Cautions in Interpreting Findings

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

• The analyses are descriptive; none of the findings should be interpreted as implying causal relationships. Neither should differences between disability categories be interpreted as reflecting disability differences alone, because of the confounding of disability and other demographic factors.
• Data presented are combined youth self-report and parent-report data. If a Wave 3 youth interview/survey was completed, youth’s responses to these items were used in this report. If a youth interview/survey could not be completed for an eligible youth or if a youth was reported by parents not to be able to participate in an interview/survey, parent responses were used. For the subsample of out-of-high school youth included in this report, the youth interview/survey was the source of data for post-high school outcomes for 84 percent of youth, and the parent interview was the source for 16 percent of youth. Combining data across respondents raises the question of whether parent and youth responses would concur—that is, would the same findings result if parents’ responses were reported instead of youth’s responses. When both parents and youth were asked whether the youth belonged to an organized community group, currently worked for pay, and worked for pay in the past 2 years, and the wages currently employed youth earned per hour, their responses agreed from 69 percent to 80 percent of the time.
• It is important to note that descriptive findings are reported for the full sample of out-ofhigh school youth; those findings are heavily influenced by information provided for youth with learning disabilities, who constitute 64 percent of the weighted sample. Comparisons also were conducted between groups of youth who differed with respect to disability category, high school-leaving status and timing, 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. These complex interactions and relationships among subgroups relative to the variables included in this report have not been explored.
• Several types of analyses were conducted for this report, including between-group means, between-group percentages, and within-subject percentages. Because of the weighted nature of NLTS2 data, equality between the mean values of the responses to a single survey item in two disjoint subpopulations was based on a test statistic essentially equivalent to a two-sample t test for independent samples using weighted data. Sample sizes for each group being compared were never less than 30. For a two-tailed test, the test statistic was the square of the t statistic, which then followed an approximate chisquare distribution with one degree of freedom, that is, an F (1, infinity) distribution.
• Although discussions in the report emphasize only differences that reach a level of statistical significance of at least p < .01, the large number of comparisons made in this report will result in some significant differences that are “false positives,” or differences mistakenly determined to be significant when they are not (i.e., type I errors). Readers also are cautioned that the meaningfulness of differences reported here cannot be derived from their statistical significance.

Top