Meta-Analyses of Gifted Education

Ability Grouping

The practice of ability grouping, in which students are placed in learning groups with others of similar aptitude, has generated extensive controversy in the field of education. Arguments in favor of the practice include statements that such grouping helps teachers more effectively meet students' needs by narrowing the focus necessary in delivering material, and that it helps facilitate curricular modifications appropriate to the type of student being taught. Arguments against the practice include concerns that grouping reduces the performance of students in lower-ability groups by reducing the expectations their teachers have of them, that it discriminates against minority students by disproportionately placing them in lower-ability groups, and that it promotes inequity through unfair allocation of resources (e.g., assigning the best teachers to the highest-ability groups).

Part of the reason why controversy persists is that research results vary across studies. Meta-analyses have been used in an attempt to zero in on well-designed studies and then quantify their results to seek a scientific answer to questions about ability grouping. Typically, these meta-analyses focus on studies that measure outcomes using scores on standardized achievement tests. Some also include studies of social and emotional outcomes, such as general self-concept or self-esteem, attitudes toward the specific subject matter taught, or attitudes toward school in general. Unfortunately, even the results of meta-analyses have been controversial, with some authors concluding that ability grouping is ill advised, and others concluding that it is a well-supported educational strategy.

To some extent, differences in conclusions are related to differences in the criteria used to qualify studies for inclusion in a meta-analysis. For instance, some researchers prioritize or even restrict studies to those that use random assignment of students to grouped versus ungrouped classes. Random assignment of participants to research conditions is a basic requirement of experimental design, and is a key element in allowing conclusions about cause and effect. For some types of ability grouping, however, random assignment does not accurately reflect the way in which students are selected for participation in groups. For example, some meta-analyses include studies of accelerative groups for gifted learners. In practice, students selected for acceleration are identified by more than just their ability level. Factors such as personality style, motivation, friendship patterns, and age (to name just a few) often are considered as well. Therefore, randomly assigning high-ability students to accelerated versus unaccelerated classes for the purpose of research may not reflect the reality of the practice as it is used in school settings.

...