Development and validation of a multidimensional, culturally and socially inclusive Child Resilience Questionnaire (parent/caregiver report) to measure factors that support resilience: a community-based participatory research and psychometric testing study in Australia

Abstract

Objective: Develop a comprehensive socially inclusive measure to assess child resilience factors. Design A socioecological model of resilience, communitybased participatory research methods and two rounds of psychometric testing created the Child Resilience Questionnaire (parent/caregiver report, child report, school report). The parent/caregiver report (CRQ-P/C) is the focus of this paper. Setting Australia. Participants Culturally and socially diverse parents/ caregivers of children aged 5–12 years completed the CRQ-P/C in the pilot (n=489) and validation study (n=1114). Recruitment via a large tertiary hospital’s outpatient clinics, Aboriginal and refugee background communities (Aboriginal and bicultural researchers networks) and nested follow-up of mothers in a pregnancy cohort and a cohort of Aboriginal families. Analysis Exploratory and confirmatory factor analyses conducted to assess the structure and construct validity of CRQ-P/C subscales. Cronbach’s alpha used to assess internal consistency of subscales. Criterion validity assessed with the Strengths and Difficulties Questionnaire (SDQ) parent report. Results: Conceptually developed CRQ comprised 169 items in 19 subscales across five socioecological domains (self, family, friends, school and community). Two rounds of psychometric revision and community consultations created a CRQ-P/C with 43 items in 11 scales: self (positive self, positive future, managing emotions), family (connectedness, guidance, basic needs), school (teacher support, engagement, friends) and culture (connectedness, language). Excellent scale reliability (α=0.7–0.9), except basic needs scale (α=0.61) (where a highly endorsed item was retained for conceptual integrity). Criterion validity was supported: scales had low to moderate negative correlations with SDQ total difficulty score (Rs= -0.2/– 0.5. p<0.001); children with emotion/behavioural difficulties had lower CRQ-P/C scores (β=−14.5, 95%CI −17.5 to −11.6, adjusted for gender). Conclusion: The CRQ-P/C is a new multidomain measure of factors supporting resilience in children. It has good psychometric properties and will have broad applications in clinical, educational and research settings. The tool also adds to the few culturally competent measures relevant to Aboriginal and refugee background communities.