A regression equation relating verbal and full-scale IQs was derived from scores of patients. Scores were highly correlated, and verbal IQ significantly exceeded full-scale IQ. Regression estimates closely estimated mean full-scale IQ, suggesting that the verbal scale serves effectively as a WAIS abbreviation. (Author)

Eight kinds of estimation activities are described and discussed. Nine activities which range from kindergarten to twelfth-grade level are suggested. (MK)

Chemistry pretest scores, high school chemistry grades and, to a greater extent, math SAT scores were useful predictors of college general chemistry grades. Regression analysis of all these predictors combined was used to construct an expectancy table which is being used to identify and advise underprepared students. (BB)

An expectancy model for predicting academic achievement was found to be a potentially useful tool for understanding factors that influence effort and performance. (MM)

The Vane Kindergarten Test (VKT) is judged to have limited usefulness in early detection of learning handicaps for two reasons: (a) Its reliability is too low to allow discrimination between individuals, and (b) The ability of the VKT to predict problem behaviors is quite limited. (Author)

This study investigates the accuracy with which occupational choices for 225 male theological school students can be predicted from scores on the Inventory of Religious Activities and Interests by two experienced counselors and by means of two statistical procedures--discriminant function and profile shape. (Author)

Examined the effect of demographic characteristics, academic endowments, course preparation, and course performance variables on the accuracy of pretest expectations when asking students to predict their performance on a regularly scheduled macroeconomics midterm examination. Finds overconfidence and misjudgments about the scope of the midterm…

The dispositionist bias manifests itself when behavior is overattributed to dispositions, and when contextual factors are underused when predicting behavior. Psychological processes underlying the former bias have been most thoroughly examined. Three studies support the hypothesis that trait implications of past behavior function as heuristics…

The test-retest reliability and predictive validity of developmental quotients (DQs) derived from the Cattell Infant Intelligence Scale was assessed with 83 children with mental retardation, age 30 months or more. Scores were impressively stable on retest. DQs derived from the Bayley Scales of Infant Development were superior to Cattell DQs in…

Examines practices that encourage false negatives and false positives on licensing and certification tests designed to protect the public. Recommends increasing the amount of testing or raising the required passing score for repeat test takers. (FMW)

Suggests that Bretz (in an earlier article) underestimated the usefulness of the college grade point average as a predictor of job success. Reviews recent findings concerning several predictors of job performance related to education. (Author/CH)

Using data published annually by the American Hospital Association, the turnover rates of chief executive officers at 800 teaching and 4,700 non-teaching hospitals were compared. Teaching hospital CEOs experienced less turnover than did nonteaching hospital CEOs. The higher-turnover hospitals typically under-perform relative to other hospitals.…

Using empirical research methods to segment alumni markets, profile donors, prioritize prospects, and project realistic budgets becomes more important as fundraising goals expand. Projecting alumni growth and its impact on program and budget planning, profiling donors and nondonors, and ranking prospects are discussed. (MLW)

Among the five rounds of labor force projections conducted between 1970 and 1980, those estimates produced in 1978 yielded results closest to actual 1985 values. (Author)

Adaptive forms of weighted structural regression are developed and discussed. Bootstrapping studies indicate that the new methods have potential to recover known population regression weights and predict criterion score values routinely better than do ordinary least squares methods. The new methods are scale free and simple to compute. (SLD)