Item analysis using Rasch models confirms that the Danish versions of the DISABKIDS® chronic-generic and diabetes-specific modules are valid and reliable

Table 3 Item fit statistics comparing the observed and expected item-rest-score correlations under the two models

Item	A: Rasch model–six items			B: Graphical log linear Rasch model^a–five items
Item	Observed γ	Expected γ	p	Observed γ	Expected γ	p
CG1	.54	.58	.48	.56	.63	.24
CG2	.62	.57	.39	.65	.69	.41
CG3	.73	.58	.017	.74	.62	.051
CG4	.68	.59	.12	.70	.70	.99
CG5	.71	.58	.052	.74	.67	.17
CG6	.27	.56	< .001

The correlations are measured using Goodman and Kruskal's gamma. Goodman and Kruskal's gamma measures rank correlation for ordinal categorical data [37]. When item 6 is excluded, and local dependency and DIF are allowed the observed correlations are the same as expected
^aThe model assumes that items 1 and 3 are locally dependent and that item five is affected by DIF depending on age

ISSN: 1477-7525