Autism screening in children: using the social communication questionnaire in a Western Cape population

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

Autism spectrum disorder (ASD) has a global prevalence of approximately one percent of all new births. There is a lack of literature on autism in South Africa. South African children are waiting years for diagnoses, despite the fact that early diagnosis and subsequent intervention appear to have a positive effect on the outcomes of the intervention. A screening device to detect ASD could be used to speed up the diagnostic process. This study tested the viability of using the Social Communication Questionnaire (SCQ) in a Western Cape state-funded hospital. This thesis describes Phase 1 of a larger study. The 40 item SCQ was adapted and translated into Afrikaans and isiXhosa. The English, Afrikaans and isiXhosa versions of the SCQ were administered to parents of very young children attending the Red Cross Children?s Hospital?s developmental clinic (N = 228, age range of children = 3.00-5.97 years). Positive results were that no relationship was found between age and SCQ score, or between SCQ language version and SCQ score were found. However there was a relationship between SCQ score and socioeconomic status, indicating a possible bias in the SCQ. Internal reliability of the SCQ versions was analysed and was satisfactory. The factor structure of the English SCQ was examined. Two and four factor solutions were explored, with the two factor solution proving the best fit with good internal reliability. This two factor solution reflected the recent changes to the Diagnostic and Statistical Manual of Mental Disorders, as well as previous findings on the SCQ and the ADI-R, the diagnostic instrument on which the SCQ was based. Preliminary results of Phase 2 of the larger study were analysed. Eighteen children received an Autism Diagnostic Observation Schedule (ADOS) assessment;; the current gold standard for diagnosing ASD. SCQ scores proved to be a good predictor of ASD diagnosis, predicting 17 out of 18 individuals correctly. Further research on isiXhosa and Afrikaans versions of the SCQ as well as the predictive power, sensitivity and specificity and cut-off scores for the SCQ is recommended.