The use of the Analytic Hierarchy Process (AHP) in assisting researchers to analyze decisions is discussed. The AHP is compared with other decision-analysis techniques, including multiattribute utility measurement, conjoint analysis, and general linear models. Insights that AHP can provide are illustrated with data gathered in an auditing context.…

Using 1985 personnel record data from the Portuguese Ministry of Labor, a recent regression analysis found that the average rates of return for schooling in Portugal are in the 9.4 to 10.4 percent range. Large gender and regional earnings differentials exist. Study implications and limitations are discussed. (21 references) (MLH)

Physical and chemical indicators of water quality monitored by Environmental Canada between 1977 and 1987 in the Niagara River at Niagara-on-the-Lake and in the Saint Lawrence River at Wolfe Island are analyzed for seasonal and annual variations. Results indicate that specific conductivity, sodium, and chloride have decreased significantly over…

A solution of the convection-dispersion equation is used to describe the solute breakthrough curves generated in the demonstrations in the companion paper. Estimation of the best fit model parameters (solute velocity, dispersion, and retardation) is illustrated using the method of moments for an example data set. (Author/MDH)

Some approximate formulas are presented for standard errors of estimated regression coefficients in two-level designs. If the researcher can make a reasonable guess as to parameters occurring in the model, this approximation can be a guide to the choice of sample sizes at either level. (SLD)

Reexamines Ashenfelter and Krueger's estimates of schooling returns (13%), using three additional years of the same twins survey. Finds a 10% return per year of schooling completed. The within-twin regression estimate of schooling's effect in the log wage is smaller than the cross-sectional estimate. (14 references) (MLH)

Data from a large sample of Australian twins indicate that 50 to 65 percent of variance in educational attainments can be attributed to genetic endowments. Only about 25 to 40 percent may be due to environmental factors, depending on adjustments for measurement error and assortative mating. (Contains 51 references.) (MLH)

Simulations were used to compare the moderator variable detection capabilities of moderated multiple regression (MMR) and errors-in-variables regression (EIVR). Findings show that EIVR estimates are superior for large samples, but that MMR is better when reliabilities or sample sizes are low. (SLD)

Compares multiple regression analysis and a simplified scale in predicting competitive employment after the provision of vocational rehabilitation services. The two prediction techniques yielded nearly identical results when applied to an independent, cross-validation sample. Discusses practical applications of the simplified procedure to client…

The quality of approximations to first- and second-order moments based on latent ability estimates is discussed. The ability estimates are based on the Rasch or the two-parameter logistic model, and true score theory is used to account for the fact that the basic quantities are estimates. (SLD)

Studied the spillover between family and work by examining the link between marital status and work performance across marriage, divorce, and remarriage. A polynomial regression model was fit to the data from 78 evaluations of an individual professor, and a cubic curve through the 3 periods was statistically significant. (SLD)

The domain specificity and generality of an important critical thinking skill was examined by administering nine reasoning and decision-making tasks to 125 adults. Performance across tasks displayed considerable domain specificity, but five tasks displayed moderate convergence. Performance on tasks that displayed domain generality was more…

Describes how the State University of New York, using large databases and regression techniques, has developed several coordinated, comprehensive efforts to evaluate and project current and future enrollments. One requires decomposition of the student population into mutually exclusive subgroups. The other, superior, approach uses logistic…

Evaluated systematic empirical evidence of how postsecondary institutions support and promote the use of student assessment information in academic decision making using linear regression techniques. Found that student assessment data has only a marginal influence on academic decision making, with slightly more influence on educationally related…

Uses longitudinal study of differences between boys and girls in levels of mathematics and science achievement across grades 7 through 10 to extend hierarchical modeling to allow for regression among latent variables using a fully Bayesian approach. (Contains 30 references.) (PKP)