Dimensionality and Construct Validity of the Romanian Self-Report Strengths and Difficulties Questionnaire (SDQ)

The Strengths and Difficulties Questionnaire (SDQ) is one of the most widely utilised measures of behavioural and emotional difficulties among children and young people. Previous research has raised concerns about the psychometric properties of the measure, particularly the internal consistency of the CP and PP subscales. Confirmatory factor analysis (CFA) has generally supported a five-factor solution that is consistent with Goodman’s (1997) original conceptualisation of the SDQ, but alternative factor structures have been validated including models with internalising and externalising factors, and a total difficulties factor. This was the first study to examine the dimensionality, construct validity and internal consistency of the Romanian self-report version of the SDQ. Based on data collected from 1,086 school children aged 9-17 years old, six alternative factor models were specified and tested using conventional CFA techniques and a confirmatory bifactor modelling approach. The five-factor model provided a better fit for the data than alternative factor structures, but was still unacceptable according to a range of overall model fit indices and individual item loadings. Model fit statistics for the five-factor solution were also notably poorer among boys than girls. Internal consistency was low for the CP, H and PP subscales among the total sample and girls only; and for the EP, CP, H and PP subscales among boys only. Results are discussed in terms of the appropriate interpretation of the Romanian SDQ.