Effects of inclusive school settings for students in six Indiana school corporations were investigated. Results reveal that students without disabilities educated in inclusive settings made significantly greater academic progress in mathematics and reading. For students with disabilities, there were no significant differences in reading and math achievement across the comparison groups. However, a review of group means and the percentage of students making comparable or greater than average academic progress when compared to students without disabilities indicates a pattern in favor of inclusive settings. The academic progress of students with specific disability labels, namely, learning disabilities and mild mental handicaps, also supported inclusive education.
Although debates rage among educational scholars regarding the inclusion of students with disabilities in general education classrooms, one thing is clear: “It is not a fad that is going to go away” (Peltier, 1997, p. 234). It has now been 18 years since the former Assistant Secretary of the U.S. Department of Education Madeline Will outlined specific problems with the special education service delivery system and proposed a consolidation of categorical programs and general education to create an educational system that recognized and taught to the individual differences of all children in the general education classroom (Will, 1986). Educators, policymakers, and parents continue to seek a common understanding of Will's vision. The debate continues.
Critics of inclusion have noted the dearth of empirical research that supports the alleged benefits of inclusive environments. Skeptics charge that efforts to include students with disabilities in general education classrooms may result in the more able students experiencing boredom, while students with disabilities may experience extreme frustration when trying to keep up with the instructional pace. There is often concern on the part of educators that the achievement scores of all students in inclusion classrooms could decline (Brockett, 1994). Those who have argued against inclusion noted that many students with disabilities were pulled from the general education classroom because they were not being well-served in those environments (Kauffman, 1995). Daniel and King (1997) used a quasi-experimental design to determine the effects of students' placement versus nonplacement in an inclusive classroom. Discriminate analysis results indicate that (a) parents of students in the inclusion classes expressed a higher degree of concern with their children's school programs; (b) teachers and parents of the students in the inclusion classes reported more instances of behavior problems; (c) students in inclusion classes were more likely to experience gains in reading scores with no noteworthy differences for math, language, and spelling; and (d) students in inclusion classes reported lower levels of self-esteem. Manset and Semmel (1997) conducted a comparative review of eight model inclusive programs for elementary students with mild disabilities. They concluded that the evidence clearly indicates that a model of wholesale inclusive programming that is superior to more traditional special education service delivery models does not exist.
Advocates of inclusion maintain that inclusion is beneficial to all students in terms of academic and social growth. Indeed, many scholars believe inclusion to be an issue of social justice and that the burden of proof should fall upon the shoulders of those who wish to segregate students with disabilities. Advocates claim that academic achievement is enhanced when students with disabilities are held to the higher standards of a general education classroom. Rea, McLaughlin, and Walther-Thomas (2002) investigated the relationship between placement in inclusive and pull-out special education programs and academic and behavior outcomes for students with learning disabilities. Results indicate that students served in inclusive classrooms earned higher grades, achieved higher or comparable scores on standardized tests, committed no more behavioral infractions, and attended more days of school than did students taught in pull-out special education classrooms. Waldron and McLeskey (1998) used a curriculum-based measure to investigate the effects of an inclusive school program on reading and math achievement of students with learning disabilities. Results revealed that they made significantly more progress in reading and comparable progress in math when compared to students with disabilities educated in resource settings. Advocates further claim that there are social benefits associated with the inclusion of students with disabilities in general education classrooms. Tapasak and Walther-Thomas (1999) summarized the first year evaluation of an inclusive education program in an urban elementary school. At the primary level, results showed significant increases in self-perceptions of cognitive competence for both students with and without disabilities. Teacher ratings showed improved social skills for both primary and intermediate students with disabilities. Vaughn, Batya, and Schumm (1996) studied the social functioning of students with learning disabilities in second, third, and fourth grade who participated in inclusive environments for the entire school year. The evidence from their study, when compared with that of other studies in which researchers evaluated similar outcomes for students with learning disabilities in resource room settings, reveals that students in inclusive settings fare at least as well socially as students from previous studies in resource room settings. There was even some evidence that they may, in fact, fare better in that they do not demonstrate high levels of loneliness and do experience increases in the number of reciprocal friendships from fall to spring.
