Academic achievement and class placement in high school: do students with learning disabilities achieve more in one class placement than another?

Abstract

The purpose of this study was to examine classroom placement, inclusive versus non-inclusive, relative to the academic performance of students with specific learning disabilities in secondary content area classrooms. Fifty-seven high school students with learning disabilities were assessed using the Grade Level Short Form of the Multilevel Academic Survey Test (MAST). Their reading and math scores were examined relative to each student's grade level, number of general and special education classes attended, and types of placement (i.e., inclusive or non-inclusive setting). The results revealed no statistically significant evidence to indicate that students' academic achievement varied based on inclusive versus non-inclusive placement. The only statistically significant differences observed regarded participants enrolled in a general education literature class compared to those participants placed in a special education setting for literature. Implications for practice, limitations of the study, and considerations for future research are discussed.

**********

Recent trends reveal that students with specific learning disabilities (SLD) are increasingly educated in general education classrooms, while the number receiving instruction in resource rooms or separate classes has decreased substantially (Holloway, 2001). Despite the increase in students with disabilities placed in general education settings there is limited research, particularly at the secondary level, to suggest where students with SLD achieve more academically. Of the many issues related to the integration or inclusion of students with disabilities into general education classrooms, there is none more important than the effects of placement on students' learning and social relations with classmates (Baker, Wang, & Walberg, 1995).

The research literature reviewed for this study was mixed regarding whether academic outcomes for students with disabilities are improved in inclusive classroom settings. Previous research suggested that students with SLD in elementary schools achieve more academically in inclusive programs versus non-inclusive programs (Banerji & Dailey, 1995; Waldron & McLeskey, 1998). Other studies suggested that middle school students achieve more academically in inclusive programs (Rea, McLaughlin, & Walther-Thomas, 2002). For example, Rea et al. (2002) reported that students with disabilities had higher grades, fewer behavioral infractions, and attended school more often than students educated in non-inclusive programs. In contrast, a number of other studies concluded that students with disabilities achieved more in inclusive classroom settings (Baker, et al., 1995; Madden & Slavin, 1983; Carlberg & Kavale, 1980).

The only certainty regarding the effects of class placement is that there is no consensus. Researchers have drawn vastly different conclusions (Carlberg & Kavale, 1980) such as: (a) Students with severe learning disabilities and emotional and behavioral disorders may achieve more in special classrooms (Carlberg & Kavale, 1980); (b) Students with disabilities may achieve more academically in programs that combine the inclusive model with the resource model as opposed to either model on its own (Holloway, 2001); and (c) Inclusive versus non-inclusive placements produce no differences in the academic achievement of students with disabilities (Affleck, Madge, Adams, & Lavenbraun, 1988; Manset & Semmel, 1997; Waldron & McLeskey, 1998). Another perspective voiced is that the frequently posed question of class placement is the wrong question to consider (Leinhardt & Palloy, 1982; Zigmond, 2003; Hocutt, 1996). Researchers taking this position point out that the evidence is inconclusive on whether students with disabilities achieve more in one instructional setting as opposed to another, and contend that questions regarding class placement for students with disabilities are at odds with a focus on individualized educational planning.

Inclusive Classroom Settings

There is a body of research suggesting that students with disabilities achieve more academically and behaviorally in inclusive classroom settings (Baker et al., 1995; Banerji & Dailey, 1995; Carlberg & Kavale, 1980; Frederickson, Dunsmuir, Lang, Monsen, 2004; Madden & Slavin, 1983; Magiati, Dockrell, & Logotheti, 2002; McDonnell, et. al., 2003; Rea, et al., 2002; Waldron & McLeskey, 1998). For example, Baker et al. (1995) reviewed three meta-analyses in the educational literature that addressed the issue of the most effective instructional setting for the education of students with disabilities. Those meta-analyses produced a small-to-moderate beneficial effect of inclusive education on the academic and social outcomes of students with disabilities. The average effect sizes ranged from 0.08 to 0.44, suggesting that students with disabilities educated in regular classes do perform better academically and socially than comparable students in non-inclusive classroom settings.

