The psychometric properties of the strengths and difficulties questionnaire for children from refugee backgrounds in Australia

Abstract

Background: The psychometric properties of the Strengths and Difficulties Questionnaire (SDQ) were explored amongst newly arrived children and adolescents from refugee backgrounds in Australia. Methods: The third wave Beginning New Life in Australia (BNLA) data was utilised. This was an Australia-wide cohort study examining humanitarian migrants settling in Australia. The SDQ was administered as part of this study. Amongst the sample of 679 children and adolescents, primary caregivers were invited to answer questions (including the SDQ) for up to two of their children (aged 5 to 17 years). Of these participants, 401, 11-17 year olds also completed the self-report version of the SDQ. Results: Mean scores for this sample were comparable or slightly lower to other Australian samples. However the prosocial behaviour scale was significantly higher than all others, with all scores in the 80th and 90th percentile suggesting a ceiling effect. The Cronbach alpha coefficient for each of the five subscales and the total difficulties scale varied markedly from 0.16 (peer problems) to 0.67 (total difficulties score). Principal components analysis revealed no meaningful factor structure amongst primary caregiver and self-report versions of the SDQ. Cross-scale correlation between each version of the SDQ revealed insignificant to medium correlations. Conclusions: This study adds to an expanding literature that suggests caution when using the SDQ with refugee and non-English speaking populations. While some elements of the SDQ may be used to guide assessment, such as the total difficulties score, its psychometric properties suggest that results should be interpreted with caution. K E Y W O R D Sasylum seekers, psychometric properties, refugees, SDQ, strengths and difficulties questionnaire

Mental health among unaccompanied refugee minors after settling in Norway: A matched cross-sectional study

“…Previous research on the SDQ in languages spoken by children and adolescents of refugee backgrounds have also found poor factor structure and internal consistency of the peer problems subscale [35]. Comparable findings were reported by a study on minor refugees in Australia [36], which also noted that the item 'gets better along with adults than peers' was particularly problematic, as in our study. One reason for the poor properties of this subscale could be due to cultural variations in response styles or conceptions of problematic behaviour [35].…”

“…One reason for the poor properties of this subscale could be due to cultural variations in response styles or conceptions of problematic behaviour [35]. For instance, the 'getting along with older people' item could, in some cultures, actually be perceived as something positive [36] and may perhaps also reflect a necessary coping strategy used by the URMs to adapt to their new host country.…”

“…A review found little support for the originally proposed five-factor model of the SDQ when used in languages spoken by children and adolescents of refugee backgrounds, suggesting that, for this group, the SDQ subscales may measure different constructs than among Western young people [35]. A similar conclusion was also reached by a study on minor refugees in Australia, finding poor psychometric properties, particularly for the prosocial (ceiling effects) and peer problems (low internal consistency) subscales [36].…”

Psycho-Emotional and Behavioral Problems in Children With Growth Hormone Deficiency

“…The use of Goodman's SDQ has become widespread internationally (22)(23)(24), especially among chronically ill patients. The clinical significance of local studies given the diversity of national, ethnic, and regional characteristics of the pediatric population has been demonstrated by a number of studies (25)(26)(27). The SDQ seems to be an appropriate instrument to assess psychosocial functioning among children with GHD, which can be used for clinical and research purposes.…”

Strengths and Difficulties Questionnaire: Preliminary Findings about Local Cut-offs, Prevalence, and Gender Differences in Emotional and Behavioral Difficulties among Indian Adolescents

“…Cut-off scores on the prosocial domain were higher but skewed among Indian adolescents with major proportion of the study sample falling in the 'low' and 'very low'range of prosocial abilities. Problematic nature of the prosocial scale has been empirically proven with established ceiling effects and social desirability issues [3,4].…”