A Limitation of the Applicability of Interval Shift Analysis to Program Evaluation

Abstract

There can be almost unanimous agreement among those of us who are involved in the evaluation of educational programs in the public schools that there is a great need for the development of innovative methods of analyzing student achievement data for the purpose of program evaluation.Most persons charged with the responsibility of project or program evaluation come equipped with the tools of behavioral research. We have attempted to apply experimental and quasi experimental designs, hypothesis testing, and inferential statistics to developmental or on-going programs and have found these methods inadequate.In his 1970 review of the literature on measurement and analysis techniques in evaluation, Sjorgren (1970) concluded that exploration with the methodologies of other fields should have a high priority:... much work is needed in adapting the methods of the social and behavioral sciences to educational evaluation. The models of anthropology, economics, political sciences, sociology, law, and journalism need to be studied for their applicability to evaluation. Some of this analysis is occurring, e.g., cost-benefit analysis. It is clear, however, that the models are not immediately applicable to educational evaluation because many variables that are used in the basic discipline are not present or must be measured in another way in education. The problem of adapting such models to educational situations creates measurement problems of its own (p. 314).