Banerji and Dailey (1995) studied the effectiveness of an inclusive educational program for elementary students with SLD. The purpose of their study was to measure academic and affective gains made by students with SLD against gains made by normally achieving students. Their study was conducted in an elementary school in a school district in the southeast. Participants were elementary students with SLD in grades 2 to 5 who were served with normally achieving students in inclusive, mixed-grade pods called "houses." Each pod was taught by a team of four teachers, along with a specialized teacher in SLD. The 5th grade students with SLD were placed in an inclusive classroom with general education peers. The SLD teacher co-taught with the general education teacher, assisting both students with SLD and normally achieving students. Instructional interventions and planning were determined collaboratively by the teams of teachers in both instructional settings. The inclusion program was considered the independent variable and the dependent measures were the academic and affective gains of students with SLD. Results suggested that students with SLD made some academic and affective gains at a pace comparable to that of normally achieving peers. In addition, parent and teacher surveys indicated improved self-esteem and motivation among students with SLD. The authors also suggested that the successful implementation of the inclusive program reduced the stigma for students with SLD because they felt more included and less isolated from their peers.

Rea et al. (2002) investigated outcomes for students with SLD in inclusive versus pullout special education programs. Their sample consisted of all 8th grade students with SLD in two middle schools in a suburban school district in the southeast. The independent variables were the educational setting, pullout or inclusive. Dependent measures included academic achievement and affective outcomes such as behavioral infractions and school attendance. The results of Rea et al.'s study indicated that the two settings differed significantly. Students served in inclusive programs achieved higher or comparable scores on standardized tests, earned higher grades, attended more days of school, and committed no more behavioral infractions than those students served in the pullout program.

Waldron and McLeskey (1998) studied the effects of an inclusive school program on the math and reading achievement of students with mild and severe SLD. The experimental group consisted of all students with SLD from three elementary schools that had recently implemented inclusive school programs. These students were compared to all of the students with SLD from three comparison elementary schools in the same school system. The students in the comparison group had volunteered to begin an inclusive school programs, but had yet to do so. Students' academic progress in math and reading were compared using a curriculum-based measure, Basic Academic Skills Samples (BASS). The independent variables were the two school programs, one an inclusive education program and the other a pullout resource model. Results indicated that students with SLD in the inclusive program made significantly more progress in reading and comparable progress in math compared with those students with SLD who received instruction in resource classes. In addition, a greater number of students with mild SLD in inclusive settings had reading performance comparable to their general education peers than those students with mild SLD educated in resource settings.

McDonnell et al. (2003) examined the achievement of students with developmental disabilities compared with that of their peers without disabilities in inclusive classroom settings. Their purpose was to evaluate the impact of inclusive educational programs on the achievement of students with and without developmental disabilities. Changes in the adaptive behavior of 14 students with developmental disabilities during one school year were measured in a quasi-experimental, pretest-posttest design. They reported gains in the adaptive behaviors of all 14 students with developmental disabilities. They also compared the achievement of 324 students without disabilities enrolled in inclusive programs with students with developmental disabilities with 221 students without disabilities who were not exposed to inclusive programs. Results indicated that the academic performance of students without disabilities enrolled in inclusive programs was no different than those who were not involved in an inclusive program.

Overall, of the studies discussed above, the results suggest that students with disabilities, particularly those students with SLD, are likely to have better achievement outcomes in more inclusive classroom settings. Nonetheless, the research literature's support of this assertion is far from unanimous.

Non-Inclusive Classroom Settings

Despite the body of research suggesting that inclusive settings may produce favorable academic gains for students with disabilities, a number of studies suggest that students with SLD fare no better in general education classes than they would in resource settings, and that certain aspects of inclusive programs are less advantageous than pull-out models (Carlberg & Kavale, 1980; Holloway, 2001; Praisner, 2003; Ross & Stevens, 2003; Van Hover & Yeager, 2003; Weiss & Lloyd, 2003). For example, Carlberg and Kavale (1980) concluded that students with learning disabilities, emotional disturbances, and behavioral disorders were better served in special education classrooms. They examined fifty primary research studies of special versus regular class placement to determine the favored educational arrangement for exceptional students. Each study provided a measure of effect size (ES), defined as the post-treatment difference between special and regular placement means expressed in standard deviation units. ES was used as a dependent variable in order to assess the effects of independent variables such as placement, type of outcome measures, internal validity, and other educational, personological, and methodological variables. Results suggested that students with learning disabilities, emotional disturbances, and behavioral disorders achieved more academically in special education classrooms.

