Alanna Donnelly1, Amanda Fitzgerald1, Mark Shevlin2, Barbara Dooley1. 1. a School of Psychology, Newman Building , University College Dublin , Belfield , Ireland , and. 2. b School of Psychology , Ulster University , Londonderry , Northern Ireland.
Abstract
Background: The psychometric properties of the Revised Child Anxiety and Depression Scale (RCADS) have been established cross-culturally, yet psychometric evidence is lacking for an English-speaking European population. Aim: This research sought to further cross-validate the measure in a non-clinical Irish adolescent sample, and to test for gender and age-based differential item functioning in depression and anxiety. Method: Participants were Irish second-level school students (N = 345; 164 male; 12-18 years, M =14.97, SD = 1.44). Confirmatory factor analysis for categorical data (confirmatory item factor analysis) and multiple-indicator multiple-cause (MIMIC) modelling to identify items displaying possible metric invariance were conducted. Results: A six-factor model fit the data well in both gender samples and both school cycles, as a proxy for age samples. Gender-based metric invariance for 5 of 47 items and age-based metric invariance for three items were identified. However, the magnitudes were small. Internal consistency and validity were also established. Conclusions: While, a number of items demonstrated minor metric invariance, there was no evidence that they influenced overall scores meaningfully. The RCADS can reasonably be used without adjustment in male and female, younger and older, adolescent samples. Findings have implications for the use of the RCADS in an English-speaking European population.
Background: The psychometric properties of the Revised ChildAnxiety and Depression Scale (RCADS) have been established cross-culturally, yet psychometric evidence is lacking for an English-speaking European population. Aim: This research sought to further cross-validate the measure in a non-clinical Irish adolescent sample, and to test for gender and age-based differential item functioning in depression and anxiety. Method: Participants were Irish second-level school students (N = 345; 164 male; 12-18 years, M =14.97, SD = 1.44). Confirmatory factor analysis for categorical data (confirmatory item factor analysis) and multiple-indicator multiple-cause (MIMIC) modelling to identify items displaying possible metric invariance were conducted. Results: A six-factor model fit the data well in both gender samples and both school cycles, as a proxy for age samples. Gender-based metric invariance for 5 of 47 items and age-based metric invariance for three items were identified. However, the magnitudes were small. Internal consistency and validity were also established. Conclusions: While, a number of items demonstrated minor metric invariance, there was no evidence that they influenced overall scores meaningfully. The RCADS can reasonably be used without adjustment in male and female, younger and older, adolescent samples. Findings have implications for the use of the RCADS in an English-speaking European population.