The report from the U.S. Department of Education’s National Center for Education Statistics, released today, examines patterns of student loan repayment for two separate groups of borrowers — those who started college in the 1995-96 academic year and those who started eight years later, in 2003-04. Twelve years after beginning their postsecondary educations, the second group had paid off a smaller proportion of their student loans and had defaulted at a higher rate on at least one loan.

In addition to the rising price of college, multiple factors may have contributed to changing profile of student loan repayment. Students who entered college in 2003 would have graduated or left college around the time the U.S. entered the Great Recession. Changes in federal policy also have made options like income-driven repayment more popular. And experts say the composition of student loan borrowers has changed, too, as enrollment at community colleges and for-profit institutions spiked in the recession’s wake.

The report found that 12 years after first entering college, the median percentage owed on student loans by the first cohort of borrowers was 70 percent. For the second borrower cohort, that number was 78 percent. Over the same time frame, 18 percent of those in the earlier cohort had defaulted on at least one student loan, while 27 percent of the more recent cohort had done so.

Experiences were not uniform among those struggling to repay their federal loans. Faring worst were students who began a degree program but never received any credential and those who attended for-profit institutions. More than half of borrowers who began college in 2003-04 and attended a for-profit institution defaulted on at least one student loan within 12 years; more than a quarter of those who attended a community college did so.

Compared to the earlier cohort of borrowers, the only student subgroups who appeared to do as well or better in making progress toward paying off student loans after 12 years were those who attended a four-year institution or those who earned a bachelor’s degree. Student borrowers who entered college in 2003 and attended a public four-year institution had paid off an average of 61.7 percent of their remaining student loans. Among the earlier borrower cohort, this group paid off 63.1 percent of loans after 12 years.

Sandy Baum, a senior fellow at the Urban Institute who studies higher education finance, said overall loan repayment rates can be misleading without examining the types of borrowers taking out student loans. She said that, over time, many more students have enrolled in community colleges and for-profit colleges, and borrowers who attend those institutions tend to perform worse in paying off student loans.

“The issue is the composition of the borrowers has changed so dramatically,” Baum said.

David Baime, senior vice president for government relations and policy analysis at the American Association of Community Colleges, said it would stand to reason that, if students are borrowing more on average to pay for college, repayment would be more drawn out.

“It’s also possible that students simply decided to avail themselves of other repayment options that allowed for more time,” he said, adding that the second cohort “obviously would have run smack-dab into the Great Recession.”

Steve Gunderson, president and CEO of Career Education Colleges and Universities, which represents for-profit institutions, said defaults in his sector likely rose due to the recession and increasing enrollment attributed to the growing popularity of online courses.

“This is also a snapshot of the past and is in no way reflective of where this sector is today,” Gunderson said. “Our sector has evolved and is now focused on outcomes.”

For example, while enrollments are down across the for-profit industry in recent years, Gunderson said degree-completion numbers have largely held steady.

Short-term training programs are more popular than ever with policy makers who are eager to boost skills training they say will lead to higher employment. But the NCES report found that, among borrowers who started a postsecondary program in 2003-04, those who earned an undergraduate certificate, as opposed to an associate or bachelor’s degree, had a default rate of 29.9 percent — higher than if they dropped out entirely (29 percent).

Jeff Strohl, director of research at Georgetown University’s Center on Education and the Workforce, said the center’s research has found that less than 50 percent of certificates have any reasonable earnings returns.

“Like all postsecondary credentials, students today really need to look at how the level of attainment and program combine to provide opportunity,” he said via email. “Our research supports the long-term findings that college is worth the cost, but that general finding does not apply to every possible credential in every field.”

The report reconfirmed findings by many who study higher education that even students with small amounts of debt can default, said Robert Kelchen, an assistant professor of higher education at Seton Hall University. But Kelchen said it’s hard to tell what specific factors are driving the higher number of defaults and slower progress repaying student loans.

“It’s hard to tell what’s because of the recession, what’s because of the increase in student debt levels and what’s a result of changes in student loan repayment options,” he said.

Kelchen said it’s important for the Education Department to continue to track default rates over the long term. And he said more data on student loan repayments could help show the effects of payment options like income-driven repayment plans and also disentangle patterns of loan repayment for graduate and undergraduate students.