This paper discusses the principles underlying discriminant analysis and constructs a simulated data set to illustrate its methods. Discriminant analysis is a multivariate technique for identifying the best combination of variables to maximally discriminate between groups. Discriminant functions are established on existing groups and used to…

A dichotomous dependent variable is used to determine a combination of variables that will predict group membership. Dichotomous variables are frequently encountered in multiple regression analysis. However, several textbooks question the appropriateness of using multiple regression analysis when analyzing dichotomous dependent variables. The…

The recent development of strong statistical techniques has made accurate predictions of recidivism possible. To investigate the utility of discriminant analysis methodology in making predictions of recidivism in juvenile offenders, the court records of 271 male and female juvenile offenders, aged 12-16, were reviewed. A cross validation group…

Using discriminant analysis, factors influencing teacher attrition were studied using three groups of University of Michigan graduates with teaching certificates: (1) those who taught continuously; (2) those who left teaching within five years; and (3) those who never taught. Implications for school administration, teacher training, and further…

A multivariate approach reexamined Lewis Terman's longitudinal study data comparing the 100 most successful and 100 least successful men identified in the 1920s as having very high intelligence. Results reaffirmed the importance of educational attainment in vocational achievement, though intelligence and amount of early acceleration also predicted…

An informal study was conducted of the courses selected by ninth grade students who later dropped out of high school. Longitudinal data were available for high school students in the Austin (Texas) Independent School District, from 1978-79 to 1982-83. The courses selected by high-risk students in ninth grade, including extracurricular activities…

A 15-item questionnaire was administered on the online catalog terminals at the University of New England (Australia) to assess student attitudes toward computer technology and the new online catalog, and identify predictors of future online catalog use. Data analyses defined two groups of respondents: "reluctant users" and "naive enthusiasts."…

Describes a survey of 356 adult students enrolled in university-level courses at the Open Learning Institute which was conducted to assess the predictive validity of Tinto's theoretical model of student dropout. It was found that, overall, the Tinto model appears to be a useful framework for such investigations. (Author/LRW)

Discussion of teamwork and group decision making focuses on study of undergraduates conducted to identify the variables that affect team cohesion in a business game. A literature review on small groups and business games is presented, hypotheses tested are described, discriminant analyses are explained, and suggestions for further research are…

This study found that a linear discriminant function was able, with 73 percent accuracy, to distinguish between learning-disabled dropouts (n=213) and learning-disabled graduates (n=92). The discriminant function was composed of six variables--student ethnicity, reading ability, family intactness, family socioeconomic status, school transfers, and…

Reviews a study of 117 students at the University of Saskatchewan which examined the relationship between high school computer experience and university achievement in introductory computer science. The pretests used are described, findings are analyzed, and student withdrawal patterns are discussed. (Author/LRW)

Random samples of 1979 and 1980 entering freshmen were given the Non-Cognitive Questionnaire (NCQ), which was designed to assess noncognitive dimensions predicting minority student academic success. The validity of the NCQ in predicting graduation after five and six years was determined. The graduation rate was found to differ significantly for…

Three possible explanations for superintendent succession focus on poor administrative performance, district response strategies, and the politics of the chief executive's relationship with the school board. To analyze succession in the context of declining enrollment, a case study survey was conducted of 56 school districts whose peak enrollment…