An alternative approach to assessing science competencies

Abstract

It is widely established that South African learners score poorly on written tests—whether standardized or locally devised tests as part of research and evaluation. Similar results were obtained in the project of which this research is a part. In spite of findings from classroom observations and interviews that teachers were competent, lessons well prepared and learners deeply engaged, learners continued to score poorly on written tests. The research reported in this paper is a response to these results where an alternative strategy to evaluate what learners had learnt during their science lessons was implemented. The SOLO Taxonomy was used as a means of categorising different levels of learners responses in an attempt to find out what learners had in fact learnt. Data were collected from classes in 10 schools. Analyses of the data show that the learners learned much more than the tests indicate, and could talk insightfully about science ideas and relationships between them. They also gave some indications of what learners find ‘important’ in their science learning and how they like to present their learning. The implications for assessment are clear: strategies that assess more than learners' written responses to questions assessing knowledge of science concepts are required if we wish to gain a better understanding of the learning that occurs in science classrooms.