Laurence Siegel

The instructional gestalt: A conceptual framework and design for educational research

SummaryThe argument is advanced in this paper that the pattern of educational investigation wherein comparisons are made between so-called “ experimental” and “ conventional” instructional procedures tends to be defective on three counts: the criteria are often inappropriate or contaminated; the assumptions of homogeneity within and independence between “ experimental” and “ control” conditions are met only when these conditions are very grossly conceived; and the results reflect masking and cancellation effects, thereby revealing relatively little about what transpires in an instructional setting.The Instructional Gestalt is proposed as a methodological framework for educational research focusing upon the interactive nature of learner, instructor, and environmental and course variables constituting the instructional setting. Data generated by explorations within this framework can lead to the formulation of hypotheses which, if subsequently supported empirically, can become principles for inclusion in a theory of classroom learning.The nature of investigations proceeding from the Instructional Gestalt framework was clarified by presentation and discussion of an illustrative study. Since this illustration was only a representation in miniature of the larger investigation, the generalizations about teaching-learning suggested by it are of limited scope and subject to change. Nevertheless, these generalizations do indicate the kinds of hypotheses resulting from studies within the Instructional Gestalt framework.

The Individual High School as a Predictor of College Academic Performance

THE usual studies forecasting student performance in college correlate a number of predictors (usually test scores and indices of high school performance) with some criterion (usually first year college grades). Multiple correlation techniques are then used to combine the predictors to obtain optimal prediction of the criterion. Sometimes an attempt is made to develop separate batteries for various divisions within a school; at other times predictions by division are made from a common battery of predictors (Eels, 1961). Most often predictions are made across university divisions-with