Focus is on two areas: (1) the factors available to the selection committee when students apply for admission into graduate programs, and (2) those factors emerging after admission, resulting from students' meeting program-imposed requirements for graduation. Program variables are shown to be more important than admission variables. (Author/LBH)

A nationwide survey of all users (400) of CAMPUS, PLANTRAN, RRPM, and SEARCH conducted in 1975 by the University of Massachusetts at Amherst examined the extent of model use and their effectiveness. This paper reports a follow-up study on user support for models and opinions of preconditions for future use. (LBH)

Based on a microeconometric examination of Philadelphia school district data, analyzes the impact on student achievement of pupil-specific variation in students' genetic endowment and socioeconomic status, teacher quality, nonteacher school quality, and peer group characteristics. Single copies may be purchased from Secretary, C. Elton Hinshaw,…

Data from the National Research Council's Doctorate Records File extract prepared for UCLA indicated that source of support was the most important variable in predicting time to doctorate. Following source of support were postdoctoral plans, number of dependents, sex, and field of study. (Author/MLW)

The results of this issue's series of articles are summarized, and implications for theory, research, and practice are discussed. Initial academic achievement was the strongest predictor of student achievement. The impact of attentiveness, motivation, student background, sex, and time on task were also described. (GDC)

An important part of the analysis of the expected effects of institutional closure is estimation of redistribution of students attending the closed school. How coefficient estimates from models of enrollment demand can be used to predict the alternative attendance choices of students enrolled at the closed school is discussed. (Author/MLW)

In a sample of undergraduates with comparable abilities, preparation levels, and interests in science, women abandoned their plans to major in science at a significantly greater rate than men did. The different factors that predict men's and women's decisions to choose scientific fields are examined. (Author/MLW)

Examined the relationship between instructional workload and budgeting of personnel positions and how this relationship changes as personnel positions become scarce. Longitudinal data suggest a direct relationship between the bureaucratic model and personnel allocations; tendencies to use the bureaucratic model increase as personnel positions…

Among high school students, males' educational and career aspirations were strongly influenced by their academic ability and achievement, while females' aspirations were influenced mainly by parental attitudes and their socioeconomic background. (Author/MJL)

Outlines a conceptual framework, the Reflective Judgement model, which describes predictable differences in the intellectual development of college students. Validation attempts are discussed, and applications leading to sound counseling and educational interventions are suggested. The model is illustrated with an example of a classroom assignment…

A multi-attribute attitude model was used to determine whether a multicriteria scale can be used to predict student preferences for and attendance at universities. Data were gathered from freshmen attending five state universities in Ohio. The results indicate a high level of predictability. (Author/MLW)

Muffo and Coccari (1982) analyzed data pertaining to the causes of variation in the number and amount of external funds for research. By using LISREL, a reanalysis of the study shows that the only significant indicators of funding are past success in securing funds and an emphasis on graduate education. (Author/MLW)

An extension of Tinto's model of the college dropout process was tested with freshmen at an Australian university. The college entrance examination was a relatively valid predictor of whether students will pass, achieve honors, or fail or withdraw. Nonintellective factors were not valid predictors of academic progress. (Author/BW)

Multiple discriminant analysis is an effective research tool to approach the problem of student attrition. Using this method, the small, liberal arts college can more accurately identify students who are likely to persist. Academic achievement and nonacademic variables are coupled to enhance the power of discriminant analysis. (RC)

A model of voluntary behavior change is developed and applied to higher education faculty and the process of instructional innovation. It identifies individual and organizational variables that determine whether a faculty member will change his instructional practices by adopting new practices, and also determines the likelihood that he will…