This Monte Carlo study was designed to investigate empirically the efficacy of three statistical strategies for estimating the relationships in extreme group designs in terms of estimation of correlation, power and mean square error for the correlation. Alf and Abrahams's "covariance information statistic" proved to be the best strategy.…

This is one of a series of eight short monographs intended to aid practicing educators in planning and conducting accountability programs in schools. This booklet discusses how to develop an experimental design appropriate to the objectives of a particular research or evaluation effort. Short sections focus in turn on the characteristics of a good…

Sampling distributions for ten tests for comparing population variances in a two group design were generated for several combinations of equal and unequal sample sizes, population means, and group variances when distributional forms differed. The ten procedures included: (1) O'Brien's (OB); (2) O'Brien's with adjusted degrees of freedom; (3)…

Six models for evaluating change are examined via a Monte Carlo study. All six models show a lack of power. A modified analysis of variance procedure is suggested as an alternative. (JKS)

This report describes cluster analysis methodology and illustrates its potential merits for educational research by describing a study designed to identify the natural subgroups existing among students beginning a secondary level remedial mathematics course. Cluster analysis forms groups of relatively homogeneous subjects represented in large data…

Beginners in many disciplines learn that correlation never proves causation, but sometimes, even in public health, correlation, mistaken for causation, becomes the basis for policy and great expenditures of public and private money. "True experiments" with random assignment to experimental and control groups hold a special place in the…