Exploring psychometric properties of the Strengths and Difficulties Questionnaire (SDQ) among adolescent Palestinian refugee girls in Palestine and Jordan

Background The Strengths and Difficulties Questionnaire (SDQ) is a commonly used screening instrument for assessing emotional and behavioral strengths and difficulties among children and adolescents. However, although the SDQ’s psychometric properties have been tested extensively in the literature, its current form is under debate as to whether it is suitable as a tool to assess adolescent mental health, especially among refugees. Therefore, this study aims to validate the SDQ scale among Adolescent Palestinian refugees in the West Bank, Palestine, and Jordan. Methods The psychometric properties of the original SDQ and other suggested alternatives were tested. First, construct validity for the original SDQ was tested through confirmatory factor analysis (CFA). Then, exploratory factor analysis (EFA) was used to evaluate the latent factorial structure for suggested alternatives, followed by CFA to provide additional evaluation of each alternative. Then, convergent, discriminant, and criterion validity were assessed. Finally, the internal consistency and reliability were assessed using multiple tests. Results This study included 2,828 adolescent girls living in Palestine refugee camps in the West Bank and Jordan. Of these, 2,725 participants were included in the final analysis based on completion of the SDQ. CFA indicated that the original five-factor SDQ demonstrated poor fit in this sample (CFI/TLI < 0.40; RMSEA = 0.081; SRMR = 0.110), with several items showing low standardized loadings. EFA and CFA analyses supported alternative factorial structures. A four-factor model showed acceptable fit (CFI = 0.896; RMSEA = 0.040; SRMR = 0.039) after removing poorly performing items, but a three-factor solution provided the best overall fit (CFI = 0.942; TLI = 0.930; RMSEA = 0.034; SRMR = 0.033). The final three-factor model retained 15 items representing Emotional Symptoms, Prosocial behavior, and Conduct Problems. The weak correlation between the difficulties did not support convergent validity score and WHO-5 well-being (r = −0.245). Criterion validity analyses showed acceptable discrimination for overall difficulties (AUC = 0.776). Reliability analyses indicated adequate internal consistency for the composite difficulties scale (α = 0.71). Conclusion The original SDQ structure validity and reliability were not supported for use among Palestinian refugees in the West Bank and Jordan. Instead, the 3-factor and 15-item alternative might be a better alternative, given its superior model fit indices and construct validity. However, it still had limitations in reliability and convergent validity, making it far from being an adequate substitute for the original SDQ. Contexts, culture, and the surrounding environment might influence how adolescents in refugee settings understand, interpret, and respond to SDQ items.