Multiple comparisons involve the examination of which group or groups are actually different from other group(s) in analysis of variance results. Such comparisons usually involve one-way analysis of variance. This monograph discusses designs more complex than one-way designs. (JKS)

An experimental sampling procedure for communities on which coal had been surface-mined yielded missing cells and caused the number of degrees of freedom to be N instead of the usual N minus one. The apparent discrepancy is explained, and a solution to the problem is presented. (Author/JKS)

Three methods of handling disproportionate cell frequencies in two-way analysis of variance are examined. A Monte Carlo approach was used to study the method of expected frequencies and two multiple regression approaches to the problem as disproportionality increases. (Author/JKS)

Issues in the application of multiple regression analysis as a data analytic tool are discussed at some length. Included are discussions on component regression, factor regression, ridge regression, and systems of equations. (JKS)

Repeated measures designs often involve dichotomization of a continuous variable in order to be amenable to the analysis of variance nature of such designs. An alternative to that approach wherein the independent variable is kept continuous is presented. (JKS)

Repeated measures designs often involve dichotomization of a continuous variable in order to be amenable to the analysis of variance nature of such designs. An alternative to that approach wherein the independent variable is kept continuous is presented. (JKS)

With even the simplest bivariate regression, least-squares solutions are inappropriate unless one assumes a priori that reciprocal effects are absent, or at least implausible. While this discussion is limited to bivariate regression, the issues apply equally to multivariate regression, including stepwise regression. (Author/CTM)

Of 50 students taking a graduate course in multiple regression analysis at a particular university, they tended to use multiple regression in their dissertations. (Author/JKS)

A Type VI error results from inconsistency between the researchers' question of interest and the statistical procedures employed to analyze the data. An example of a research problem is analyzed to show the increase in statistical power resulting from improved research design, using multiple regression instead of analysis of variance. (CTM)

A model for research on teacher effects in which the variance in student outcome post-test performance is attributed to pre-test perfomance; to separate construct domains of student, instructional, and teacher variables; and to interactions between variables in these three construct domains is presented and tested. (Author/JKS)

The inclusion of unmeasured variables in path analyses in educational research is discussed. The statistical basis for inclusion is presented, along with several examples. (JKS)