In another review, Holloway (2001) also concluded that students with learning disabilities are not necessarily better off in more inclusive programs. Holloway reviewed several studies from the late nineties examining academic achievement and class placement. The focus was on studies with class placement as the independent variables that included academic achievement as a dependent measure. Holloway concluded that inclusive programs are not always the most appropriate instructional setting for students with learning disabilities. Although two decades apart, the studies by Carlberg and Kavale's (1980) and Holloway (2001) questioned the assertion that inclusive classroom settings produce superior academic achievement. Carlberg and Kavale argued that students with mild disabilities were better served in special education classrooms. Likewise, Holloway found that, in many instances, students with disabilities achieved more in special education classrooms.

Inclusive versus Non-inclusive Classroom Settings

Other studies suggest that inclusive versus non-inclusive programs produced no difference in the academic achievement of students with learning disabilities (Affleck, et al., 1988; Manset & Semmel, 1997; Waldron & McLeskey, 1998). For instance, Affleck et al. (1988) found no significant differences in performance among elementary school students with learning disabilities in integrated classrooms compared to resource classroom settings. Manset and Semmel (1997) reviewed eight different inclusive models for elementary students with learning disabilities. To determine whether inclusive programs for students with mild disabilities were effective, these authors reviewed studies published between 1984 and 1994 that included measures of academic outcomes, delivered interventions school-wide, and specified as a primary intent the full-time mainstreaming (i.e., placement in general education) of students with mild disabilities. Independent variables included classroom setting (i.e., inclusive versus non-inclusive) and the dependent measure of interest was academic achievement. Methodological concerns prevented conclusions about the superiority of inclusive programming over pullout programs. There were indications however, that inclusive programming was effective for some, although not all, students with mild disabilities.

Waldron and McLeskey (1998), as reviewed previously, did find that students with mild learning disabilities were more likely to perform comparable to their peers without disabilities when placed in inclusive settings. The same study also concluded, however, that students with severe learning disabilities made comparable progress in reading and math, regardless of the setting in which they were served. Results from Affleck et al. (1988); Manset and Semmel (1997); and Waldron and McLeskey (1988) suggest that the placement of students with disabilities (i.e., inclusive versus pull-out) does not produce differences in academic performance.

Where should students with disabilities receive services?

Some scholars assert that where special education services should be received is the wrong question to ask (Leinhardt & Palloy, 1982; Hocutt, 1996; Zigmond, 2003). A decade ago Hocutt (1996) asked whether placement was the critical factor in determining the effectiveness of special education. In a review of research directly relevant to inclusion, including research on the effectiveness of special education in general, Hocutt concluded that various program models, whether implemented both in general education or special education, can have moderately positive academic and social impacts for students with disabilities. However, it was pointed out that no intervention has been designed that eliminates the impact of having a disability and that most students with disabilities have not achieved commensurately with their nondisabled peers. According to Horcutt (1996), the most effective interventions for students with disabilities, whether in general education or special education settings, have employed intensive and reasonably individualized instruction, combined with frequent and careful monitoring of student progress.

In a literature synthesis, Zigmond (2003) concluded that research evidence on the relative efficacy of one special education placement over another is scarce, inconclusive, and methodologically flawed. Zigmond (2003) also argued that querying where students with disabilities should be educated was the wrong question to ask. According to Zigmond (2003), focusing on placement conflicts with the individualized planning that should be embodied in decision-making for students with disabilities. Students have a right to be placed in the least restrictive environment and placement decisions must be made on an individualized basis.

Despite the focus on individualized educational programming for students in special education and the need to make individualized decisions, many school districts across the nation are adopting models in which students with disabilities are being educated in more inclusive settings. A major concern is that some students with disabilities may not achieve at optimal levels in more inclusive settings because they require specialized instruction. On the other hand, many students with disabilities in specialized settings may fall short off their academic potential because they lack access to the general curriculum and, in many cases, a teacher with expertise in the content area they are studying. Therefore, the purpose of this descriptive exploratory study was to examine differences in academic achievement based on classroom placement for students with SLD in secondary classroom settings.

