The Postsecondary Achievement of Participants in Dual Enrollment: An Analysis of Student Outcomes in Two States

What is Dual Enrollment, and Why is it Appropriate for CTE Students?

Dual enrollment programs are collaborative efforts between high schools and colleges in which high school students (usually juniors and seniors) are permitted to enroll in college courses. These programs provide students with a challenging academic experience and the opportunity to earn college credit prior to high school graduation. Unlike in other programs such as Advanced Placement and International Baccalaureate, dual enrollment students take actual college courses with a college syllabus, often on a college campus, rather than a college-level course intended to be taken by high school students.

Today a variety of policymakers, authors, educators, and foundations argue that many students, not only those with outstanding educational credentials, may benefit from participation in a dual enrollment program. In their view, dual enrollment is presumed to lead to a long list of positive outcomes for all participating youth, including increasing the academic rigor of the high school curriculum; helping low-achieving students meet high academic standards; providing more academic opportunities and electives in cash-strapped, small, or rural schools; reducing high school dropout rates and increasing student aspirations; helping students acclimate to college life; and reducing the cost of college for students.

This enthusiasm for expanding access to dual enrollment comes in part because dual enrollment mirrors the goals of two prominent trends in education, both driven by low rates of student success in college: increasing the rigor of secondary education and strengthening links between the secondary and postsecondary sectors. Dual enrollment is seen as addressing the need for increased high school rigor because it enables students to take college courses prior to high school graduation. And, as high schools and colleges partner to offer dual enrollment programs, proponents hope that communication and collaboration between the two sectors will increase in general.

Dual enrollment has also become a component of many CTE programs because its goals mirror those of a variety of CTE reforms, including upgrading the CTE curriculum and building upon the promise and addressing the shortcomings of previous CTE reform efforts.

How Many Students Participate in Dual Enrollment?

Data on student participation in dual enrollment are only beginning to be collected. According to two 2005 reports from the U.S. Department of Education, 71 percent of U.S. high schools and 51 percent of U.S. postsecondary institutions permitted high school students to take college courses in 2002-03. In total, 813,000 secondary school students took a college-credit course during the 12-month 2002-03 school year.

Though longitudinal data are unavailable, program-level data indicate that participation in dual enrollment has increased in recent years. This growth is likely to continue. According to a 2006 scan of state policies, 42 states have policies pertaining to dual enrollment. Some states with standing legislation are considering revisions that would make dual enrollment accessible to more students. At the federal level, the Secretary of Education’s Commission on the Future of Higher Education has expressed support for the expansion of dual enrollment programs.

The Effectiveness of Dual Enrollment in Encouraging Postsecondary Persistence

Despite the popularity of dual enrollment, little is known about its effectiveness as a strategy for increasing students’ postsecondary attainment. Two extensive reviews of the literature found almost no evidence that dual enrollment contributed to students’ college access or academic success. Recent research has not increased our knowledge regarding dual enrollment’s effectiveness either, as much of this work has been qualitative, exploring state policies or program features. Although these studies have identified promising program practices and indicate that dual enrollment may help students transition into college, they do not measure program effectiveness.

Studies that have attempted to look at dual enrollment outcomes tend to suffer from two shortcomings. First, they generally lack comprehensive data to include in their outcomes analyses, as few programs or states have comprehensive K-16 data systems. Second, they often do not use rigorous statistical methods to control for preexisting student characteristics, even when such data are available. Therefore, it remains unclear whether dual enrollment participation increases students’ likelihood of entering college, preparedness for college-level work, or attainment of a college degree. Moreover, no research has focused on its impact for CTE students.

Organization of the Report

This report uses statistical methods to examine the outcomes of dual enrollment participation for students in two large, well-established programs, one in Florida and one in New York City. We use datasets that span both high school and college student outcomes, enabling us to control for student characteristics. We pay particular attention to the impact of dual enrollment participation on students enrolled in CTE courses of study in high school.

The report is organized as follows. After describing the methodology used, we turn to the findings for each program separately. The third section focuses on students in the State of Florida. The outcomes analyses in this section address dual enrollment participation generally as well as dual enrollment participation for CTE students specifically. The fourth section examines outcomes for CTE students who participated in New York City’s College Now program. The fifth section of the report examines whether certain groups of students, such as males or those from low socioeconomic backgrounds, have different outcomes of participation in dual enrollment programs. In the final section, we offer conclusions and implications for programs and policy.

Methods

The research questions for this study include the following:

1. What are the short-term effects of participation in a dual enrollment program, for all students and for CTE students, as measured by high school graduation and college enrollment rates?
2. What are the effects of participation in a dual enrollment program on all students’ and CTE students’ initial entry into postsecondary education, such as enrollment intensity, first-semester grade point average, and persistence to the second semester?
3. What are the long-term effects of participation in dual enrollment for all students and for CTE students, as measured by their persistence into the second year of postsecondary education, grade point average, and credit accumulation?
4. Do program effects vary by race/ethnicity, gender, socioeconomic status, or number of dual enrollment courses taken?

