Effects of rating criteria order on the halo effect in L2 writing assessment: a many-facet Rasch measurement analysis

Language Testing in Asia

Table 13 Rater fit statistics of novice raters after anchoring

Rater	Standard				Reverse				Random
Rater	Severity	Infit	Outfit	Variance^a	Severity	Infit	Outfit	Variance^a	Severity	Infit	Outfit	Variance^a
01	0.06	1.01	1.04	0.78	0.44	1.11	1.13	0.89	0.81	0.94	0.94	0.66
02	0.01	1.14	1.17	0.66	0.89	0.84	0.84	0.57	1.86	1.13	1.06	0.54
03	0.47	0.73	0.74	0.52	0.32	0.40	0.40	0.36	0.89	0.69	0.69	0.46
05	0.54	1.33	1.31	0.67	1.27	1.62	1.55	0.54	1.08	1.35	1.35	0.78
06	− 1.57	0.84	0.79	0.44	− 0.66	0.60	0.61	0.31	− 0.62	0.55	0.55	0.27
07	0.64	1.04	1.04	0.57	0.71	0.89	0.89	0.60	1.04	1.30	1.28	0.81
08	0.00	1.45	1.42	0.87	− 0.53	1.26	1.24	0.58	0.19	0.82	0.82	0.55
09	− 0.04	0.61	0.61	0.55	1.16	0.86	0.88	0.50	0.34	0.98	0.98	0.73
10	− 0.04	1.10	1.10	1.05	0.59	1.09	1.10	0.94	1.01	1.16	1.16	0.54
11	0.29	0.86	0.88	0.67	1.22	1.40	1.33	0.43	0.54	1.42	1.44	0.81
12	− 0.36	0.69	0.69	0.43	− 0.08	0.87	0.87	0.48	1.04	0.54	0.54	0.31

Note: Halo-exhibiting raters whose fit statistics is smaller than 0.80 are in bold
^aAverage within-test taker variances