Abstract
Introduction: Robust measures of children’s participation and environments are critical to understanding the impact of health and social conditions on children and young people. One widely used measure, the Participation and Environment Measure for Children and Youth (PEM-CY) has undergone preliminary psychometric analysis that appears favourable. This study extends these analyses and examines the construct validity of the PEM-CY through analysis of its convergent and divergent validity.
Methods: PEM-CY scoring was modified to enable comparison with other measures. A priori predictions of correlations between other measures of participation (Child and Adolescent Scale of Participation) and environments (Child and Adolescent Scale of Environment) and parental competence (Parenting Sense of Competence Scale) were examined using Spearman’s rank correlation. Participants were university staff with children aged 5 to 17 years, recruited through an email invitation to participate in an online questionnaire.
Results: Based on responses from (n=67 to 100) parents, construct validity criteria were not met for the majority of PEM-CY summary scores, using either the recommended or modified scoring. Overall, >75% of predicted relationships were observed for some of the Environmental summary scores but for none of the Participation summary scores.
Conclusions: The convergent and divergent construct validity of summated scores from the PEM-CY was not confirmed. Findings reflect the inherent difficulty of measuring multi-dimensional constructs such as participation. Formative models of measurement design, developed using conjoint analysis of discrete choice data are suggested as a way to design more valid participation measures.