Skip Navigation

National Center for Special Education Research

Facts From NLTS2: Secondary School Experiences of Students With Autism
NCSER 2007-3005
April 2007

Instructional Settings

Most students with autism take classes in both general and special education settings, although they are more likely to take courses in a special than a general education setting (figure 2). Sixty-two percent take at least one general education course in a given semester, whereas 86 percent take at least one course in a special education setting, (p < .001). 10 On average, general education courses make up one-third of the kinds of courses students with autism take, and special education courses comprise 62 percent.

Figure 2. Course taking in a semester by students with autism, by instructional setting

Secondary school students with autism are more likely to take nonacademic courses other than vocational education (e.g., physical education, study skills) in a general education setting (52 percent) than to take academic (36 percent, p < .001) or vocational courses (31 percent, p < .001) in this type of setting.


10 Testing for the significance of differences in responses to two survey items for the same individuals involves identifying for each youth the response to the two items. Responses to each item (e.g., taking at least one course in a special education setting compared with taking at least one course in a general education setting) are scored as 0 or 1. The difference between these scores produces values for individual students of +1 (responded affirmatively to the first item but not the second), 0 (responded affirmatively to both or neither item), or -1 (responded affirmatively to the second item but not the first). The test statistic for the null hypothesis of a mean of zero for the difference score is the square of a ratio, where the numerator of the ratio is the weighted mean change score and the denominator is an estimate of the standard error of that mean. Since the ratio approaches a normal distribution by the Central Limit Theorem, and sample sizes are at least 30, this test statistic approximately follows a chi-square distribution with one degree of freedom (i.e., an F(1, infinity) distribution).