Method

Participants

Fifty-seven high school students with SLD were the subjects of this study. The participants represented all students from two suburban high schools in the southeastern United States who met both the federal and state criteria for SLD. Forty-two of the participants were male and 15 were female. Participants received special education services in inclusive and non-inclusive settings, varying in the number of inclusive classes that each was scheduled. Of the participants, 19 were in 9th grade, 18 in 10th grade, 13 in 11th grade, and 7 in 12th grade. The ethnic backgrounds of the students included 50 Caucasian, 5 African-American, and 2 Hispanic-American students. Eighty percent of the participants were reported to have reading disabilities and 20% had math disabilities. There were no instances of comorbidity (i.e., diagnosis of both reading and math disabilities). Approximately 10% of the student participants had an additional diagnosis of attention deficit hyperactivity disorder (ADHD).

Settings

Each of the students' classes was characterized either inclusive or non-inclusive. Inclusive classes were those that were offered to general education students and taught by a general education teacher. Most inclusive classes contained approximately 25 students with no more than 20% identified as having a disability. Some (e.g., mathematics and literature classes), but not all, of the inclusive classes also had a special education teacher. Conversely, classes deemed as non-inclusive were those that were taught by a special education teacher that occurred in settings other than a general education classroom. Non-inclusive classrooms contained no general education students and were only taught by special education teachers.

Dependent Measures

The students were asked to complete the Grade Level Test Short Form of the Multilevel Academic Survey Test (Howell, Zucker, & Morehead, 1985). The Multilevel Academic Survey Test (MAST) consists of two 20-item multiple choice maze tasks and 24 math computation items. The MAST is intended for use by school personnel to make decisions about student performance in mathematics and reading. These professionals include school psychologists, special education teachers, educational diagnosticians, and teachers in special programs. Teachers in general classrooms may use the MAST with normally achieving students, but the test is primarily intended for those educators who instruct and assess students exhibiting academic difficulties. The criterion-related validity of the MAST was established with 300 students with the Iowa Test of Basic Skills (ITBS). The coefficients in mathematics range from a low of .54 to a high of .85. The coefficients in reading range from a low of .59 to a high of .81. Inspections of the patterns of coefficients for the various measures indicate a high degree of agreement between the MAST and the ITBS in specific content subdomains. The predictive validity of the MAST was studied with 39 students in grades 4th to 12th of the standardization sample for the MAST Grade Level Test that had previously been labeled as learning disabled (LD), were matched with peers who had not been labeled as learning disabled (non-LD). The results showed that 74% of the students classified by the school systems as LD were also classified as LD by the Short Form, while 72% of the students classified as non LD by the school systems were also classified as non LD by the Short Form. The reliability of the MAST Grade Level Test scores was studied in a test-retest study involving 366 students in grades 3 through 8. The same teachers administered the test on both occasions within a two-week period. Raw test scores from the two administrations were correlated and the data suggest that the MAST has good reliability over a period of time. The correlation coefficients reported in the technical manual were moderately high to high in nearly all cases (only 2 of the coefficients were below .60 and only 6 of 24 coefficients were below .70). The median coefficients across grades were .68 in Short Form Reading and .71 on the Short Form Mathematics.

Procedures

Carefully trained graduate students administered all measures to student participants. The graduate students established a testing schedule with the participating school and administered the MAST to groups of students in accordance with the test's guidelines. The protocol provided by the MAST administration manual was followed for both the reading and math subtests. On the reading portion of the MAST, the students are given five minutes to complete the first MAZE task. Students are then instructed to stop and turn to the next MAZE task. Students are then given five minutes to complete the second MAZE task. Students are then instructed to complete the arithmetic problems on their answer sheet. After twelve minutes, they are asked to stop and the administrator collects the test.

Results

The purpose of this study was to examine classroom placement, inclusive versus non-inclusive, relative to the academic performance of students with SLD in secondary content area classrooms. The Grade Level Short Form of the Multilevel Academic Survey Test (MAST) was administered to 57 high school students. Descriptive statistics and significance levels are reported for gender and grade level (Table 1), number of general education classes attended (Table 2), and type of placement (Table 3). There were no statistically significant differences in the performance for gender on the MAST Reading and Math subtests. In addition, student performance on the MAST subtests did not differ by grade level. As displayed in Table 4, no significant differences were observed for reading across grade levels with a small effect size. In addition, no significant differences were observed for math across grade levels, also with a small effect size. No significant differences were observed for the MAST reading or math across class schedules, with a small effect size observed. For type of class taken, significant differences were observed between those students who were, versus were not taking the general education literature class for reading (Table 5). However, the effect size for that difference was small. This was the only significant difference found for type of class taken for reading as well as math (Table 6).

