Showing papers by "Donald B. Rubin published in 1980"

Bias Reduction Using Mahalanobis-Metric Matching

Abstract: SUMMARY Monte Carlo methods are used to study the ability of nearest available Mahalanobis metric matching to make the means of matching variables more similar in matched samples than in random samples.

Using Empirical Bayes Techniques in the Law School Validity Studies

Abstract: The law school validity studies are primarily concerned with the prediction of first-year average in law school from Law School Aptitude Test score and undergraduate grade point average. Traditionally, a separate admitting equation is estimated in each law school by the method of least squares based on data from students who attended the law school in recent years. These least squares equations can fluctuate rather wildly from year to year. This study employs empirical Bayes techniques to obtain admitting equations that are better than the least squares admitting equations in two ways: for each law school, the empirical Bayes admitting equations are more stable in time than the least squares admitting equations; and the empirical Bayes admitting equations predict student performance more accurately than the least squares admitting equations.

Composite Points in Weighted Least Squares Regressions

Abstract: In a p-variate weighted least squares multiple regression based on n points, each of the p regression coefficients can be calculated from a univariate weighted regression through the origin using N = ( n p ) composite points. This result can be applied to (1) display the p-dimensional multiple regression as p univariate regressions each based on N composite points, (2) isolate collections of points with large leverage, and (3) characterize the region of regression coefficients that can be obtained from fixed data with varying weights. Consequently, composite points are diagnostically useful in the context of robust regression.

Comparing Within- and Between-Subjects Studies

Abstract: Studies employing within-subjects designs may be compared with those employing between-subjects designs in a variety of ways. We discuss and illustrate the comparisons of variabilities, including within-condition variances and precisions as well as the comparisons of means and of mean differences. Our discussion emphasizes the importance of trying to understand the sources of differences.