The presumption that children with disabilities be educated in general education classrooms with appropriate supports and services has recently been strengthened. The Individuals With Disability Education Act—IDEA (1997) requires that a detailed explanation be given to justify any exclusion of the child, for one or more periods or activities, from the general education classroom.
An increasing number of schools are adopting inclusive education models in which students with disabilities receive special education support services in general education classrooms (McLesky, Henry, & Hodges, 1999). Many disagreements about the merits of inclusive schools hinge on the lack of empirical evidence; studies are few and quite dated, comparing inclusive services and pull-out services that were prevalent in the 1970s through the mid-1980s (Rea et al., 2002). As more students are served in inclusive environments, and as educators continue to make school-based decisions that are best for all students, there is a need to evaluate achievement outcomes not only for students with disabilities, but also for students without disabilities who are receiving their education in these settings (Waldron, 1997).
In this study we investigated the effects of inclusive programs on the academic progress of students without disabilities and students identified with mild disabilities in six Indiana school corporations. Students' academic progress in reading and mathematics were compared using a curriculum-based measure, the Basic Academic Skills Sample (BASS).
We addressed the following guiding research questions: (a) How does the academic progress in reading and mathematics of students with mild disabilities who are educated in inclusive settings compare to the progress made by students who are educated in traditional resource/pull-out settings? (b) How does the academic progress of students with mild disabilities who are educated in inclusive and traditional resource settings compare to students without disabilities? How does the academic progress of students without disabilities who are educated in inclusive general education classrooms compare to the progress made in noninclusive general education classrooms?
Six school corporations/special education cooperatives from across the state of Indiana participated. These corporations/cooperatives were selected to represent various geographic regions of the state and to reflect school locations that were urban, suburban, and rural. From these school corporation/cooperatives, we selected inclusive and traditional elementary schools. The definition provided to school districts for selection of inclusive schools was strictly based on setting. An inclusive school was one in which students with disabilities received their reading and math instruction in a general education, age-appropriate classroom; a traditional school was one in which students with disabilities received their reading and math instruction from a special education teacher in a pull-out setting.
Demographic data were collected for each of the participating schools (see Table 1), including student population, ethnic composition, per pupil expenditure, number of students receiving free lunch, and identification rate for students with mild disabilities. Data analysis revealed a significant difference in per pupil expenditure between inclusive and traditional schools, t(1, 1032) = 9.92, p < .01. The effect of per pupil expenditure will be figured into the following analyses.
All students identified with mild disabilities in Grades 2 to 5 from 23 elementary schools participated in the study. This resulted in a sample size of 429 students with mild disabilities: 235 students (54.8% of the sample) were served in special education resource settings and 194 students (45.2% of the sample) were served in inclusive settings. Demographic data were collected on all students with mild disabilities (see Table 2) to ensure that the two groups (inclusive and traditional special education) were comparable on variables such as gender, grade, ethnic background, disability label, and special education services received. Scores on Verbal IQ, Performance IQ, Full Scale IQ, math achievement, and reading achievement tests were also obtained (recorded) for students with disabilities in both traditional and inclusive schools. A t-test analysis revealed no significant differences in any of these five measures across the two settings.
To obtain the sample of students without disabilities, we randomly selected 35 classrooms from the participating schools. Classrooms were selected to yield an equal number for Grades 2 through 5, half being inclusive general education classrooms and the other half, traditional noninclusive general education classrooms. Inclusive classrooms were in inclusive schools in which students with disabilities were included for reading and math instruction. Noninclusive classrooms were those in which students with disabilities were not included for reading and math instruction. Each school was asked to nominate classrooms for participation in the study and identified each classroom as inclusive or noninclusive. From this nominated group, one or two classrooms were randomly selected from each school to make up the representative state sample. This resulted in 606 students without disabilities being included in the study, with a comparable number representing each grade level. Demographic data were not collected for students without disabilities who participated in the study; only information required to match student fall and spring test scores based on class, grade, and school was used.
