C. Thomas Holmes

The Effects of Nonpromotion on Elementary and Junior High School Pupils: A Meta-Analysis:

In this study data from all studies identified as meeting the selection criteria were mathematically integrated to determine the effect of grade-level retention on elementary and/or junior high school pupils. When each effect size calculated was treated equally, a grand mean effect size of −.37 was obtained indicating that, on the average, promoted children scored .37 standard deviation units higher than retained children on the various outcome measures. When the effect sizes within each study were first averaged so that each study could be given equal weight, a grand mean of −.34 was obtained. By using the effect sizes from only those studies in which the promoted and nonpromoted pupils had been matched, a grand mean of −.38 was calculated. The high degree of consistency in these measures lends credibility to the validity of these findings. In addition to the grand means, effects sizes were calculated on various dependent variable measures, including academic achievement (further subdivided into various areas), personal adjustment (which included self-concept, social adjustment, and emotional adjustment), and attitude toward school, behavior, and attendance. In all cases, the outcomes for promoted pupils were more positive than for retained pupils.

Effect Size Estimation in Meta-Analysis

As the rate of growth of the literature in education has grown exponentially, it has become recognized that methods of integrating like studies mathematically are needed. Meta-analysis is one technique which has been developed to fulfill this need. For the application of meta-analysis, it is important that as many studies as possible be included. Methods for estimating effect sizes when complete data are not reported are presented. When not precise, these methods provide a conservative estimate and, therefore, allow for the inclusion in a meta-analysis of relevant studies whose data might otherwise be discarded.

ACHIEVEMENT GAINS OF LOW-ACHIEVING STUDENTS USING COMPUTER-ASSISTED VS REGULAR INSTRUCTION '

One school used computer-assisted instruction for 63 low-achieving students in Grade 5. Another school used traditional instruction for 48 low-achieving students in Grade 5. Higher posttest scores for the former were attributed to daily computer-assisted instruction.

Implications of Regional Cost Adjustments to School Finance Plans

The purpose of this article is to present evidence against the use of regional cost adjustments in state schoolfinance plans. Basic assumptions underlying proposals to compensate for the presumed effects of differences among districts in the prices paid for educational resources are contrasted with research evidence.

A Tool for Principals A Method for Assessing Teacher Motivation

Administrators now have a tool for efficiently assessing critical aspects of teacher motivation, according to these writers. They ex plain this as sessment instru ment on these pages.

Effects of Ability Grouping on Academic Achievement and Self-Concept of African-American and White Students

persist. A great deal of sentiment and research, both pro and con, about the merits of ability grouping are recorded in the literature (Noland and Taylor 1986). Though recent syntheses and reviews of past research criticize many research designs and do not support the practice of ability grouping in increasing academic achievement (Drahozol and Hanna 1978; Epstein 1980; Esposito 1971; Kirp and Youdof 1974; Nevi 1987; Persell 1976; Slavin 1987, 1988, 1990), the average educator seems to be convinced of its merits. For this reason, the practice of ability grouping continues throughout the United States (Glickman 1991). Theoretically, the homogeneous ability grouping of students for instruction should help reduce the wide range of variability in instructional classes and permit teachers to work more effectively with groups of children nearly equivalent in academic abilities. Achievement testing is the major tool for assigning ability levels. Because test scores tend to be different for students along lines of race and social class (Epstein 1980; Hawley 1981; Oakes 1985; Rossell et al. 1981), one unfortunate side effect of ability grouping has been the concentration of African-American students into

Improving Test Scores--How to Develop Comprehensive Plans.

To ensure an orderly process, the writers identify the steps one should take in de veloping com prehensive plans for improving stu dent achieve ment.