Evaluating an assessment battery for a multi-disability commercial college

A research report submitted to the Faculty of Education, in partial fulfilment of the requirements for the degree of Master of Education (Educational Psychology), University of the Witwatersrand, Johannesburg, 1995.The present research study evaluated an unresearched assessment battery used to determine student trainability at a multi-disability commercial college. The issues of validity - specifically criterion-related validity and content-related validity and dynamic assessment were the main focus of the study. Revisions to the assessment battery where necessary were recommended, Fifty-two students from the multi-disability college formed the sample population for establishing the criterion-related validity of the assessment battery. Students assessment results were correlated with the number of credits .hey obtained on course completion using the Pearson Product Moment Correlation Coefficient. This research established that useful inferences about student trainability could be made from the existing assessment battery scores. Establishing the content-related validity of the assessment battery was a two foid process. Firstly, course lecturers completed a self-administered questionnaire to determine what skills/ attributes were necessary for success in the four courses offered at the College. Secondly, a workshop was held with a multi-disciplinary team of professionals to determine what skills/ attributes were being tapped by the various sub-tests of the battery. It was established that twelve of the skills/attributes necessary for course success were not tapped on the existing assessment battery. Measures which tap these skills/attdbutes need to be included in the assessment battery so that its content validity can be established. Given the controversy relating to standardised/static testing it was important to research the relevance and usefulness of the introduction of a dynamic assessment component to the existing battery. Dynamic assessment would allow for learning potential to be tapped. Analyses of variance were conducted to determine the relationships between the variables - race, educational level and type of disability with performance on the assessment battery. The significant relationships between educational level and performance and type of disability and performanc