The Effectiveness of Alternative Instructional Media: A Survey

This survey provides an overview of research on the effectiveness of alternative instructional media. The media discussed are traditional classroom instruction (TI), instructional radio (IR), instructional television (ITV), programmed instruction (PI), and computer-assisted instruction (CAI). The effectiveness of these media is examined from a reasonably macroscopic point of view; the psychology of pupil-teacher interaction or the 'content variables' of ITV, to take two examples, are at a micro-level not considered. Achievement test scores constitute the measure of effectiveness most frequently used in this survey though, where available, results concerning the affective impact of the various media of instruction are included. Achievement test data, in most cases, were collected only on an annual basis, so they reveal no fine-grained detail about the learning process. Since this survey is relatively brief and its scope broad, a few caveats are in order. First, where literature surveys are available, their results have been cited to the extent possible and, frequently, original sources remain unchecked. Second, available knowledge of the effectiveness of the various media varies considerably; much more is known about TI and ITV than about the others. For this reason a survey such as this is inherently spotty in its conclusions. The third warning, related to the second, is that many of the evaluations fall short of (or lack entirely) scientific standards of analysis and reporting. For this reason, it was sometimes necessary to attempt to cull conclusions from essentially journalistic accounts of projects. Fourth, it should be noted that this survey is limited to instruction within a school setting. Finally, the survey excludes information on costs. Before beginning the literature survey, we present a paradigm for measurement of effectiveness and then discuss several less desirable alternatives that have actually been employed. In the surveys of the individual methods where adequate prior surveys are unavailable, results from a representative sample of individual evaluations are discussed.