Academic Progress Measures
The academic progress of students was evaluated using a curriculum-based measure, the Basic Academic Skills Samples—BASS (Espin, Deno, Maruyama, & Cohen, 1989). Data are available to support the technical adequacy of the BASS when used with students in Grades 2 to 6 (Espin et al., 1989; Jenkins & Jewell, 1992). The BASS has been used frequently to measure the progress of students with mild disabilities in inclusive school settings (Klingner, Vaughn, Hughes, Schumm, & Elbaum, 1998; Waldron & McLeskey, 1998; Zigmond et al., 1995).
The BASS is a group-administered instrument designed to assess student achievement in the academic skill areas of mathematics and reading. The mathematics section consists of two 1-minute probes with a variety of mathematical problems, including addition, subtraction, multiplication, and division. Students are assessed according to the average number of digits correct on the two probes. The first math probe has a total of 100 digits; and the second probe, a total of 103. The highest possible score for a student on the BASS math is 101.5.
The reading section consists of a modified cloze procedure. Three reading passages are used, with every seventh word deleted and three choices offered to the students, only one of which makes sense in the passage. Students are allowed one minute for each probe, and scores are calculated by taking the total number of correct responses after a ceiling of three consecutive incorrect responses is reached. The three probes have a total of 24, 33, and 26 correct responses, respectively. The student's score is the sum of the number of correct responses across the three probes. The highest possible score on the BASS reading is 83. According to Jenkins and Jewell (1992), the BASS is a stable and valid instrument that correlates well with norm-referenced measures of academic achievement.
The BASS was administered to participating students in fall 1998 and spring 1999 to assess academic progress in reading and math during the course of one school year. One investigator was assigned to each of the six corporations/cooperatives, and she administered the BASS to all second through sixth grade participants, using standardized instructions (see Espin et al., 1989). Group administration of the BASS occurred in all general education classrooms that participated in the study. For students identified with disabilities, administration of the BASS occurred either in their inclusive general education classroom or as part of a small group in their special education resource room or another location in the school building. The procedures and times for administration of the BASS were the same for all students with and without disabilities included in the study. Total administration time for the reading and mathematics portions of the BASS was 15 to 20 minutes. Administration procedures were the same during both the fall and spring administrations.
A scoring protocol was used to score the two math probes and the three reading probes. Scoring was done among 3 individuals for triangulation. Score reliability across the math and reading probes was 90% or better.
Student achievement gains on the BASS were analyzed in three ways: (a) to determine whether significant differences existed in reading and math scores for the two comparison groups (inclusion and traditional resource/pull-out), (b) to examine the educational achievement of students without disabilities over the course of the school year, and (c) to compare the educational achievement of students with disabilities to that of students without disabilities (McLeskey & Waldron, 1995; Waldron & McLeskey, 1998; Zigmond et al., 1995).
Test Results With the BASS
Student without disabilities
Table 3 presents the pre- and adjusted posttest scores in math and reading BASS measures of the students without disabilities, who were educated in inclusive and traditional classrooms. Analysis of covariance (ANCOVA), covarying pretest scores, and per pupil expenditure was used to compare students without disabilities across the two settings. The results demonstrate that in both math and reading, students educated in inclusive schools made significantly greater progress than did their peers educated in traditional schools: for math, F(1, 605) = 23.74, p < .01; for reading, F(1, 605) = 6.83, p < .01.
Students with disabilities
Table 3 also presents the pre- and adjusted posttest scores in math and reading BASS measures of all students with disabilities and separately for those with learning disability and those with mild mental handicap who were educated in inclusive and traditional classrooms. Analysis of covariance, covarying pretest scores, and per pupil expenditure were used to compare students with disabilities across the two settings. The results reveal no significant differences between the students educated in inclusive settings and those educated in traditional settings regardless of disability.
Educational Achievement of Students Without Disabilities
Our second focus in this research was to examine the educational achievement of students without disabilities over the course of the school year. One method used in previous research is to compare “students' test standings relative to their grade level peer group at the beginning and end of the school year, to determine whether students actually start to catch up to their peers who are achieving at an average level” (Zigmond et al., 1995, p. 539). Standard scores (Z scores, in this case) were used to examine student progress and determine the percentage of students that made progress comparable to or greater than their typical grade level peers.
