Suggests that research on small group behavior, in a wide variety of settings, will be enhanced by rethinking alternative modes of enquiry, e.g., a focus on induction, a study of whole systems of human enterprise, the utilization of subjective experience, and trans-causal thinking. (Author/RC)

Contains seven articles discussing the application of meta-analysis to research on outcomes in psychotherapy. Provides a framework for understanding meta-analytic procedures and outlines major techniques. Reviews criticisms of integration of psychotherapy outcome research. Evaluates comparative therapy outcome research and discusses limitations of…

Problems with funding, politics, and policy formation for research in the social sciences raise the question of what results are expected from such research. Action based on research results usually does not follow closely the completion of research projects, and policy is often out of step with research. (MSE)

Discusses the current meager state of knowledge of sex differentiation in language variation and concludes that poor theorizing and inadequate methodology are to blame. Describes a study of pronoun usage by male and female Japanese speakers to show that research on sex differentiation in language is possible. (EKN)

Studies of second language acquisition often employ analytical concepts defined relative to the target language. This practice can be a serious obstacle to investigation of crucial questions about the nature of the learners' languages. This comparative fallacy's drawbacks are illustrated through a framework designed to study interlanguage…

Examines some problems with school effectiveness research, including small samples, indentification error, and inappropriate comparisons. Nevertheless, the article concludes that characteristics of effective schools can be useful in school improvement as long as facile solutions are avoided in favor of incremental long-term cultural change.…

Calls attention to limitations and dangers in routine application of multivariate analysis of variance (MANOVA), describes some alternative procedures, and laments the necessarily pervasive character of the statistical problem. (Author)

Critically reviews an earlier published study that assessed the sensitivity of the cloze procedure to intersentential integration. (AEA)

Recent research data contest Piaget's conclusion that preschool children are totally egocentric, incapable of taking different perspectives, and prevented from acting altruistically. Children are able to decenter when experiments enable children to use their knowledge of very basic human purposes, intentions, and interactions. (Author/DB)

Health education researchers have occasionally used measuring tools inappropriately. Test validity and the use of criterion-referenced tests are discussed in view of this need to improve health education research. Also described are handbooks containing measuring devices for the use of health education researchers. (CJ)

The gender-linked character of descriptive terms recurrent in studies reporting gender differences in children's play is examined. It is suggested that such terms are linked to gender as a socially determined classification, influencing observation and interpretation of gender-related behaviors. (Author/GK)

Describes the evolution of national educational assessment in the United States, its present stage of development, methodological and reporting problems, and approaches applicable to assessment data. Examines how proposed approaches can be applied in the California Assessment Program and in the National Assessment of Educational Progress. (MJL)

Different experimental techniques for empirical assessment of children's linguistic ability have produced conflicting evidence. A number of such contradictions are discussed, in the context of an analysis of the potential weaknesses, and consequent requirements, of both comprehension and production tasks. (A commentary by P. L. Harris appears on…

Notes that after 20 years of research on advance organizers there is little unequivocal evidence regarding their efficacy and optimal use. Discusses two aspects of this research that are probably responsible for these findings: lack of scientifically objective advance organizer construction and definition and the lack of objective and valid…

LISREL V computer program is applied to a weak quasi-experimental design involving the Head Start program, as a multiple analysis attempt to assure that differences between nonequivalent control groups do not confound interpretation of a posteriori differences. (PN)