Three versions of a probabilistic model adapted from the theory of information retrieval--a binary version, a version using the full value of the data, and a version using principal components--were tested and applied to data available from application forms to predict graduate school performance of library school students. (EM)

Two competing definitions of test fairness and their differing implications are explained and illustrated, and data from a number of published studies on test bias are reanalyzed; the focus is on unfair bias that may exist in tests that are approximately equally valid for both majority and minority groups. (Author/JM)

Multinomial logic and multiple discriminant analyses are used to predict probabilities that individuals will attain several occupational categories based on individual characteristics and qualifications. The authors found that their results vary substantially from women's actual occupational distribution. They conclude that a significant portion…

A visual display is introduced that enables one to compare the predictive validities of all possible non-negatively weighted sums of two predictors. This two-predictor validity curve assesses the level of predictability of a criterion from composites of two predictors. Derivations are presented for the general case. For illustrative purposes, the…

Results are described for the first two phases of a study to develop and evaluate models of student progress in a technical training course being offered in a computer managed instructional environment at Lowry Air Force Base, Colorado. Several categories of models were developed in Phase I and defined by model type and by the type of variables…

Employed were a small sample of fifth graders to examine techniques for assessing the validity of psychosocial maturity (PSM) and the scales used to measure it. Three validation techniques were examined: (1) Data obtained from self-report scales, teacher ratings, and peer ratings were used to generate three multitrait-multimethod matrices (total…

This paper offers a preliminary analysis of the effects of a semi-segregated school system on the IQ's of its students. The basic data consist of IQ scores for fourth, sixth, and eighth grades and associated environmental data obtained from their school records. A statistical model is developed to analyze longitudinal data when both process error…

Admissions criteria used for three classes of occupational therapy students were subjected to statistical analysis. Weighted factors included cumulative, science, and behavioral science grade point average, reading and mental ability test scores, and essay and interview scores. Revised factors eliminated the reading test and subject-specific grade…

Evaluates the validity of using circulation data as the index to the total utilization of a library collection and tests the hypothesis that the proposed mixture of Poisson distributions can describe and predict various use distribution. Seven statistical analysis listings, six algorithms, 20 tables, and 43 references are provided. (Author/RBF)

Evaluates the validity of the test used to place students in the comprehension-based courses offered by the Second Language Institute of the University of Ottawa. The test is described in the context of the target student population and the comprehension-based courses. Different factors of evaluating the concurrent and predictive validity were…

Admissions data and first-year grade point average (GPA) data from 11 graduate management schools were analyzed to evaluate the predictive validity of Graduate Management Admission Test[R] (GMAT[R]) scores and the extent to which predictive validity held across sex and race/ethnicity. The results indicated GMAT verbal and quantitative scores had…

A study was conducted at Scottsdale and Paradise Valley community colleges to assess the importance of math for the study of economic principles and the extent to which a student's math background determines performance in micro- and macro-economic courses. Students (N=156) from six economic principles classes were surveyed regarding their high…

TWO TYPES OF COUNSELING PROGRAMS WERE COMPARED BY RANDOM ASSIGNMENT OF 30 COUNSELORS TO TWO PRACTICUM TREATMENTS. FIFTEEN COUNSELORS RECEIVED 60 HOURS OF COUNSELING EXPERIENCE CONCENTRATED INTO THE LAST 3 MONTHS OF A YEAR-LONG TRAINING PROGRAM. ANOTHER 15 RECEIVED 60 HOURS DISTRIBUTED IN THE LAST 6 MONTHS. SUPERVISING STAFF AND METHODS OF…

The suppressor variable, a variable wholly uncorrelated with a criterion, but which nevertheless improves prediction because of its relationship with a predictor, is critically examined. For a suppressor so defined, formal identities are shown with part, partial, and multiple correlational procedures. It is demonstrated that if maximum prediction…

Examines variables used to predict the academic success or failure of college football players. Valid predictors include the following: (1) high school grades; (2) repeating a year in school; (3) feelings towards school; (4) discipline history; (5) mother's education; and (6) high school background. (FMW)