Using the procedure described above (Zigmond et al., 1995), we determined the percentage of students without disabilities in inclusive and traditional settings who made progress over the course of the year. In math, 58.8% of students without disabilities in inclusive schools made progress on the BASS as compared to 39.0% of students without disabilities in traditional schools. Comparing the two groups in reading, 50.7% of the students without disabilities in inclusive schools and 47.1% of students without disabilities in traditional schools made progress over the course of the school year.
Our third focus in this research was to compare the educational achievement of students with disabilities to that of students without disabilities (McLeskey & Waldron, 1996; Waldron & McLeskey, 1998; Zigmond et al., 1995). The method used to achieve this goal in previous research is to compare “students' test standings relative to their grade level peer group at the beginning and end of the school year, to determine whether students actually start to catch up to their peers who are achieving at an average level” (Zigmond et al., 1995, p. 539). Standard scores (Z scores, in this case) were used to examine student progress and determine the percentage of students that made progress comparable to or greater than their typical grade-level peers.
Educational Achievement of Students With Disabilities
Using the procedure described above, we found that 43.3% of students with disabilities who were educated in inclusive classrooms made progress comparable to or greater than the progress made by students without disabilities in math. In comparison, 35.9% of the students with disabilities who were educated in traditional or resource programs made comparable or greater progress in math. In reading, 45.9% of students with disabilities educated in inclusive settings and 41.9% of those educated in pull-out resource programs made comparable or greater progress.
Students with learning disabilities
The results for students with learning disabilities are comparable to those obtained for all students with mild disabilities included in the study. In math, 41.7% of the students with learning disabilities in inclusive settings and 34.0% of the students with learning disabilities in traditional settings made progress comparable to or greater than their peers without disabilities. In reading, a comparable percentage of students with learning disabilities made progress in inclusive and traditional settings (48.2% and 47.8%, respectively). These percentages clearly indicate that the difference in math gains is larger than that in reading gains across the two settings.
Students with mild mental disability
The difference across the inclusive and traditional settings was even more pronounced for students identified with mild mental handicaps included in the study. A greater percentage of students with mild mental disabilities educated in inclusive classrooms made progress in math and reading than the students with mild mental disabilities educated in traditional classrooms. In math, 50.0% of the students with mild mental disabilities in inclusive settings made progress as compared to 37.7% in traditional settings. In reading, 40.0% of the students in inclusive settings and 29.5% in traditional settings made progress comparable to or greater than their peers without disabilities.
The results of this investigation reveal that students without disabilities educated in inclusive settings made significantly greater academic progress in both reading and mathematics than did students without disabilities in traditional classrooms. For students with disabilities, there were no significant differences in reading and math achievement across the comparison groups. However, a review of group means and the percentage of students making comparable or greater than average academic progress when compared to students without disabilities indicates a pattern in favor of inclusive settings. This finding was also supported when considering the academic progress of students with specific disability labels, namely, learning disabilities and mild mental handicaps. These results are similar to those of previous investigators, who have found small or no significant differences on measures of academic achievement for students with mild disabilities in inclusive classrooms when compared to students who were placed in more traditional special education classes (Waldron & McLeskey, 1998).
These results speak well for the inclusive school programs in the six Indiana corporations/cooperatives involved in this study and the positive impact they have on the academic achievement of students with and without disabilities. This investigation makes it clear that for students with mild disabilities, the inclusive school programs in the six participating districts provided an instructional experience that was at least as good, and in many cases better, than the education these students would receive in a traditional resource/pull-out setting. In addition, if students with disabilities make comparable progress in the two settings, one can argue that they should be educated in the setting that best meets the letter and intent of IDEA 1997 and the least restrictive environment provision.
We are encouraged that the differences across settings were even more pronounced for students identified as having a mild mental disability. Often, it is this group of students that schools assume need a separate, special education classroom in order to attain basic skills in the area of reading and math. Results of the present study show that in the six Indiana school districts, students with mild mental disabilities clearly benefit from inclusive classroom instruction in reading and math.
