Skip to content
ascd logo

Log in to Witsby: ASCD’s Next-Generation Professional Learning and Credentialing Platform
December 1, 1994
Vol. 52
No. 4

Synthesis of Research / The Effects of Inclusion on Learning

Recent research, legislation, and a court decision support the case for inclusion of special-needs students in regular classrooms.

Of the many issues related to the inclusion or integration of children with disabilities into regular classrooms, none is more important than the effects on students' learning and social relations with classmates. Following is a summary of studies on this topic.

Classifying Special-Needs Students

A report of the National Academy of Sciences (Heller et al. 1982) prompted early research on inclusion. The panel found classification and placement of children in special education ineffective and discriminatory. It recommended that children be given noninclusive or extra-class placement for special services only if (a) they can be accurately classified, and only if (b) noninclusion demonstrates superior results.
Several major studies in the 1980s showed that it is difficult to classify children accurately and that the classification systems for placing students in special programs are seriously flawed (Reschly 1987, Wang et al. 1992, Ysseldyke 1987). Even if these classification problems could be solved, what evidence exists to show the benefits of noninclusive educational practices compared to those of inclusive practices? This informational need is pressing, particularly in light of recent legislation—such as Goals 2000 and the Improving America's Schools Act of 1993—that call for an inclusive approach to achieving higher educational outcomes for all students, including those with special needs.

Adopting New Research Methods

Until very recently, the methods available to social scientists for establishing the relative merits of educational practices were somewhat limited, and some would even say counterproductive for real knowledge generation (Schmidt 1992). Unlike the physical sciences, in which one experiment can sometimes demonstrate a significant effect, social science relies on an accumulation of evidence. This presents a challenge to social scientists because it is virtually impossible to synthesize the results of a large number of studies without resorting to structured, statistical methods.
More explicit and rigorous techniques that use statistical and experimental methods today can supplement or replace traditional narrative-based methods of research reviews. The general term for these more objective research integration methods is meta-analyses. The structured nature of meta-analytic techniques helps reduce the potential for reviewer bias and provides a statistical mechanism for handling large amounts of data. In fact, the application of meta-analytic research methods to social science issues has produced effect estimates that in some well-developed education areas are as consistent as those found in areas of physical science (Hedges 1987). As such, meta-analysis provides a basis for a more rational approach to decision making, especially decisions involving highly emotional issues such as the education of special-needs students.
Three meta-analyses in the educational literature address the issue of the most effective setting for the education of special-needs students (Baker 1994, Carlberg and Kavale 1980, Wang and Baker 1985–1986). These meta-analyses generate a common measure, called an effect size. The effect sizes shown in Figure 1 were estimated from the available research that compared the effects of inclusive versus noninclusive educational practices for special-needs students. These effect sizes demonstrate a small-to-moderate beneficial effect of inclusive education on the academic and social outcomes of special-needs children. Academic outcomes are learning measures generated by standardized achievement tests, whereas social outcomes are obtained by self, peer, teacher, and observer ratings of special-needs students' success in relating with others in the classroom.

Synthesis of Research / The Effects of Inclusion on Learning-table


Carlberg and Kavale

Wang and Baker


Year Published-1980(1985–1986)-1994
Time PeriodPre-19801975–19841983–1992
Number of Studies501113
Academic Effect Size0.150.440.08
Social Effect Size0.110.110.28

The average effect sizes range from 0.08 to 0.44 (and all are positive), which means that special-needs students educated in regular classes do better academically and socially than comparable students in noninclusive settings. The average of the six inclusion effects, 0.195, is near the average effect for effective instructional practices (Walberg 1986). Although estimated effects vary across individual studies, they have rarely shown negative effects for inclusion. The meta-analyses attempted to discern whether other factors influenced the effect size estimates related to inclusion (for example, type of special-needs student, or grade level). These secondary analyses revealed no consistent pattern of results that differ from the overall effects. The effects of inclusion are positive and worthwhile, but they are not huge. To reduce the gap between special and regular students requires both inclusion of special-needs students and effective educational methods for all students.

Making Policy Decisions

In the early 1990s, social and historical influences contributed to the creation of the separate system for special-needs students. Although well intentioned, the separate system has not resulted in improved learning for such students. Considerable evidence from the last 15 years suggests that segregation of special students in separate classrooms is actually deleterious to their academic performance and social adjustment, and that special students generally perform better on average in regular classrooms.
These findings are underscored by recent educational legislation and a court decision, Oberti v. Clementon 1993. In Oberti, the federal court upheld the right of children with disabilities to be educated in regular classrooms with their nondisabled peers. More specifically, this judicial decision clearly placed the burden of proof on school districts that continue to remove special-needs students from regular classes. That is, school districts, as mandated by the National Academy of Sciences panel, need to demonstrate that a segregated special education placement is the best educational approach for the individual students when making such placement decisions.
Parents and legal experts are increasing their demands for schools to address the scientific and legal basis for noninclusive practices. As schools are challenged to effectively serve an increasingly diverse student population, the concern is not whether to provide inclusive education, but how to implement inclusive education in ways that are both feasible and effective in ensuring schooling success for all children, especially those with special needs.

Baker, E. T. (1994). “Meta-Analytic Evidence for Non-Inclusive Educational Practices: Does Educational Research Support Current Practice for Special-Needs Students?” Doctoral diss., Temple University, Philadelphia.

Carlberg, C., and K. Kavale. (1980). “The Efficacy of Special Versus Regular Class Placement for Exceptional Children: A Meta-Analysis.” Journal of Special Education 14: 295–309.

Hedges, L. V. (1987). “How Hard Is Hard Science, How Soft Is Soft Science? The Empirical Cumulativeness of Research.” American Psychologist 42: 443–455.

Heller, K., W. Holtzman, and S. Messick. (1982). Placing Children in Special Education: A Strategy for Equity. Washington, D.C.: National Academy of Science Press.

Oberti v. Clementon, 995 F.2d 1204, (3rd Cir. 1993).

Reschly, D. J. (1987). “Learning Characteristics of Mildly Handicapped Students: Implications for Classification, Placement, and Programming.” In Handbook of Special Education: Research and Practice: Vol. 1. Learner Characteristics and Adaptive Education, edited by M. C. Wang, M. C. Reynolds, and H. J. Walberg, pp. 253–271. Oxford, England: Pergamon Press.

Schmidt, F. L. (1992). “What Do Data Really Mean? Research Findings, Meta-Analysis, and Cumulative Knowledge in Psychology.” American Psychologist 47: 1173–1181.

Walberg, H. J. (1986). “Synthesis of Research on Teaching.” In Handbook of Research on Reaching, 3rd ed. edited by M. C. Wittrock. New York: Macmillan.

Wang, M. C., and E. T. Baker. (1985–1986). “Mainstreaming Programs: Design Features and Effects.” The Journal of Special Education 19: 503–521.

Wang, M. C., H. J. Walberg, and M. C. Reynolds. (1992). “A Scenario for Better—Not Separate—Special Education.” Educational Leadership 50, 2: 35–38.

Ysseldyke, J. E. (1987). “Classification of Handicapped Students.” In Handbook of Special Education: Research and Practice: Vol. 1. Learner Characteristics and Adaptive Education, edited by M. C. Wang, M. C. Reynolds, and H. J. Walberg, pp. 35–58. Oxford, England: Pergamon Press.

Edward T. Baker has been a contributor to Educational Leadership.

Learn More

ASCD is a community dedicated to educators' professional growth and well-being.

Let us help you put your vision into action.
From our issue
Product cover image 194214.jpg
The Inclusive School
Go To Publication