The present study investigated the effectiveness of two short relevance interventions (writing a text or evaluating quotations about the utility of mathematics) using a sample of 1,916 students in 82 math classrooms in a cluster randomized controlled experiment. Short-term and sustained effects (6 weeks and 5 months after the intervention) of the…

Paper proposes a set of techniques for measuring the importance" of each independent variable in a multivariate causal law (i.e., a law showing the combined effect of several independent variables on a single dependent variable). (Authors)

(See also TM 501 087, TM 501 088, TM 501 090.)

(See also TM 501 087, TM 501 088, and TM 501 089.)

Regions of significance in aptitude-by-treatment-interaction studies are examined by the traditional statistical approach and an alternative approach which integrates: (1) testing for the parallelism of two or more regression lines; (2) testing for their identity; and (3) Scheffe's theorem. (Author/RL)

The causal modelling technique is extended to nonrecursive causal models that involve feedback and/or reciprocal causation. Three least squares techniques for estimating parameters are described, and data from an empirical study illustrate the causal modelling technique. (Author/GDC)

The purpose of this study was to compare five predictor models, including two least-square procedures, two probability weighting (semi-Bayesian) methods, and a Bayesian model developed by Lindley. (See also TM 501 088, TM 501 089, and TM 501 090) (Author/NE)

A series of comments, corrections, and errata are presented regarding homogeniety of regression (see also EJ 001 198). (CK)

Kaufman and Sweet's article on the regression analysis of unbalanced factorial designs (EJ 111 767) is reviewed. A number of errors are noted, and relevant literature is cited. (GDC)

(See also TM 501 087, TM 501 089 and TM 501 090.)

Contrary to Anderson (EJ 187 936), his rule for equation identification is a necessary but not sufficient condition; furthermore, the choice of two-stage or ordinary least squares depends on results and not on methodological properties of estimators. Modification of Anderson's rule and a means for choosing between estimates is offered. (Author/CP)

Similarities between regression analysis and analysis of variance are noted and it is argued that the former has advantages over the latter. It is also argued that canonical correlation analysis is more suitable than factor analysis in certain cases. The argument is illustrated with four recent pieces of educational research. (DG)