One can also conclude that clear achievement benefits accrue to students without disabilities who receive their education in inclusive general education classrooms. One of the most persistent myths about inclusive education has been that the presence of students with disabilities in general education classrooms interferes with the academic achievement of students without disabilities (Staub & Peck, 1994). One of the strongest conclusions from this study is that, in fact, students without disabilities in inclusive classrooms made significantly greater progress in reading and math than did their peers in noninclusive classrooms. Although individual classrooms were not analyzed in this study, other researchers have speculated that benefits to typical students are likely the result of additional supports provided in inclusive classrooms to all students. The assumption can be made here that the purposes of inclusion are highly relevant to the needs of all children (Staub & Peck, 1994).
We note that although we believe that the results from this study clearly support inclusive school settings, we also acknowledge our disappointment in the fact that less than half of the students with disabilities made as great or greater progress than did their peers without disabilities in either setting. This is not acceptable, regardless of the setting.
Limitations of the Study
Although in this study we gathered data on the achievement of students in inclusive and traditional elementary schools, we did not investigate the beliefs, values, understandings, and practices of the educators or analyze individual classrooms. Questions as to whether the placement setting is the critical factor rather than the quality of instruction within the setting are valid. In addition, the schools, whether inclusive or traditional, may not be a representative sample of the schools across the state, making it difficult to draw conclusive inferences about the results.
Implications for Practice
Nearly 30 years ago, political, social, and historical influences contributed to the creation of a separate system of education for students with disabilities. Though well-intentioned, the separate system has resulted in fragmentation, separation, and a significant increase in the number of students identified as “disabled.” This investigation is broad-based with a large sample size and, coupled with other recent research, adds to the growing body of evidence that supports inclusive schools. The results from this study for both groups of students would indicate a need for schools to begin to spend the time and resources necessary to develop quality inclusive programs for all students. The focus and discussion should begin to shift from whether to provide inclusive education to how to develop and implement quality inclusive classrooms that are effective in ensuring school success for all children.
As mentioned earlier, we have concern that the number of students with disabilities making as great or greater academic progress compared to their peers who do not have disabilities was less than half. We must do better. In an age of high stakes accountability and the inclusion of students with disabilities in state and districts assessments, it becomes even more imperative that all students have access to high expectations, rich curricula, and varied instructional strategies for learning. Schools must collect and use well the various achievement data available to them to ensure that the achievement gap between students with and without disabilities closes.
This study also has important implications for teacher education programs. As schools continue to develop classrooms that meet the needs of all students, general and special educators will have to expand their repertoire of skills and shift away from traditional roles. Teacher education programs must restructure to merge professional training programs so that general and special educators can participate and learn together the skills necessary to teach to all students.
Finally, there are implications for policymakers. As more schools accept the challenge of educating students with disabilities in inclusive classrooms, both equity and excellence for all students should be placed high on the public policy agenda. This will require that schools and states ensure that the support and resources necessary for inclusive education be available: adequate time for professional development and shared planning, use of multiple assessments designed to improve student performance rather than an overreliance on standardized tests, and removal of disincentives to teacher professionalism that flow from overregulation and standardization of school structures and teacher evaluation.
During the 1999–2000 school year, data from three of the original six school districts were gathered and analyzed. The research questions, achievement measure, and procedures from the first year of the study were used in Year 2. A report is pending.
Although it is clear that “good” inclusive programs can and are being developed, much information is available to indicate that “poor” inclusive programs that do not meet the needs of students are being implemented throughout the country (Baines, Baines, & Masterson, 1994; Shanker, 1994–1995, Vaughn & Shumm, 1995). With this in mind, we have used the results and data from the first 2 years of the Indiana Inclusion Study to develop a qualitative study in three inclusive schools that participated in the first 2 years of the study. The purpose of the study is to enable us to examine and describe the teaching practices and school structures that exist within three inclusive elementary schools in which students demonstrated high rates of academic progress in the first 2 years of this study. A report is pending. The information collected in this study will contribute to a growing body of research regarding effective teaching and school practices within inclusive school arrangements.
Authors: Cassandra M. Cole, EdD ( firstname.lastname@example.org), Director, and Massoumeh Majd-Jabbari, PhD, Research Associate, Indiana Institute on Disability and Community, Indiana University, Center on Education and Lifelong Learning, 2853 E. 10th St., Bloomington, IN 47408. Nancy Waldron, PhD, Professor, College of Education, University of Florida, Box 117047, Gainesville, FL 32611