Discussion

Inclusion of students with disabilities in general education classes continues to be a contentious area of special education. In this study, we examined the placement patterns of 57 high school students with SLD. Overall, with the exception of one comparison, we found no statistically significant differences in the academic performance of students with SLD for reading or math. Our findings are consistent with previous research reporting that class placement for students with disabilities did not correlate with academic achievement (Affleck et al., 1988; Manset & Semmel, 1997; Waldron & McLeskey, 1998). In this study, it was anticipated that at the very least, students who were included in more general education classes would have had higher reading and math scores than those who were in special education. In this study, the only statistically significant difference we found were those students who were placed in a general education literature class and their mean reading scores on the MAST, however, even the effect size was small for this one significant result. In addition, while this finding somewhat aligns with previous research asserting that students placed in more inclusive settings achieve more academically (Baker, et al., 1995; Banerji & Dailey, 1995; Carlberg & Kavale, 1980; Madden & Slavin, 1983; Rea, et al., 2002; Waldron & McLeskey, 1998), given the number of analysis conducted, any significant result is suspect due to the presence of family-wise error.

Limitations and Implications for Future Research

Regarding the optimal academic achievement of students with SLD, the results of this study do not provide any statistical support for one class placement over another. However, even before discussing this study's apparent lack of relation between class placement and performance, several of its limitations must be addressed. First, we should acknowledge that the analyses conducted as part of this study were exploratory in nature. Moreover, there were only 57 student participants. This small sample size impedes our ability to make generalizations to a larger population.

Another limitation of this study is the possible heterogeneity of the groups divided by class schedule. One could argue that more capable students tend to be placed in a greater number of general education classes and while lower performing students tend to be placed in a greater number of special education classes. This particular limitation would be difficult to overcome even in the most rigorous efforts of inquiry. It stands to reason that the intent behind Individualized Education Plans would prohibit random placement of students into class schedule groups.

The lack of information regarding classroom teachers' backgrounds, professional experiences and instructional practices is another limitation of this research. Although, when questioned, the school assumed there were no differences between teachers in inclusive versus non-inclusive classes, these data were not collected for examination. In addition to careful analysis of such critical variables, detailed information regarding the type of instruction delivered across class-room settings is needed.

Results from this study revealed no significant differences between academic achievement and class placement for students with SLD in two Southeastern high schools. Future research should care-fully examine variables much more distinct than the dichotomous division of inclusive versus non-inclusive class setting. More care must be taken to query the extent to which students with SLD are achieving optimally regardless of setting.

Implications for Practice

It remains important for educators, parents, and administrators to consider the potential benefits and shortcomings of class placement options. Results from this study have three major implications for educators, administrators, local education agency representatives, and other individuals involved in the educational planning of students with learning disabilities. First, educators must carefully consider student outcomes when placement decisions are made. In many cases, the percentage of time a student will spend in general education classes is the primary factor in determining class schedules. Educators also must determine whether students can meet individualized goals and objects in general education classes. Even though this should be standard practice as mandated by IDEA and its reauthorizations, it is easy for IEP committees to lose sight and allow the tail to wag the dog by first determining an optimal percentage of time for general education placement without careful consideration of whether each situation is likely to allow the student to accomplish all the goals and objectives of the IEP. Likewise, if students are being placed in more restrictive settings, the committee should judiciously consider how each student will receive access to the general curriculum. Perhaps the most important implication is that educators must pay closer attention to how placement decisions are made. Individualized planning should include consideration of each learner's unique needs when determining class placement and should not automatically assume that any student will necessarily learn more or less in a general education class.

Conclusions

The results from this study lend support to the proposition that where students with disabilities receive special education services may not be the most relevant question within the context of student achievement. Students with SLD may achieve comparably in general and special education classes. However, empirical studies controlling for the seemingly infinite combinations of extraneous variables are needed to support that assertion as well.

Researchers must continue to contemplate the numerous other variables that likely weigh more heavily on the performance of students with SLD than placement. As others have suggested, student achievement may not be heavily influenced by location alone. Given the limited information comparing the content and instructional techniques between inclusive and general education settings, the present study does not offer any alternative hypotheses. It does, however, underscore the need for future research that examines the content and instruction across settings and their potential effect on student achievement.

