Stigma toward individuals with mental illness: validation of the Arabic version of the Attribution Questionnaire in a university student population

Purpose This study aims to examine the reliability and validity of the Arabic version of the attribution questionnaire (AQ). Design/methodology/approach The AQ is designed to assess attitudes, affects and behavioral intentions related to a hypothetical person diagnosed with schizophrenia. The original English version was translated into Literary Arabic. A total of 310 students registered in different universities, with medical and paramedical establishments excluded completed the Arabic version of AQ. Reliability was tested using Cronbach’s alpha coefficients. Structural equation modeling was used to test hypothesized paths. Correlations among exogenous (e.g. responsibility) and endogenous (e.g. help) variables in the path were examined. Fit indicators were then examined for equations that were identified. Findings The results revealed that the Arabic version of AQ showed acceptable psychometric properties in the assessment of stigma in the Tunisian population. All factors of this Arabic version showed Cronbach’s alpha values equal to or greater than 0.72. Structural equation models for the responsibility and dangerousness models were mostly supported. The Arabic version of AQ is valid and reliable for the assessment of stigma in Tunisian and Arabic-speaking populations. Practical implications The Arabic version of AQ may be used to promote research on stigma toward people with mental illness in larger and more representative Tunisian and Arabic-speaking populations, which will help to further address the complex and multifaceted phenomenon of stigma toward people with mental illness. Originality/value This is the first validated stigma measure in the Tunisian socio-cultural context.