Dropout-Adjusted Outcomes

A new paper measuring the economic outcomes for students who drop out of community college programs finds that earnings vary based on the number of credits earned and courses taken. 

February 16, 2015

Most research on the payoff of attending community college actually doesn’t measure the effect of attending, but rather what happens for those who graduate.

Yet when the majority of students who enroll in community colleges don’t complete their programs, the financial benefit should be adjusted given the likelihood of failure.

That’s the philosophy driving a recently published report that tries to measure the economic benefit of two-year college for the mass of dropouts. The report was published by the Center for Analysis of Postsecondary Education and Employment at Columbia University’s Teachers College.

“People look at what the completers do and they think, ‘Oh, well, that’s what I’m going to get,’ and that’s a somewhat inaccurate picture,” said Clive Belfield, one of the report’s authors and an associate professor of economics at the City University of New York's Queens College.

Relatively few individuals who enroll in community college go on to earn a credential, although the available national data is limited. Based of federal graduation rates, 22 percent of first-time, full-time students who enroll in community colleges earn an associate degree within three years, according to the center.

In general, community college provides fairly high labor-market returns for those who complete. Based on previous studies, women with an associate degree earn an average of 21 percent more than those with a high school diploma, and men earn an average of 13 percent more, according to the report.

But there’s a far more complex and varying breakdown of the labor market returns based on the number of credits earned and type of courses taken, Belfield said.

If colleges or prospective students are evaluating programs and majors based on their labor market returns, they need to look at the rate of completion so students know what their chances of actually seeing those payoffs are.

That sounds obvious, Belfield said. And yet, most research measures only the earnings of completers. Part of that is explained by the relative difficulty of gathering data on noncompleters. 

In this case, to measure the outcomes of students who enroll -- not just those who complete -- the researchers created an algorithm that predicts which award (associate degree, certificate or a one-year job-training credential) a student was most likely to receive based on his or her transcript data.

The report also compares the outcomes of completers and noncompleters based on the students’ stated intent or goal. The algorithm, intent and goal models delivered different outcomes, demonstrating that students often don’t pursue the path they intended. (Sometimes there’s a deliberate change of paths, but often this reflects students’ confusion about what courses they need to take, the study said.)

The researchers used data from all first-time, credit-seeking students within the North Carolina Community College System between the 2002-03 academic year through 2004-05. Students who transferred and went on to earn bachelor’s degrees were taken out of the sample. The student transcript data was merged with earnings data from the state’s unemployment office.  

For students who end their postsecondary education without an award, those who were predicted to complete certificate programs had worse labor market outcomes that those who were on the associate degree or diploma track.

The differences in earnings also vary depending on the field of study. In most of the 13 fields researchers looked at, there wasn’t a significant difference between students who earned an associate degree and the noncompleters, though that wasn’t true across the board.

In general, fields that are relatively high paying remain lucrative after dropouts are included in the data, but not by as much, according to the report.

The clearest example is in nursing. The earnings gain for female students who complete their nursing credentials is 104 percent. But when noncompleters are factored into the equation, the earnings advantage drops to 37 percent. It’s still positive, but much less so.

“To the student, if you’re really going to pin your hopes on nursing, you’ve got to make sure you complete,” Belfield said. “Because those big payoffs are only going to happen if you complete.”

Earnings differentials might be influenced by how skills can be applied across disciplines, Belfield said. For example, a noncompleter in a business program is more likely to have learned skills that can be applied elsewhere than has a nursing student. 

“The more specific the skill, the bigger the risk if you don’t complete,” he said.

One surprising finding of the study was that there wasn’t any evidence that progression mattered. Students who were farther along in a specific course of study were no better off than students who simply earned a collection of general college credits.

In fact, after controlling for the number of credits earned, the labor market returns actually declined for noncompleters who progressed farther into their studies. In other words, students who dropped out after taking a somewhat random assortment of classes were just as well off as those who followed a strict path of study.

The finding certainly needs to be looked at in further detail, Belfield said. But it could make a big difference in the way programs are structured and the way students sign up for courses.

This story has been updated to correct the name of the center that released the report.


Back to Top