Skip to main content

Table 5 Global Tests-of-fit and differential item function for the final models of the Parental Stress and the Lack of Parental Satisfaction subscales to the models for the separate behaviour and ordinary samples depicted in Fig. 2

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

Tests Behaviour sample   Ordinary sample  
PS (GLLRM)a LPS (RM) PS (GLLRM)b LPS (GLLRM)c
CLR df p CLR df P CLR Df p CLR df p
Global homogeneity 5.0 10 .89 8.1 6 .23 17.2 11 .10 1.8 8 .99
DIF relative to:
 Children age group 27.2 20 .13 8.5 12 .74 25.2 16 .07 26.2 16 .05
 Parent 16.8 10 .08 3.3 6 .77 7.1 11 .80 4.0 8 .85
 Parent education 6.6 6 .36 2.6 6 .85 23.3 11 .02+ 11.3 8 .18
 Parent age groups 31.1 20 .05 14.3 12 .29 24.5 22 .32 19.1 16 .26
  1. 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
  2. Global homogeneity test compares items parameters in approximately equal-sized groups of high and low scoring parents
  3. + The critical 5% limit for the p-values after adjusting for FDR was p = .0100
  4. aThe GLLRM for the PS scale assumed that item 15 and 16 functions differentially relative to parent educational level
  5. bThe GLLRM for the PS scale assumed that items 3 and 4 are locally dependent, and that item 4 functions differentially relative to child age
  6. cThe GLLRM for the LPS subscale assumed that items 1 and 6, and item 1 and 17 are locally dependent