References

Affleck, J., Madge, S., Adams, A. & Lavenbraun, S. (1988). Integrated classroom versus Resource model: Academic viability and effectiveness. Exceptional Children, 54, 331-348.

Baker, E.T. Wang, M.C., Walberg, H. J. (1995). The effects of inclusion on learning. Educational Leadership, 52, 33-35.

Banerji, M. & Dailey, R.A. (1995). A study of the effects of an inclusion model on students with specific learning disabilities. Journal of Learning Disabilities, 28, 511-523.

Carlberg, C. & Kavale, K. (1980). The efficacy of special versus regular class placement for exceptional children: A meta-analysis. Journal of Special Education, 14, 295-309.

Frederickson, N., Dunsmuir, S., Lang, J., & Monsen, J.L. (2004). Mainstream-special school inclusion partnerships: pupil, parent and teacher perspectives. International Journal of Inclusive Education, 8, 37-58.

Hocutt, A.M. (1996). Effectiveness of special education: Is placement the critical factor? The Future of Children, 6, 77-102.

Holloway, J.H. (2001). Inclusion and students with learning disabilities. Educational Leadership, 58, 86-87.

Howell, K. W., Zucker, S. H., & Morehead, M. K. (1985). Multilevel Academic Survey Test: San Antonio, TX: The Psychological Corporation.

Leinhardt, G., & Palloy, A. (1982). Restrictive settings: Exile or haven? Review of Educational Research, 52, 557-578.

Madden, N.A. & Slavin, R.E. (1983). Mainstreaming students with mild handicaps: Academic and social outcomes. Review of Educational Research, 53, 519-569.

Magiati, I., Dockrell, J.E., & Logotheti, A.E. (2002). Young children's understanding of disabilities: The influence of development, context, and cognition. Journal of Applied Developmental Psychology, 23, 409-431.

Manset, G. & Semmel, M.I. (1997). Are inclusive programs for students with mild disabilities effective? Journal of Special Education, 31, 155-181.

McDonnell, J., Thorson, N., Disher, S., Mathot-Buckner, C., Mendel, J., & Ray, L (2003). The achievement of students with developmental disabilities and their peers without disabilities in inclusive settings: An exploratory study. Education and Treatment of Children, 26, 224-236.

Praisner, C.L. (2003). Attitudes of elementary school principals toward the inclusion of students with disabilities. Exceptional Children, 69, 135-145.

Rea, P.J., McLaughlin, V.L. & Walther-Thomas, C. (2002). Outcomes for students with learning disabilities in inclusive and pullout programs. Exceptional Children, 68, 203-223.

Ross, A.H. & Stevens, K.B. (2003). Teaching spelling of social studies content vocabulary in inclusive learning environments. Journal of Behavioral Education, 12, 287-309.

Van Hover, S.D., & Yeager, E.A. (2003). Secondary history teachers and inclusion of students with disabilities: An exploratory study. Journal of Social Studies Research, 27, 36-45.

Waldron, N.L. & McLeskey, J. (1998). The effects of an inclusive school program on students with mild and severe learning disabilities. The Council for Exceptional Children, 64, 395-405.

Weiss, M.P., & Lloyd, J. (2003). Conditions for co-teaching: Lessons from a case study. Teacher Education and Special Education, 26, 27-41.

Zigmond, N. (2003). Where should students with disabilities receive special education services? Is one placement better than another? Journal of Special Education, 37, 93-199.

Cecil Fore III

The University of Georgia 

Shanna Hagan-Burke

Mack D. Burke 

Texas A & M University

Richard T. Boon 

Steve Smith

The University of Georgia

Correspondence to Cecil Fore, III, The University of Georgia, 548 Aderhold Hall, Department of Special Education, Athens, GA 30602-7153; e-mail: [e-mail].

