Abstract
M.A.
South Africa is perceived as a country with many economic and social problems. These are stressors that put individuals at risk of developing emotional disturbance. Children, as the most vulnerable sector of society, have the least access to resources such as psychological intervention. A measurement instrument that can identify emotional disturbance within this high-risk group; that minimises the effects of formal learning, cultural and language barriers; and utilises the minimum of resources is needed. The Draw A Person: Screening Procedure for Emotional Disturbance (DAP: SPED) (Naglieri, McNeish & Bardos, 1991) is a screening test for emotional disturbance that seems to fulfil these criteria.
The present study aimed to explore how useful the DAP: SPED is to identify children with emotional disturbance in South Africa. This involved comparing the scores of the DAP: SPED with those of an independent measure for emotional disturbance, the Beck Youth Inventory (BYI-II). To further explore the usefulness of the DAP: SPED, internal reliability and inter-rater reliability were examined. The DAP: SPED and the BYI-II were administered to a group of English-literate Grade 4 learners from two mainstream schools.
The emotional status of the children was unknown, and scores from the BYI-II were used to provide this information. As the BYI-II was used as a control measure, the internal reliability coefficients of its subscales were investigated within the context of the present research. Internal reliability coefficients were good (r=.823 to r=.929, p<.01). A third measure, the Qualitative Questionnaire for Observed Behaviour in the Classroom (QQOBC) was completed by the teachers of participating learners. This measure was used to provide depth to the results as well as independently rate the children’s behaviour.
xiv
Results showed that the DAP: SPED achieved low correlations with the BYI-II (rho=.024 to rho=.76, p<.01), thereby supporting the null hypothesis. The DAP: SPED was also unable to distinguish between emotional disturbance Present and Absent groups, created post facto according to scores on the BYI-II. A statistical analysis of the internal reliability showed that the DAP: SPED had low internal reliability (α=.448), and that the omission of the measurement items increased the internal reliability (α=.525). Inter-rater reliability was significantly high (rho=.687 to rho=. 816, p<.01), thereby rejecting the null hypothesis. Despite the good inter-rater reliability, certain scoring problems were encountered that showed that a margin for scoring bias could influence the usefulness of the measure.