A procedure for testing heterogeneity of variance is developed which generalizes readily to complex, multi-factor experimental designs. Monte Carlo studies indicate that the Z-variance test statistic presented here yields results equivalent to other familiar tests for heterogeneity of variance in simple one-way designs where comparisons are…

Piagetian research provides the focus of this article in which Professor Thomas C. O'Brien interviews Professor Hermina Sinclair. Aspects of children's thinking that prefigure scientific theory construction are described and the influence of schooling in inhibiting or suppressing a theory-building human tendency is briefly discussed. The…

Too often researchers rely upon the classical normal theory parametric tests to analyze non-normal data, even though the tests may not be robust to violations of that assumption. Fligner's class of two-sample tests for scale is an important development because the test is distribution-free and has desirable properties. This paper outlines the…

The results of a study of find alternative techniques for testing distributional normality are presented. A group of statistical techniques--some established and some new--were compared using empirical techniques. One new technique which appears to have higher power than the Lilliefors test was subjected to a better definition. Distributions under…

A method is proposed and illustrated for estimating the degree to which a factor rotation to a hypothesized target represents an improvement over rotation to a random target. (Author)

Configural frequency analysis (CFA), a new method for identifying types, is illustrated numerically. Relations to latent class analysis and to factor analysis are discussed. It is suggested to use CFA as a type-defining method instead of factor analysis if the variables are linked not only by first but also by higher-order associations. (RC)

A new test for increasing or decreasing trend in a set of observations is presented. Rather than making specific paired comparisons, as in the Cox and Stuart tests for trend, the user of the new test makes the comparisons of all possible pairs of observations. (Author)

Preparation of this paper aided by Grant MH-11521 from the National Institute of Mental Health, U.S. Public Health Service.

A theory of policy implementation, as opposed to policy results, would allow analysts to estimate how government programs would actually work. To help achieve such a theory, a general framework is presented along with testable propositions about the implementation of federal programs. Implementation is defined as including two components: the…

Several aspects of interview research heretofore receiving little attention are discussed. A brief description of the different types of interview formats and levels of analysis is presented. Following a discussion of the problem of analyzing protocol data, some suggestions are offered about analysis procedures that derive from constructionist…

Researchers often use the analysis of variance to test hypotheses about the means, followed by a multiple comparison technique when the F-test is significant. The technique is this study was developed by Newman (1939) and Keuls (1952). A flaw in the rationale underlying this technique was evaluated to determine whether the flaw is sufficiently…

This study shows that the test statistic for Analysis of Covariance (ANCOVA) has a noncentral F-districution with noncentrality parameter equal to zero if and only if the regression planes are homogeneous and/or the vector of overall covariate means is the null vector. The effect of heterogeneous regression slope parameters is to either increase…