Table 1 Descriptive Statistics for MAST Reading and Math Subtests by
Gender and Grade Level

                           Descriptive Statistic
Variable     MAST Subtest  M      SD     N

Gender
Male         Reading       31.79  6.898  42
             Math          14.36  3.392  42
Female       Reading       32.87  4.642  15
             Math          14.67  3.599  15

Grade Level
9th          Reading       31.16  6.405  19
             Math          13.47  3.306  19
10th         Reading       31.72  7.450  18
             Math          14.94  3.208  18
11th         Reading       32.69  4.250  13
             Math          14.54  4.196  13
12th         Reading       34.29  7.181   7
             Math          15.57  2.573   7

Table 2 Descriptive Statistics for MAST Reading and Math for Class
Schedule

                               Descriptive Statistics for MAST Subtests
                                   Reading         Mathematics
Class Schedule                 M      SD     N   M      SD     N

Schedule One (1 general and 3  28.67  8.682  15  13.40  3.019  15
  special)
Schedule Two (2 general and 2  33.00  4.541  17  14.65  3.936  17
  special)
Schedule Three (3 general and  32.89  5.496  19  15.16  3.371  19
  1 special)
Schedule Four (4 general and   35.33  3.777   6  14.17  3.061   6
  no special)

Table 3 Descriptive Statistics for MAST Subtests by Class Subject Matter
and Placement Type

                                 Reading        Mathematics
Subject and Placement       M      SD     N   M      SD     N

General Ed. Literature
  In class:                 35.38  3.429  13  14.38  3.355  13
  Not in class:             31.09  6.716  44  14.45  3.474  44
Special Ed. Literature
  In class:                 30.33  7.556  15  13.27  3.882  15
  Not in class:             32.69  5.854  42  14.86  3.182  42
General Ed. Mathematics
  In class:                 33.10  5.438  20  15.05  3.832  20
  Not in class:             31.51  6.813  37  14.11  3.178  37
Special Ed. Mathematics
  In class:                 30.45  7.702  20  14.10  3.370  20
  Not in class:             32.95  5.421  37  14.62  3.475  37
General Ed. Science
  In class                  33.90  4.045   9  14.78  3.866   9
  Not in class:             31.73  6.684  48  14.38  3.368  48
Special Ed. Science
  In class:                 31.18  8.268  11  14.82  3.894  11
  Not in class:             32.28  5.909  46  14.35  3.335  46
General Ed. Social Studies
  In class                  32.86  5.659  22  14.41  2.789  22
  Not in class              31.57  6.797  35  14.46  3.799  35
Special Ed. Social Studies
  In class                  31.44  6.167   9  14.56  2.603   9
  Not in class              32.19  6.453  48  14.42  3.572  48
General Ed. Elective
  In class                  31.96  6.403  53  14.55  3.484  53
  Not in class              33.50  6.455   4  13.00  2.160   4
Special Ed. Elective
  In class                  31.12  6.776  41  14.27  3.421  41
  Not in class              34.50  4.457  16  14.88  3.481  16

Table 4 Results for Grade Level and Class Schedule

Subject           F      df     p     [[eta].sup.2]

Reading
  Grade Level      .459  3, 53  .712  .025
  Class Schedule  2.343  3, 53  .084  .117
Mathematics
  Grade Level      .891  3, 53  .452  .048
  Class Schedule   .766  3, 53  .518  .042

Table 5 MAST Reading Results by Type of Class

                           Descriptive Statistics for MAST Reading
                Education  Results
Type of Class   Placement  F      df     p     [[eta].sup.2]

Literature      General    4.890  1, 55  .031  .082
                Special    1.532  1, 55  .221  .027
Mathematics     General     .805  1, 55  .374  .014
                Special    2.036  1, 55  .159  .036
Science         General     .872  1, 55  .355  .016
                Special     .262  1, 55  .611  .005
Social Studies  General     .553  1, 55  .460  .010
                Special     .102  1, 55  .751  .002
Elective        General     .214  1, 55  .645  .004
                Special    3.384  1, 55  .071  .058

Table 6 MAST Math Results by Type of Class

                           Descriptive Statistics for MAST Math
                Education  Results
Type of Class   Placement  F      df     p     [[eta].sup.2]

Literature      General     .004  1, 55  .949  .000
                Special    2.456  1, 55  .123  .043
Mathematics     General     .986  1, 55  .325  .018
                Special     .299  1, 55  .587  .005
Science         General     .104  1, 55  .749  .002
                Special     .166  1, 55  .686  .003
Social Studies  General     .003  1, 55  .959  .000
                Special     .012  1, 55  .912  .000
Elective        General     .759  1, 55  .387  .014
                Special     .359  1, 55  .552  .006