The comparability of experimental and comparison groups, problems in sampling and matching, existence of uncontrolled differences and their statistical treatment, and problems of client mobility and experimental control are discussed and illustrated by examples from an evaluation project. (Author)

Accurate knowledge of the institution's past policies and composition considered within a broader state, regional, or national historical context can be an advantage in coping with complex institutional concerns. Implications for researchers are discussed. (Author/LBH)

The author provides a brief view of issues and developments in the field of problem solving in the light of progress made as well as implications for the future. (MN)

Discusses developments in earthquake prediction programs. (SL)

This paper describes a two-stage sampling design for obtaining probability samples of the elderly and other specialized populations. Results indicate elderly individuals will respond, use of the design does not appear to increase sample bias, and the method is extremely inexpensive. (Author)

Advocates according a central role to context-relevant beliefs in the operation of attitudinal saliency through contrasting Delia et al's cognitive-constructivist analysis with Cronen and Conville's operant conditioning and aggression displacement accounts. (Author)

The field of comparative administration should relax its preoccupation with ecological theories and aggregate data technologies and move on to systematic field research in multiple settings. (Author)

The hypothesis that differences in definitions of perceived internality in Piagetian and locus of control viewpoints account for paradoxical findings was tested empirically using two independent samples of children. Results suggest that perceived internality depends on the aspects of means-ends beliefs on which one focuses.

Presents observations of a researcher on state of educational research. States that some issues and questions have not changed in 50 years. Cites the arguments over the difference between basic and applied research, and the hereditary basis of intelligence. Concludes with remarks about inappropriate data analysis, the use of technology, and the…

A two-stage generalized least squares model is developed for estimating the linear regression of an individual outcome on a group characteristic in studies of multilevel data. Results of this model are compared to the results of analytic methods, and formulas are developed for assessing the accuracy of the traditional approaches. (Author/JAZ)

Discusses several features of the feminist critique of objectivity (part of the ongoing project of feminist theory and practice), as well as two possible responses to the questions it raises. Comments on some of the implications of the feminist critique of objectivity for communication research and proposes criteria for feminist communication…

Resolves a hypothetical editorial dilemma of whether or not to reject a research manuscript on the grounds that the statistical test applied to the data may have been inappropriate. (HTH)

Bureau of Labor statistics prove that schools are not top-heavy with administrators, contrary to the myth and Secretary William Bennett's assertion. Administrators comprise 6.6 percent of school employees and public education ranks 28 out of 35 occupations in terms of the percentage of administrative personnel. Accounting and bookkeeping lead with…

Three self-report methods for rating secondary sexual characteristic growth were compared to physician ratings; the accuracy of self-reported height and weight was also assessed. The usefulness of the Pubertal Development Scale, Tanner ratings, and self-reported height and weight as adequate estimates of pubertal development is discussed. (PCB)

Defines structural equation modeling (SEM) and points out its relation to other more familiar data-analytic techniques, as well as some of the potentials and pitfalls of SEM in the analysis of developmental data. Discussion focuses on causal modeling, path diagrams, ordinary least-squares regression analysis, and powerful methods for model…