The affect and arousal scales: psychometric properties of the Dutch version and multigroup confirmatory factor...

Psychometric properties of the Dutch version of the Affect and Arousal Scales (AFARS) were inspected in a combined clinical and population sample (N = 1,215). The validity of the tripartite structure and the relations between Negative Affect, Positive Affect, and Physiological Hyperarousal (PH) were investigated for boys and girls, younger (8-11 years) and older (12-14 years) children, and for children with high versus low levels of emotional/behavioral problems separately. Results demonstrated an adequate fit of the tripartite structure in each of the subgroups. Multigroup confirmatory factor analysis demonstrated parameter equivalence across sex, age, and psychopathology status. Furthermore, the assumption of invariant error variances of the indicator variables was tenable across sex, age, and psychopathology status. Invariant variances and covariances of the latent constructs (i.e., the tripartite factors) were supported across sex and psychopathology status but not across age. Finally, adequate item and scale properties and good convergent and moderate divergent validity of the Dutch AFARS subscales were demonstrated, except for the divergent validity of the PH subscale.