This final report indicates the general areas of investigation in a program of research in the prediction of personal adjustment, conducted over a seventeen-year period. The crucial problems in the prediction of personal adjustment are fundamentally concerned with the data matrix which reflects the experimental designs and displays the essential…

The major purpose of this study was to determine whether data from the 1970 Census could be used to account for at least 70 percent of the variance in grade-level achievement scores in the elementary schools of a big city, without including variables directly denoting race or ethnicity. This goal was attained when it was found that a regression…

The purpose of this study was to generate descriptive statistical estimates regarding the expected proportion of occurrence of each of the PROANA 5 (Process Analysis) variables (line usage, clique group, detrimental clique group, leadership, and dominance) in small group communication. A second purpose was to determine the expected pattern of…

The optimum weighting of variables to predict a dependent-criterion variable is an important problem in nearly all of the social and natural sciences. Although the predominant method, multiple regression analysis (MR), yields optimum weights for the sample at hand, these weights are not generally optimum in the population from which the sample was…

The contributions of several core variables to the white suicide rates in 24 Florida counties were examined. A prediction equation including age, crime rate, insanity, and alcoholism was found that accounted for 79 percent of the variation in the suicide rates across counties. (Author)

A study examined whether information about student approaches to making their course choices could have been used to identify those who subsequently changed or withdrew from their courses. Multivariate analysis indicated that combinations of the information could discriminate between students who persisted and those who withdrew. (Author/MLW)

Analyzes the impact of expenditures per year of schooling on earnings, based on data from a longitudinal study of 1,633 male American World War II veterans. (Available from North-Holland Publishing Company, P.O. Box 211, Amsterdam, the Netherlands; $13.50 annually, plus $4.00 postage and handling) (JG)

(See also TM 501 087, TM 501 089 and TM 501 090.)

Realizing the need to select success crtieria for use in a longitudinal vocational development study, the authors developed a model which depicted the relationships between several available success criteria. Hopefully, these hypothetical relationships would act as a framework for further research. (DS)

Johnson-Neyman, Potthoff, and Erlander and Gustavsson techniques for obtaining confidence regions for the difference between regression surfaces are compared. (DG)

Explores problems involved in the quantitative verification of independent variables in investigations of the effects on student outcomes of planned variations in instructional behavior. Specifically addresses (1) gathering of data through direct systematic observation, and (2) analysis of those data through the use of inferential statistics. (GC)

Describes a study conducted to determine the accuracy of the Cohort-Survival Method of Enrollment Forecasting, based on enrollment data collected over a ten-year period from a random sample of 42 Missouri school districts. Compares forecast accuracy with the enrollment stability and location of each of the districts. (JP)

Presents a method to improve client classification. Reviews clinical prediction models and statistical prediction issues. In classifying clients for remedial services, utility of the prediction scheme depends on gains accrued by its use over program base rates and relative cost. This assumes that differential treatments parallel diagnostic…

Multiple regression is used to discover interconnections between IQ and vocabulary test scores as one variable, and socioeconomic factors as the other. Results show total variance as explained by predictors is never more than eight per cent, indicating differences in IQ scores are not attributable to environmental factors. (RW)