Test-level and Item-level Model Fit Comparison of General vs. Specific Diagnostic Classification Models: A Case of True DCM

Language Testing in Asia

Table 4 Absolute item-level fit indices

Model		Mean[stats]	Max[stats]	Max[z.stats]	p value	Adj. p value
G-DINA	Proportion correct	0.0012	0.0041	0.3764	0.7066	1
	Transformed correlation	0.0258	0. 0928	3.9027	0.0001	0.669
	Log odds ratio	0.1142	0.4625	3.7646	0.0002	0.1173
LLM	Proportion correct	0.0013	0.0032	0.2732	0.7847	1
	Transformed correlation	0.0273	0.0983	4.1359	0.000	0.249
	Log odds ratio	0.1210	0.5072	3.9982	0.001	0.449
ACDM	Proportion correct	0.0018	0.0085	0.7535	0.4511	1
	Transformed correlation	0.0283	0.1117	4.6980	0.0000	0.0018
	Log odds ratio	0.1252	0.5093	4.5975	0.0000	0.0030
RRUM	Proportion correct	0.0012	0.0051	0.4499	0.6528	1
	Transformed correlation	0.0290	0.1169	4.9177	0.0000	0.0062
	Log odds ratio	0.1275	0.5283	4.7776	0.0000	0.0012
DINA	Proportion correct	0.0012	0.0030	0.2570	0.7972	1
	Transformed correlation	0.0390	0.1485	6.2483	0.0000	0
	Log odds ratio	0.1722	0.7906	5.6413	0.0000	0
DINO	Proportion correct	0.0012	0.0035	0.3092	0.7571	1
	Transformed correlation	0.00403	0.1603	6.7426	0.0000	0
	Log odds ratio	0.1787	0.8604	6.2667	0.0000	0