Colleges Matter

At the end of Anya Kamentez’s new book DIY U (look for a review here in the next few days), there’s a fascinating quote from Dennis Littky, a psychologist who runs 50 charter schools across the country:

“When kids drop out of high school people blame the high school. When they drop out of college they blame the high school or the kid. What about the college? Do you take any responsibility?”

Littky is getting at one of the biggest disconnects between how we think about K-12 and higher education. In the former, schools matter immensely, and student failure can be largely blamed on bad teaching and schools. In the latter case, the university matters little—low-income students fail because of their background and they are just harder to educate.

But what makes this belief even stranger is it’s not backed up by data. Research published in Crossing the Finish Line and elsewhere has concretely demonstrated that the odds of graduating for two academically identical students will vary significantly based upon the type and quality of the school in which they first choose to enroll. Students who attend a school above their academic qualifications are more likely to graduate than the academically identical student who attends a school of lesser quality.

A study released this week about the increasing length of time students need to complete their bachelor’s degree lends further credence to the notion that schools matter. (Unfortunately, I can’t distribute the full copy of the paper, “Increasing Time to Baccalaureate Degree in the United States,” but the abstract is here and some media coverage is here.)

Using data from a 1972 and 1988 study that tracked recent high school graduates for eight years, the authors looked at the average amount of time it took bachelors degree recipients to obtain their credential. The authors found that the amount of time it takes to earn a degree went up significantly over the 20 year period between the two studies:

We find evidence of large aggregate shifts in time to degree: in the 1972 cohort, 58% of eventual BA [bachelor's] degree recipients graduated within four years of finishing high school, but the for the 1992 high school cohort only 44% did so.

As a result, the average time to degree increased from 4.69 to 4.97 years.

But what’s most interesting is the second part of their findings:

This extension of time to degree, and the associated reduction in “on-time” degree completion, did not occur evenly over the different sectors of higher education. Time to degree increased mostly among students beginning college at less selective public universities as well as community college.

In other words, where a student started his or her education mattered in the amount of time it took them to earn a degree. (Remember, these are only recipients, so the lower graduation rate at these schools does not matter.)

Here’s table showing the breakdown in time to degree changes by type of first institution:

The first and likely explanation for the different increases would be that the longer time to degree is a result of the greater heterogeneity of today’s students. The increases in the number of minority and low-income students from 1972 through the present would suggest that less qualified students may simply take longer to get a degree.

But the research doesn’t bear out this finding. The authors looked at both student preparedness and demographics and found that neither had changed substantially since 1972, though test scores had improved slightly. The explanation simply cannot be explained away by blaming the student.

So then what does explain the increase in time to degree? The authors find two things primarily: increases in student-faculty ratio and longer hours worked by enrolled students.

But how does that apply to schools? Student-faculty ratio can be affected by two issues: state support for the college and enrollment expansion. If state support drops, then the student-faculty ratio is likely to increase because the institutions cannot hire as many instructors. Second, the student-faculty ratio could increase if the amount of instructors is held constant but a state ends up with more students enrolled. The authors looked at the increase of 18 year-olds in a state and found that population growth can create crowding effects at schools—making it harder to get necessary classes, straining resources, and having other effects that make it take longer to get a degree on time.

What about student work? In general, students hold a job while enrolled because the cost of higher education is too much for them to support without a job. During the period covered, the amount of financial aid available decreased in real terms, while the cost skyrocketed. Though the students may not have been less prepared, they had to work more for their education, decreasing the likelihood of success.

As this study and others have shown, universities matter in student outcomes. Schools’ financial aid allocation and state’s subsidy decisions have a real and dramatic effect on how quickly their students are able to degrees (to say nothing of whether or not they are able to earn one at all). In that regard, one must wonder whether re-purposing the  significant amounts of merit aid that public colleges and universities dole out to non-needy students each year might not have more beneficial effects elsewhere.

Tagged as: Colleges and Universities, Student Financial Aid, Time to Degree