The use of planned, or "a priori," and unplanned, or "post hoc," comparisons to isolate differences among means in analysis of variance research is discussed. Planned comparisons typically involve weighting data by sets of "contrasts." Planned comparison offer more power against Type II errors. In addition, they force…

In his early exposition of path analysis, Duncan (1966) noted that the method "provides a calculus for indirect effects." Despite the interest in indirect causal effects, most users treat them as if they are population parameters and do not test whether they are statistically significant. Sobel (1982) has recently derived the asymptotic…

See TM 501 202-3 and EJ 060 883 for related articles. (MLP)

Treats the problem of testing an ordered hypothesis based on the ranks of the data. Statistical procedures for the randomized block design with more than one observation per cell are derived. Multiple comparisions and estimation procedures are included. (Author/RC)

Log linear models are proposed for the analysis of structural relations among multidimensional developmental contingency tables. Model of quasi-independence are suggested for testing specific hypothesized patterns of development. Transitions in developmental categorizations are described by Markov models applied to successive contingency tables. A…

The purposes of this study are: (1) to demonstrate the superiority of simple ridge regression over ordinary least squares regression through theoretical argument and empirical example; (2) to modify ridge regression through use of the variance normalization criterion; and (3) to demonstrate the superiority of simple ridge regression based on the…

JOINTLY CONSIDERING THE FACTORS OF SAMPLE SIZE (N-10 TO N-1000) AND THE STANDARD DEVIATION OF THE CRITERION MEASURE (SD-1 TO SD-25), A TABLE SHOWS THE DIFFERENCE BETWEEN TREATMENTS NEEDED FOR STATISTICAL SIGNIFICANCE AT THE .05 LEVEL. THUS, IF THE EDUCATIONAL RESEARCHER HAS APPROPRIATE INFORMATION ON PAST PERFORMANCE, HE CAN CONSULT THE TABLE TO…

The advantages of Goodman's Gamma, a measure of association, are discussed in reference to the Pearson coefficient of contingency. Both theoretical and practical advantages and disadvantages are discussed. An empirical comparison of the two measures shows that gamma detects significant relationships which chi square does not, and that gamma is…

Techniques for detecting synergistic effects in analysis of variance designs are presented and discussed. These techniques make it possible to apply some kinds of theoretical insights to the data analysis phase of a study: either by seeking synergistic effects implied or predicted by theory, or by seeking evidence of synergies as alternative…

Multivariate models are demonstrated to analyze repeated measures profile and growth curve data when univariate or multivariate mixed model assumptions are not tenable. Standard mixed model tests are recovered from certain multivariate hypotheses. The procedures are illustrated using numerical examples. (Author/RC)

Games (1973) criticized the Marascuilo and Levin (1970) "nested" or simple effects design as a poor substitute for the usual logic of a factorial analysis of variance. Marascuilo and Levin's use of "nesting" is completely idiosyncratic and contrary to well-established usage, thus confusing the student. Proper usage is…

Three exact probability tests, counterparts of t tests for single sample, independent samples, and dependent samples, are described for data obtained from ratings on "m" scales by a single rater or on a single scale by "n" raters. (LMO)

This study compared two approximations for normalizing noncentral F distributions: one based on the square root of the chi-square distribution (SRA), the other derived from a cube root of the chi-square distribution (CRA). The CRA was superior, and generally provided an excellent approximation for noncentral F. (Author/BW)