Technical Notes

Readers should remember the following issues when interpreting the findings in this report:

• Purpose of the report. The purpose of this report is descriptive; as a nonexperimental study, NLTS2 does not provide data that can be used to address causal questions. The descriptions provided in this document concern the self-reported perceptions of youth. No attempt is made to "validate" these self-perceptions with information on youth's understanding of the survey items or with direct assessments of students' abilities or behaviors. Further, the report does not attempt to explain why youth responded as they did or why youth in different subgroups (e.g., disability categories) differ in their responses.


• Subgroups reported. In each chapter, the descriptive findings regarding youth's self-representations are reported for the full sample of youth; those findings are heavily influenced by information provided by youth with learning disabilities, who constitute 63 percent of the weighted sample (see appendix B). Youth with mental retardation, emotional disturbances, or other health impairments, and speech/language impairments constitute 12 percent, 12 percent, 5 percent, and 4 percent of the weighted sample, respectively. The other seven categories together make up less than 6 percent of the weighted sample. Findings then are reported separately for youth in each federal special education disability category. Comparisons also were conducted between groups of youth who differed with respect to age, gender, race/ethnicity, and household income. These bivariate analyses should not be interpreted as implying that a factor on which subgroups are differentiated (e.g., disability category) has a causal relationship with the differences reported. Further, readers should be aware that demographic factors (e.g., race/ethnicity and household income) are correlated among youth with disabilities, as well as being distributed differently across disability categories (e.g., youth in the category of mental retardation are disproportionately likely to be African American, and those in the other health impairment category are disproportionately likely to be White, relative to the general population; see appendix B table B-5, for percentage of youth within each disability category, by demographic characteristics).²⁰


• Findings are weighted. NLTS2 was designed to provide a national picture of the characteristics, experiences, and achievements of youth with disabilities in the NLTS2 age range as they transition to young adulthood. Therefore, all the statistics presented in this report are weighted estimates of the national population of students receiving special education in the NLTS2 age group who could describe their own perspectives, and of each disability category individually.


• Standard errors. For each mean and percentage in this report, a standard error is presented that indicates the precision of the estimate. For example, a variable with a weighted estimated value of 50 percent and a standard error of 2.00 means that the value for the total population, if it had been measured, would, with 95 percent confidence, lie between 46 percent and 54 percent (i.e., within plus or minus 1.96 x 2 percentage points of 50 percent). Thus, smaller standard errors allow for greater confidence to be placed in the estimate, whereas larger ones require caution.


• Small samples. Although NLTS2 data are weighted to represent the population, the size of standard errors is influenced heavily by the actual number of youth in a given group (e.g., a disability category). In fact, findings are not reported separately for groups that do not include at least 30 sample members because groups with very small samples have comparatively large standard errors. For example, because there are relatively few youth with deaf-blindness, estimates for that group have relatively large standard errors. Therefore, readers should be cautious in interpreting results for this group and others with small sample sizes and large standard errors.


• Significant differences. A large number of statistical analyses were conducted and are presented in this report. Since no explicit adjustments were made for multiple comparisons, the likelihood of finding at least one statistically significant difference when no difference exists in the population is substantially larger than the Type 1 error for each individual analysis. This may be particularly true when many of the variables on which the groups are being compared are measures of the same or similar constructs, as is the case in this report. To partially compensate for the number of analyses that were conducted, we used a relatively conservative p value of .01. The text mentions only differences that reach a level of statistical significance of at least p < .01. If no level of statistical significance is reported, the group differences described do not attain the p < .01 level of statistical significance. Readers also are cautioned that the meaningfulness of differences reported here cannot be inferred from their statistical significance.

Top