When the process of publication favors studies with small p-values and large effect estimates, combined estimates from many studies may be biased. This paper describes a model for estimation of effect size when there is selection based on one-tailed p-values. (SLD)

A Monte Carlo study considers the use of meta analysis with the Solomon four-group design. Experiment-wise Type I error properties and the relative power properties of Stouffer's Z in the Solomon four-group design are explored. Obstacles to conducting meta analysis in the Solomon design are discussed. (SLD)

Problems and illustrations are presented from a meta-analysis of crime reduction programs. Eight criteria are specified for including studies in a meta-analysis, and some problematic studies are discussed using the criteria. Rules for handling problem studies in order to retain consistency throughout the analysis are discussed. (SLD)

Conducted meta-analysis on 67 studies of impact of career education on academic achievement in elementary and secondary schools. Found evidence of small positive effect, with career education interventions improving student academic achievement levels an average of 0.16 standard deviations over alternative or control conditions. Results support…

Examined attention given in 185 studies to individual differences and empirical patterns of variability reported in studies presenting measures of dispersion (43 percent of gerontological studies reviewed and 24 percent of developmental studies reviewed). Most gerontological studies presenting data reported increases in variability with increasing…

Effects of choice of data analysis program on outcomes of meta-analysis were studied by considering the extent to which four commonly used Schmidt and Hunter validity generalization-based meta-analytic software programs produce identical outcomes. Differences were minor and did not influence outcomes. (SLD)

Arguments supporting the use of the exploratory case method in evaluating reviews of multivocal literature are repeated, and points of general agreement with M. Q. Patton and R. F. Elmore are reviewed. Disagreement is expressed with R. K. Yin about the value of grounded theory in analyzing multivocal literature. (SLD)

Discusses issues attendant to meta-analytic reviews of the research literature concerning factors influencing persuasive effectiveness. Discusses issues connected with (1) the literature search (such as identification of relevant studies); and (2) the analysis of the collected studies (including extraction of effect sizes from studies, analyses of…

Twenty-four studies on creativity and bilingualism are compared in a table and text. Twenty of the studies found bilinguals to be more creative than monolinguals. Four studies found an interaction between sex and bilingualism. (DB)

Because of the importance of single-subject studies in mental retardation research, a meta-analysis of the literature examining interrater agreement for visual analysis of single-subject data was conducted. Analysis revealed an overall itinerant agreement value of 0.58 for 14 studies representing 789 raters. Effects of data type, design type,…

Argues that there are two central reasons for concern about the adequacy of M. Allen's metanalysis of the research literature concerning the relative persuasive effectiveness of one-sided and two-sided messages: (1) the classification of the message types in a number of studies is questionable; and (2) the literature search underlying the review…

Responds to an analysis by D. O'Keefe (in the same issue of this journal) of a research study by the author. States that most of what O'Keefe suggests contains merit but that acceptance of the suggestions still generates a conclusion that two-sided refutational messages constitute the most persuasive form of message. (SR)

Conducted meta-analysis of 163 randomized trials to examine differences in outcome associated with different theoretical orientations, differences between marital and family therapies versus individual therapies, clinical significance of therapy outcome, differences between marital versus family therapies in both outcomes and problems treated, and…

An analysis of 50 years of research reveals that direct influences such as classroom management affect student learning more than indirect influences such as policies. Findings based on standardized tests and essay examinations showed student aptitude as the most influential variable, followed by classroom instruction and climate, context, program…

Analysis of 20 research studies of critical thinking in nursing, 1977-1995, found that the Watson Glaser Critical Thinking Appraisal was used in 18. There was no consistent evidence that nursing education increases critical thinking. Research design flaws, inconsistent definition of critical thinking, and lack of an appropriate measurement tool…