Problems in the evaluation of focussing objectives and their implications for the design of systems models of the curriculum with special reference to comprehensive examinations

The author considers the use of comprehensive examinations as vehicles for multiple strategy in engineering. Evaluations of A-level Engineering Science examinations in England show that subtests of examinations to test specific domains (focussing objectives) do not show consistently high levels of certainty that these domains were tested. Factorial analyses suggest that, unless the instructional strategy is specifically designed to obtain the required objectives, no high expectation can be held that they are obtained in student learning. Therefore a major task for the designer of multiple-strategy assessment procedures is the selection of objectives for assessment, since, by definition, outcomes can only be determined which are well specified. A model of the curriculum, assessments, and teaching process is derived. It implies that an assessment movement which does not take into account the design of instruction will not necessarily bring about an improvement in the quality of education provided. Support for this view is to be found in studies of comprehensive examinations in engineering science, mathematics, and history, undertaken in Britain and Ireland. The findings of these studies are discussed