Archived Information

Summary and Conclusions

Data from the National Education Longitudinal Survey of 1988 8th Graders (NELS:88) transcript study and student cognitive testing administration were used to assess the effect of vocational programs and vocational courses on achievement in mathematics, science, and reading, and on student dropout rates. Analyses were conducted which took into account differences in student characteristics, school type, and student achievement.

Results of the effects of curricular program on achievement indicate that students in vocational programs do not score as high in mathematics, science, and reading as students in academic programs, supporting findings by Alexander, Cook, & McDill (1978) and Alexander & Cook (1982, but see Jencks and Brown, 1975 and Meyer; 1992, who show no difference and a positive influence of certain vocational courses, respectively). Vocational program students' scores in mathematics and science do not differ significantly from students in general high school programs. Vocational program students' reading scores are significantly lower than those of students in general programs. These results hold even when a number of student background and educational characteristics are included as control variables in the analysis.

Results of the effects of vocational coursetaking on achievement in mathematics, science, and reading indicate no impact of vocational courses on gain in performance in these areas. With a few scattered and inconsistent exceptions, vocational courses taken in the 9th and 10th grade had no effect on achievement gain from 10th to 12th grade.

The analysis of dropout rates indicates that vocational education may play a role in keeping students in school. However, that role may be indirect, operating through the mechanism of improving students' grades and standing with respect to others in their school. Though it was not explicitly tested in the present analysis, it is possible that vocational courses serve to raise poorly-performing academic students' sense of accomplishment and improve their image of their capability. These factors may be enough to keep some students from dropping out of high school. Direct effects on reduction of the dropout rate were found for vocational courses in agriculture and technical/communications. Based on the present analysis, it is not unreasonable to assume that there were aspects of the curriculum or course content in these areas that stimulated students' interests in high school.

The analyses presented here suggest that vocational courses, as they are grouped and studied here, have small and inconsistent effects on gain in mathematics, science, and reading achievement. When using these results to gauge the worth of vocational education, it is important to remember that the results do not speak to other utilities of vocational courses. For example, the results of this study do not examine whether the vocational courses effectively prepare students for the work force. If vocational courses do prepare students adequately for the labor market, that may make up somewhat for the lack of an effect on academic achievement.

The results suggest that vocational programs may have some effect, either indirect or direct, on reducing the dropout rate. This is an intriguing finding, and bears further exploration, especially since the current literature presents confusing and contradictory results. In the process of conducting the research for this report, a number of methodological issues have surfaced which have been useful in thinking about both previous research and the results of the analyses presented here. They are mentioned here in the hope that they will be useful in structuring future research on the relationship between vocational education and the dropout rate.

The first issue is whether vocational program participation or vocational courses are used to predict dropout rates. If vocational program participation is used, one must be very careful that students can be accurately placed in vocational programs very early in their high school experience in order to avoid the confound between length of high school tenure and program participation. If it is impossible to make accurate program participation categorizations until later in high school, then it is better to use the number of vocational courses taken in a previous time period to predict dropout status in a subsequent time period.

The second issue concerns which control variables are used in dropout prediction models. To yield an unbiased assessment of the effect of vocational education on dropout rates, models should contain other factors knows to affect the dropout rate. For example, because dropout rates are low among students in academic programs, it is important to include academic coursetaking in any model assessing the effect of vocational courses on dropping out. Similarly, because poor performance and skipping school are related to dropping out, these factors should be included. Other factors representing differences in school or social experiences that are associated with dropping out should be included in the model predicting dropout rates from vocational coursetaking. Part of the confusion in the current literature may be due to a lack of consistency in measuring and including control variables in the analyses.

The third issue concerns whether vocational education has a direct or indirect effect on the dropout rate (Pittman, 1991). This issue is related to the second issue because it concerns whether other, intermediate, outcomes are included. As the analysis in this paper showed, excluding class rank resulted in a positive effect of vocational education on the dropout rate. Including rank eliminated that effect. This suggests both that the model excluding rank may have resulted in a biased parameter estimate for the vocational education effect, and that the effect of vocational education on the dropout rate is mediated through its effect on class rank. This finding

raises the issue of how vocational education may have a non-curricular influence on the high school student's school experience to keep the student engaged in school. This issue could be explored by examining the effect of vocational education on other aspects of the student's school experience available in the NELS:88 data, for example, sense of self-esteem, interest in school, attendance, behavioral problems, aspirations, and other school-related motivational factors. In turn, the effect of these non-curricular factors on the dropout rate could be modeled. The result would be a sophisticated picture of the total vocational education experience, and how this experience can account for keeping students in school.

It should be noted that this approach could be used advantageously to understand direct curricular effects of vocational education as well. The analysis presented in this paper suggests curricular effects on the dropout rate for courses in the agriculture and technical/communications areas, however, the level of the present analysis does not specify what those curricular effects are (see Meyer, 1992 for an example of studying curricular effects of vocational education on achievement). If there is a way to use the NELS:88 data to ascertain the nature of the curriculum in these vocational areas, it would be well worth the time it takes to do so.

In addition, it should be noted that 8th-grade achievement scores were available and could have been used in the analysis. The decision not to include them was based on two considerations: (1) no instrumental variables were available, and (2) a weight allowing the use of the 8th- to 12th-grade panel sample along with transcripts was not available. NELS:88 also contains information on a number of other potential predictors of success in high school, such as student aspirations, attitudes and motivations, and parental support. Researchers who wish to explore this topic further might consider these and other variables available from this rich data source.

The analyses reported here failed to find an effect of vocational performance on achievement. However, they do suggest that vocational education has an effect on the dropout rate, although the results do not clearly show the mechanisms underlying that effect. Even if taking vocational courses do keep students in school, a hidden cost of vocational education is that it limits students' opportunities to take traditional academic courses and may limit students' access to traditional academic knowledge. This is the case because the total number of courses students can take in high school is more or less fixed. For the most part, taking a vocational education course means trading off some other, perhaps traditional academic, course. It appears that this tradeoff, and its effect on academic learning, is being recognized to some extent by students themselves or by school administrators.

Coursetaking patterns including a concentration both in vocational and academic courses were found. Three different vocational program types were identified: (1) a standard vocational track, (2) a vocational and academic track, and (3) a vocational and rigid academic track. The standard vocational track pattern was found among a large number of students with vocational concentrations (43.41 percent). Of all the students with vocational concentrations, these students are least likely to take traditional mathematics and science courses. Fortunately, this is less true for students with the vocational and academic pattern, and fortunately this pattern also makes up an even larger portion of students with vocational concentrations (50.14 percent). It is encouraging to see the emergence of a third combined vocational/academic pattern, students whose curriculum accommodates vocational courses and the more challenging of the traditional academic courses, although this pattern is seen for a very small portion of students with vocational course concentration (6.45 percent).

[Results ]

[References ]