To answer these questions, we analyzed two existing large-scale administrative datasets using non-experimental methods, including ordinary least squares and logistic regressions.

The longitudinal nature of the data enabled us to control for preexisting student characteristics. Limiting analyses to only CTE students also provided us with an additional level of control for preexisting characteristics, because both CTE dual enrollment students and their non-participating peers had technically-oriented goals while in high school. Although we assume that these students are similar to one another in terms of motivation, career and academic aspirations, and high school experiences, this restriction will not necessarily eliminate all preexisting differ-ences. Without a randomized design, we were unable to control for all possibly important preex-isting characteristics. Positive findings may therefore be due to unmeasured factors that are not accounted for in our models rather than to dual enrollment participation. By not controlling for important factors affecting the decision to participate in dual enrollment, it is possible that our models may generate what appear to be positive impacts when in fact there are no such impacts or there are negative impacts. Future research should seek additional control variables as well as use experimental and quasi-experimental designs to establish a causal relationship between dual enrollment participation and educational outcomes.

Findings: Florida

Florida has a long-standing statewide dual enrollment program that is supported by state legislation. All high school students in the state who meet eligibility criteria (a 3.0 grade point average and passing the appropriate college placement exam for general education courses) must be offered the opportunity to participate in dual enrollment.

The State of Florida maintains a comprehensive student unit-record system for all students enrolled in the public education system. Using unique student identifiers, the data track students over time and across secondary and postsecondary public institutions in Florida. For this study, we had access to student records for the 2000-01 and 2001-02 high school graduating cohorts. Variables in the dataset included all courses taken in high school and college; dual enrollment courses and achievement in them; academic achievement measured as final high school grade point average and semester grade point averages in postsecondary education; and demographic information including age, gender, race/ethnicity, English language proficiency, and citizenship. The State of Florida does not identify students as CTE concentrators. In order to determine which students were in CTE programs, we relied on the National Center for Education Statistics’ definition of CTE participants: those who had completed three or more credits in a Specific Labor Market Preparation (SLMP) area (such as Technology and Communications, Health Care, or Business).

The analyses examined the impact of participation in dual enrollment on subsequent ma-triculation into and persistence in Florida postsecondary institutions. We conducted the analyses twice, once for all students in the state (to determine the effect of dual enrollment participation generally) and once for only students in CTE programs (to determine the effect of dual enrollment participation for this subgroup of students). The analyses included an array of control vari-ables in order to minimize the influence of student characteristics (other than dual enrollment participation) on outcomes.

We ran each model for (1) the entire sample and (2) only students categorized as being in CTE programs of study. Then, because some authors have suggested that dual enrollment programs spanning multiple semesters may be more effective in improving outcomes for middle-achieving students, we conducted a second set of analyses that accounted for students’ participation intensity, defined as the number of dual enrollment courses taken.

We found a positive relationship between dual enrollment participation and short- and long-term outcomes for both the full sample and the CTE sub-sample. Dual enrollment was posi¬tively related to students’ likelihood of earning a high school diploma. For the full sample, students were 4.3 percent more likely than their peers to earn a diploma. CTE students who participated in dual enrollment were 1 percent more likely than their peers to earn a high school diploma.

Participation in dual enrollment was positively related to enrollment in college for both the full sample and the CTE students. Dual enrollment participation also increased the likelihood of initially enrolling in a four-year institution (by 7.7 percent for all students and 8.6 percent for CTE students). For students who enrolled in postsecondary education, dual enrollment participation was also positively related to their likelihood of enrolling full-time.

Dual enrollment students, whether in the full sample or the CTE sub-sample, were statistically significantly more likely to persist in college to a second semester. They also had statistically significantly higher postsecondary grade point averages one year after high school graduation. The difference ranges from as low as 0.21 points for all students to as high as 0.26 points for CTE students only.

Of those students ever enrolled in postsecondary education, dual enrollment participation was positively associated with their likelihood of remaining enrolled two years after graduating from high school. Dual enrollment students’ grade point averages after two years of college were also statistically significantly higher than their non-participating peers. Both of these relationships held true for the full sample and the CTE sub-sample.

The relationship between dual enrollment participation and grade point average continued throughout students’ postsecondary careers. Dual enrollment students’ cumulative college grade point averages three years after high school graduation were statistically significantly higher than those of their non-participating peers. Finally, dual enrollment students had earned more postsecondary credits three years after high school graduation (indicating that they had made more progress toward a degree). For the full sample, dual enrollment students had earned 15.1 more credits than their non-dual enrollment peers. CTE dual enrollment students had earned 15.2 more credits than their non-dual enrollment CTE peers. Although some of these credits were likely earned through dual enrollment, it is also likely that some were earned after matriculation into postsecondary education.

Participation intensity had little impact on short- and long-term outcomes, however. The statistically significant effect of dual enrollment participation versus non-participation generally remained the same, regardless of whether students took one, two, three or four, or five or more dual enrollment courses.

