Investigating the psychometric properties of the Qiyas for L1 Arabic language test using a Rasch measurement framework

Language Testing in Asia

Table 2 Overall person- and item-fit statistics and reliability analysis of the Qiyas for L1 Arabic language test

Total score	Count	Measure	Model error	Mosq	Infit ZSTD	Mnsq	Outfit ZSTD
Summary of 271 measured persons
Mean	28.3	50.0	0.33	0.33	1.00	0.0	1.02	0.1
SD	8.7	0.0	0.92	0.03	0.13	0.9	0.26	1.1
Max.	46.0	50.0	2.78	0.54	1.43	2.5	2.29	3.4
Min.	9.0	50.0	− 1.75	0.31	0.69	− 2.7	0.51	− 2.6
Real RMSE = 0.34		True SD = 0.85		Separation = 2.48		Person reliability = 0.86
Model RMSE = 0.33		True SD = 85		Separation = 2.55		Person reliability = 0.87
SE of person mean = 0.06
Summary of 50 measured items
Mean	153.1	271.0	0.00	0.14	0.99	0.0	1.02	0.1
SD	46.5	0.0	0.90	0.01	0.10	1.8	0.20	2.0
Max.	229.0	271.0	2.17	0.18	1.26	4.4	1.62	4.5
Min.	46.0	271.0	− 1.63	0.13	0.78	− 4.1	0.72	− 3.7
Real RMSE = 0.15		True SD = 0.89		Separation = 6.10		Item reliability = 0.97
Model RMSE = 0.14		True SD = 0.89		Separation = 6.22		Item reliability = 0.97
SE of item mean = 0.13