gamma }$	0.0614	0.0612	0.0609	0.0670	0.0650	0.0623
		$\beta$	0.0010	0.0010	0.0010	0.0011	0.0011	0.0010
		$\sigma$	0.0016	0.0017	0.0016	0.0017	0.0017	0.0018
	Large $d_j$	$\mathit{a}$	0.0562	0.0552	0.0552	0.0589	0.0575	0.0560
		$\mathit{d}$	0.0467	0.0467	0.0470	0.0475	0.0476	0.0476
		$\mathit{\gamma }$	0.0873	0.0854	0.0834	0.1089	0.1009	0.0903
		$\beta$	0.0013	0.0012	0.0012	0.0011	0.0012	0.0013
		$\sigma$	0.0014	0.0014	0.0015	0.0015	0.0015	0.0015
$N=1000$	Small $d_j$	$\mathit{a}$	0.0227	0.0222	0.0222	0.0223	0.0221	0.0222
		$\mathit{d}$	0.0145	0.0145	0.0145	0.0145	0.0145	0.0145
		$\mathit{\gamma }$	0.0291	0.0289	0.0287	0.0335	0.0320	0.0298
		$\beta$	0.0004	0.0004	0.0004	0.0005	0.0005	0.0004
		$\sigma$	0.0004	0.0005	0.0005	0.0005	0.0005	0.0005
	Large $d_j$	$\mathit{a}$	0.0263	0.0261	0.0262	0.0270	0.0267	0.0264
		$\mathit{d}$	0.0223	0.0224	0.0225	0.0227	0.0224	0.0226
		$\mathit{\gamma }$	0.0412	0.0401	0.0392	0.0500	0.0461	0.0418
		$\beta$	0.0005	0.0005	0.0005	0.0006	0.0005	0.0005
		$\sigma$	0.0005	0.0005	0.0005	0.0006	0.0005	0.0005

Mean-squared errors are first evaluated by averaging out of 100 replications and then averaged across 25 items to obtain the average mean-squared errors for $\mathit{a}$, $\mathit{d}$ and $\mathit{\gamma }$. The mean-squared errors for $\beta$ and $\sigma$ are presented