Findings: New York City

New York City’s public university system, the City University of New York (CUNY), has a long-standing dual enrollment program, College Now. Every two- and four-year college in the CUNY system and nearly 300 high schools are involved. Between 2001 and 2006, 113,796 stu-dents participated in College Now.

The data collected from College Now come from two sources—the College Now office and the City University of New York’s Office of Institutional Research. We had access to data on students who graduated from any of the 19 vocational high schools in New York City and enrolled in CUNY in 2001 and 2002. The dataset included demographic variables, College Now courses taken by students and grades earned, information on students’ high school academic per-formance, and semester-by-semester information on credits attempted, credits earned, and grades in all courses taken throughout the CUNY system.

The analyses examined the impact of participation in College Now on matriculation into and persistence in CUNY for CTE students. We ran each regression for the entire sample, com¬paring students with any degree of participation in College Now to their nonparticipating peers. Then, as with Florida, we conducted a second set of analyses that accounted for students’ participation intensity.

Though not as consistently as in Florida, we also found positive short- and long-term out¬comes of dual enrollment participation in New York City. College Now participants were more likely than their peers to pursue a bachelor’s degree. College Now participation was also positively related to students’ first-semester grade point averages. College Now participants had first-term grade point averages 0.133 points higher than those of non-participants. Finally, College Now participation was positively related to students’ overall progress toward a de¬gree. Three-and-a-half years after their initial enrollment in postsecondary education, College Now participants had earned significantly more college credits than their non-participating peers.

Unlike in Florida, we did find some influence of participation intensity in New York.

Specifically, the positive relationship between College Now participation and first-semester GPA seems to be due to the impact of taking two or more courses, rather than on participation more generally. Additionally, although College Now participation by itself did not influence student full-time enrollment, students who took two or more College Now courses were 3.5 percent more likely to enroll full-time than non-participants (a statistically significant difference), whereas students who took one course were no more likely to do so. Intensity of participation appears to be more important for long-term outcomes. All three long-term outcome variables (persisting to the second year of college, grade point average after four semesters, and progress toward a degree) were positively related to College Now participation when intensity was taken into effect.

Findings: Outcomes for Sub-groups

Part of the argument for expanding access to dual enrollment programs is based on an as-sumption that some types of students, particularly low-income or low-achieving students, may benefit from early exposure to the demands of college courses. This argument is based on evidence that such groups typically have less-positive postsecondary outcomes than their more ad-vantaged peers and on a desire to help eliminate these gaps in college achievement. Thus, we examined whether the positive findings were particularly strong for certain sub-groups of students. We ran separate regressions of the impact of dual enrollment for each of three sub-groups and then tested whether each group demonstrated similar marginal effects, using a standard t-test at 5 percent levels. Statistical differences in effects can be treated as differences in gains from dual enrollment participation for the various sub-groups. Our analyses focused on differences in terms of gender (since males are increasingly underrepresented in higher education), high school achievement, and socioeconomic status.

Given the limited size of our New York City sample, we could examine differences only in outcomes in terms of gender. We found no significant differences between males and females. In Florida we were able to run analyses for all of the sub-groups. We found that, in many cases, male and low-income students benefited more from dual enrollment participation than their peers. On some measures, students with lower high school grades also benefited to a great¬er extent than students with higher grade point averages. On some measures, these sub-group differences were true for both the full sample and the CTE sub-sample; on other measures, the differences were found only for the full sample.

Conclusions and Implications

The findings provide an encouraging, though not definitive, picture of dual enrollment as a strategy for encouraging student access to and persistence in postsecondary education. We find positive effects from participation in all areas, and though we would encourage future research to use additional control variables for student background and motivation, we believe there is evi-dence that dual enrollment can be an effective transition strategy for a range of students.

The findings have implications for policy and programs. First, they indicate that the spread of dual enrollment programs may be warranted and that states and programs should consider ways to encourage participation for a broad range of students. However, additional research is needed to further establish the efficacy of dual enrollment as a promising high school and CTE reform strategy, including analyses using randomized or quasi-experimental designs to eliminate some of the shortcomings of our regression analyses and further examining sub-group differences in outcomes in order to better understand which groups of students may benefit most from dual enrollment participation.

The implications of the findings presented in this report can be seen as pertaining to two separate arenas: dual enrollment generally, and CTE reform. These are as follows:

Dual Enrollment

1. Because dual enrollment participation can benefit a range of students, expand currently restrictive eligibility requirements for dual enrollment.
2. Consider creating dual enrollment sequences since our findings suggest that students benefit from taking more than one dual enrollment course.
3. Expand outreach to underserved populations and provide dual enrollment courses tuition-free for low-income students (if not for all students) in order to ensure that such students are able to take advantage of dual enrollment opportunities.

CTE Programs

1. Expand dual enrollment options for CTE students, particularly in places where these students are not currently offered dual enrollment opportunities.
2. Continue to integrate dual enrollment into CTE pathways and programs.

In conclusion, we present very encouraging, though not definitive, findings that dual enrollment participation is related to positive postsecondary outcomes. This positive association is particularly strong for groups of students who are struggling in postsecondary education, especially males and low-income students.