AccessMyLibrary provides FREE access to millions of articles from top publications available through your library.
125 total articles
Set up an RSS feed
Educational and Psychological Measurement back issues
MANOVA SIMULTANEOUS TEST PROCEDURES: THE POWER AND ROBUSTNESS OF RESTRICTED MULTIVARIATE CONTRASTS.
December 1, 1998... Multivariate analyses in educational and psychological research have become much more prevalent since the 1970s (Maxwell, 1992). Emmons, Stallings, and Layne (1990) surveyed 16 years of research and determined that "the multivariate characteristic of the social science research environment...
MEASURING CHANGE CONVENTIONALLY AND ADAPTIVELY.
December 1, 1998... Educators and psychologists have long recognized the problems inherent in the measurement of change, particularly in the use of a difference score. A principal problem of difference scores is their unreliability in certain circumstances. Another problem, the fact that difference scores...
SINGLE-ITEM VERSUS MULTIPLE-ITEM MEASUREMENT SCALES: AN EMPIRICAL COMPARISON.
December 1, 1998... It has become almost a maxim in psychometrics that, holding all else constant, the more items in self-report measures of psychological constructs, the better. For example, Nunnally (1978) noted that "other things being equal, a long test is a good test" (p. 243). This recommendation is...
CA AND SPOD FOR THE ANALYSIS OF TESTS COMPRISED OF BINARY ITEMS.
December 1, 1998... It is well known that the regression of a dichotomous item on a continuous latent variable is necessarily nonlinear (McDonald, 1980; Mislevy, 1986). Hence, the fitting of a linear model to a set of dichotomous items results in misspecification that undermines the making of meaningful claims...
AN ANALYTIC PROCEDURE TO ESTIMATE THE BENEFITS OF A PARTICULAR TYPE OF PERSONNEL CLASSIFICATION DECISION.
December 1, 1998... Recently, Alley and Darby (1995) proposed the use of Monte Carlo simulation methods to estimate the benefits of a subset of personnel classification decisions. The subset consists of decisions for which it is assumed that the available criterion estimates are both equi-correlated and...