Introduction to the Meta-Analysis Special Issue

Meta-analysis is a powerful research synthesis tool. Meta-analysts can advance theory and influence policy to a greater extent than individual researchers because their results synthesize the results of many individual studies. Meta-analysis may be particularly important in gifted education research because samples in individual studies are often small and context-bound, and the results of individual studies are typically quite heterogeneous. Meta-analysts can combine such studies to determine whether there are trends that hold across contexts and, through moderator analyses, indicate which variables are contributing to the heterogeneity. The articles we have included in this special issue provide examples of the power of meta-analysis to enlighten the field and to provide guidance for how to conduct a meta-analysis. More specifically, we have included four articles. First, we present two applied studies, one of identification methods and another on enrichment programming. A basic research study on the relationship between psychopathologies and creativity completes the substantive examples of the importance of meta-analysis for the field. Two of these studies also provide an illustration of the use of multilevel models to handle dependencies among effect sizes within studies. The special issue concludes with a methodological brief that provides guidance for gifted education researchers who would like to learn how to conduct a meta-analysis.

In the first study, Acar, Sen, and Cayirdag (2016) investigated the consistency between performance (e.g., intelligence, achievement, and creativity tests) and nonperformance methods (e.g., parent and teacher rating scales) of identifying gifted children to provide guidance on how to use these methods in designing identification procedures. Meta-analysis was appropriate because it allowed the researchers to combine the results of 35 studies and to examine two moderator variables: grade level and type of measure. The results suggested that the two types of measures identify different students so should be used simultaneously, not sequentially. In the moderator analyses, somewhat in contradiction to prior research, the researchers found that teacher ratings tended to be more consistent with performance ratings than parent ratings. This study suggests that nonperformance measures should not be used as a screening tool to identify children to be tested on performance measures and supports the simultaneous use of multiple measures to be sure that all gifted children in a given context are identified.

In the next study, Kim (2016) investigated the effects of enrichment programs on achievement and social/emotional development. Dr. Kim identified 26 studies that met the inclusion criteria. Meta-analysis was appropriate for this study because there have been a sufficient number of studies of enrichment programs to conduct a meta-analysis, yet the programs studied and the ages of participants were quite heterogeneous. Meta-analysis enabled the researcher to determine that across this wide variety of programs there were strong positive effects on academic achievement and moderately positive effects on social emotional development. Moderator analyses indicated that both the type of program and grade level of the students made a difference, although all program types and age-groups exhibited positive effects on the outcome variables. The most powerful intervention for increasing academic achievement was a summer residential program. High school students were the participants that were affected the most. For social/emotional development, the program type that was most effective also was a summer residential program, but the students most affected were middle school students. This study complements the earlier meta-analysis by Steenbergen-Hu and Moon (2011) of acceleration. There was no overlap between the studies included in this study and the earlier meta-analysis of studies of acceleration. Taken together, the two meta-analyses provide nuanced support for gifted programming and can be useful in summarizing the effectiveness of gifted programs for policymakers.

In the final meta-analysis, Hyeon Paek, Abdulla, and Cramond (2016) investigated the relationships among three common types of psychopathology (attention-deficit hyperactivity disorder, anxiety, and depression) and little-c creativity. Their pool of studies was larger than that of the previous two studies, so it allowed for the examination of five moderators: assessment methods for both psychopathology and creativity, age, gender, and level of intelligence. Again, the studies they included in their meta-analysis were quite heterogeneous, making the moderator analyses particularly important. They found no overall relationship between these three psychopathologies and creativity, perhaps because of the tremendous heterogeneity in the design and results of the studies included (e.g., results of the studies ranged from −.97 to +.95). However, they were able to determine that the results varied systematically by assessment method and level of intelligence. A great deal of the heterogeneity in these studies could be explained by assessment method, suggesting that assessment methods for both psychopathology and creativity are very important in determining relationships among these variables. There were higher relationships with creativity for nonclinical populations, those whose creativity was assessed by person rather than by process, and those with higher intelligence. The results suggest that severe psychopathology is likely to inhibit creative processes and that high intelligence can facilitate creativity in the presence of psychopathology.

For readers interested in conducting a meta-analysis, we include a methodological brief that provides guidelines for conducting one in the field of gifted education. Steenbergen-Hu and Olszewski-Kubilius (2016) explain why meta-analysis is needed in gifted education and provide a brief history of meta-analyses that have been conducted in the field. Then they provide details on the various steps in conducting a meta-analysis including search procedures, coding, primary analyses followed by heterogeneity and moderator analyses, and reporting of results. Many examples are used to illustrate complex concepts, and a detailed list of resources for further study is provided. It is hoped that this methodological brief will encourage more researchers to conduct meta-analyses in the future.

In conclusion, the articles in this special issue illustrate the importance of meta-analysis for the field of gifted education and the value of meta-analysis in synthesizing both applied and basic research in the field. In order to conduct a quality meta-analysis, especially one with moderator analyses, a sufficient body of high-quality research studies must be available. Hence, indirectly, the issue provides encouragement for more high-quality, quantitative research in the field to build the foundational studies for meta-analysts to synthesize.