The current state of usage of regression models in analysis of variance (ANOVA) designs is empirically examined, and examples of several statistical errors made in usage are presented. The assumptions of the general linear model are that all predictors are known without error of measurement and are fixed with no replication or sample variation; in…

This investigation examines the impact the l98l Graduate Record Examination (GRE) General Test Format Revision had on the stability over time of the verbal, quantitative, and analytical scores. Scores were used from the self-selected group of repeaters who took the GRE General Test twice between October 1980 and June 1982. Examinees were divided…

Factor analysis and multiple regression are two major approaches used to look at functional age, which takes account of the extensive variation in the rate of physiological and psychological maturation throughout life. To examine the role of racial or cultural influences on the measurement of functional age, a battery of 12 tests concentrating on…

The success with which freshmen cumulative point averages may be predicted from a combination of a student's high school rank, Scholastic Aptitude Test (SAT) score (Verbal) and SAT score (Mathematics) was investigated. A multiple regression formula developed by the College Entrance Examination Board was used to combine the three predictors into a…

In 1976, the Association of American Medical Colleges developed a map-like model to describe the global picture of the U.S. medical school similarities with respect to two loosely defined concepts: an institutional emphasis on research and an emphasis on clinical and graduate medical training. This study is an attempt to replicate the results of…

Using construed data, it is demonstrated that the commonly-used method of showing mean salary by sex or mean salary by rank by sex can be misleading, and that a more realistic picture can be obtained using multiple regression analysis. Faculty staffing patterns were designed for two simulated institutions and is presented showing means and…

A stepwise regression technique was used to analyze assessment data while taking differences in nonschool variables across districts into account. The primary purpose of this investigation was to determine whether the inclusion of quadratic and/or interaction terms in a regression model would improve the prediction of school district average…

Multiple sources of complications in the analysis of multilevel educational data are described with emphasis on the determination of a method of analysis that will provide accurate estimates of teacher/class effects even when there are systematic differences in within-class regressions of outcome on input associated with teacher/class…

This study involved the investigation of the differential validity hypothesis as it pertained to the disadvantaged college student. It evaluated, by means of a multiple regression analysis, the use of SAT scores (Math and Verbal) and the Nelson-Denny Reading Test as predictors of college success, as measured by semester grade point average, for…

This report explains why the Management Implications of Team Teaching (MITT) project chose multiple linear regression and path analysis to analyze through-time relationships among variables, and why it rejected repeated-measures analysis of variance (ANOVA) and difference scores over time. Project MITT examined governance and work structures for…

This is the second volume of a two volume study. The first volume examined the literature to identify authoring aids for developing instructional materials, and to identify information clearing houses for existing materials. The purpose of this volume was to develop a means for assessing the cost versus expected benefits of innovations in…

A search for better predictors of incomes of high school and college graduates is described. The accuracy of the prediction, implicit in the work of John R. Walsh of Harvard University, that the income differences in a given year are good indicators of income differences in future years, was tested by applying standard statistical procedures to…

The development of a predictive equation capable of explaining a significant percentage of enrollment variability at Florida International University is described. A model utilizing trend analysis and a multiple regression approach to enrollment forecasting was adapted to investigate enrollment dynamics at the university. Four independent…

This paper reviews issues pertaining to the relationship between child mortality and fertility and examines the fertility-mortality relationship of women who emigrated to Israel from various countries in Asia, Africa, and Europe and continued child bearing in Israel. Data from the 1961 Israel census of population is used. Among issues addressed in…

A strategy is proposed for dealing with educational research problems, using as an example a study based on data that had already been collected and using that data to discover how differences in school characteristics may result in differences in their students. The research method is based on concepts of analysis of covariance and is concerned…