The Parental Stress Scale revisited: Rasch-based construct validity for Danish parents of children 2–18 years old with and without behavioral problems

Table 3 Global Tests-of-fit and differential item function for the Parental Stress and the Lack of Parental Satisfaction subscales for the total sample

Tests	PS (RM)			PS (GLLRM)^a			LPS (RM)			LPS (GLLRM)^b
Tests	CLR	Df	p	CLR	df	p	CLR	Df	p	CLR	df	p
Global homogeneity	13.1	8	.11	15.4	15	.43	14.6	6	.02⁺⁺⁺	10.3	9	.32
DIF relative to:
Child sample	106.0	8	<.0005⁺	16.4	11	.16	44.2	6	<.0005⁺⁺⁺	22.7	7	<.01⁺⁺⁺⁺
Children age group	45.5	16	<.0005⁺	47.8	30	.02⁺⁺	42.1	12	<.0005⁺⁺⁺	33.7	18	.01⁺⁺⁺⁺
Parent	10.1	8	.26	18.8	15	.22	8.4	6	.21	11.0	9	.27
Parent education	73.6	8	<.0005⁺	17.5	9	.04⁺⁺	17.9	6	<.01⁺⁺⁺	18.7	9	.03⁺⁺⁺⁺
Parent age group	22.1	16	.14	51.9	30	<.01⁺⁺	24.5	12	.02⁺⁺⁺	28.5	18	.05

Notes. PS Parental stress, LPS Lack of parental satisfaction, RM Rasch model, GLLRM Graphical loglinear Rasch model, CLR Conditional likelihood ratio, df degrees of freedom, p p-value, DIF differential item function
Global homogeneity test compares items parameters in approximately equal-sized groups of high and low scoring parents. The critical limits for the p-values after adjusting for false discovery rate were: ⁺ 5% limit p = .0083, 1% limit p = .0017. ⁺⁺ 5% limit p = .008. ⁺⁺⁺ 5% limit unaltered, 1% limit p = .0017. ⁺⁺⁺⁺ 5% limit p = .0083, 1% limit p = .0017
^aThe GLLRM for the Parental Stress subscale assumed that some items pairs are locally dependent (items 3 and 4, and items 9 and 12), that item 9 and 13 functions differentially relative to child sample, and that item 4, 15 and 16 functions differentially relative to parent educational level
^bThe GLLRM for the Lack of Parental Stress subscale assumed that items 1 and 6, and items 1 and 17 are locally dependent and that item 1 functions differentially relative to child sample

ISSN: 1477-7525