The evaluation of qualitative examination questions at matriculation level in physics

摘要

The style and format of Physics examination papers has changed markedly over the past hundred years. Physics was regarded formerly as a sub-division of Mathematics; hence, apart from giving formal statements of scientific laws, and some account of experimental procedures, examination candidates were required to spend most of their time, and to earn most of their marks by solving numerical problems. Most Physics examination papers today retain the emphasis on problem-solving by calculation. It has been recognised, however, that the ability to obtain correct numerical answers by substition in a formula does not necessarily imply understanding of the physical principles which underly the problem. There has been a reaction against the awarding of marks for algebraic or arithmetical manipulation. Bloom and his disciples, by calling on examiners to define precise behavioural objectives, have encouraged the development of qualitative questions. These require no calculation nor numerical answer, but do aim to test the candidate's insight, by requiring him either to explain phenomena in scientific terms, or to predict the outcome of changing conditions. In this investigation samples of both quantitative and qualitative questions were extracted from Physics papers past and present. A closer study was made of qualitative questions used in the Natal Senior Certificate Physics papers (both Higher Grade and Standard Grade) in November 1987. The examiners were asked to state the objective of each question, as well as its categorization in terms of a simplified Bloom taxonomy. Candidates' answers to these questions were extracted from the scripts written at a representative sample of Natal schools. In the case of multiple-choice questions, an item analysis was performed, and discrimination indices were calculated . The responses of individual candidates to the longer questions were collected, classified and discussed. In each case the effectiveness of the question was studied, as to whether or not it was successful in detecting the presence of correct scientific concepts in the candidates' thinking. It was concluded that most of the qualitative questions used by the examiners were effective. Suggestions were made as to how their efficacy could be improved. Much further study and experimentation are needed to develop the effective use of this type of question