Student Success Charts

Was this forwarded to you by a friend? Sign up for the On Student Success newsletter and get your own copy of the news that matters sent to your inbox every week.

As a break from the heat across much of the northern hemisphere, here are some charts on affordability, the wider returns to education, graduate financing, part-time enrollment, cheating and transfer. A surprising number of them are also reminders that charts answer the question they were designed to answer, which is not always the question we most want answered.

The best deal in town

In a recent Gallup poll on confidence in higher education, the difference between public perceptions of affordability at two-year and four-year institutions is striking.

First, two-year institutions.

By contrast, four-year institutions are not doing well on perceptions of affordability.

This restores my faith in opinion polling and humanity a little. Two-year institutions generally are much more affordable than four-year institutions. They are also fundamentally different institutions, so this is not a straightforward comparison of two versions of the same product.

What strikes me is that the public appears to have a relatively coherent idea of what two-year colleges are for and believes they perform those functions reasonably well. Its judgment of four-year institutions is much more fractured. The four-year sector’s affordability problem may therefore be compounded by uncertainty about what people believe they are receiving in return.

The returns we leave out

Discussions of higher education ROI almost always frame it as a return to the individual. This chart is a reminder that the potential return to society may also be substantial, including through reduced reliance on social services and the safety net.

The chart shows that participation in most means-tested programs declines with educational attainment. It does not establish that education itself caused the difference. Age, income, labor-market attachment, health, and selection all matter.

But even allowing for those limitations, the chart is a reminder that the returns to education do not accrue only to individual graduates. If public policy evaluates higher education primarily through graduate earnings, it is counting only the most visible and easily measurable portion of the return.

California has a Grad PLUS problem

With Grad PLUS loans coming to an end in the US, exposure varies considerably across states. California residents received far more Grad PLUS funding than those in Texas or Florida, the two other largest states.

California’s exposure may reflect its size, high graduate enrollment, expensive programs, or a larger role for private institutions than we might have guessed. Those possibilities imply very different policy consequences. The dollar total identifies where to look; it does not yet tell us what we will find.

New York and Pennsylvania are also more exposed than I might have predicted.

The aggregate exposure is highly concentrated geographically, but these charts cannot tell us where the removal of Grad PLUS will be most disruptive. For that, we would need to know how many students and institutions depend on the program, and how easily they can replace the financing.

Definitions all the way down

In a recent Ithaka S+R report on part-time students, a map of part-time enrollment rates yields some perplexing insights.

New Hampshire has the highest rate of part-time study largely because the online giant Southern New Hampshire University (SNHU) is located there. As the report notes.

This is no surprise. The big surprise is Utah where Western Governors University (WGU) is located and where we might expect a big part-time student population. But because of how terms and courses of study are structured at WGU, many students count as full-time. In the latest IPEDS data, all its undergraduates are listed as full-time.

Arizona was also not as heavily part-time as I would have expected, given the number of large, online-heavy institutions based there. ASU Digital Immersion reports that just 60 percent of its students are part-time.

This makes state comparisons far less straightforward than the map suggests. New Hampshire looks exceptionally part-time because SNHU classifies many online students that way. Utah looks exceptionally full-time because WGU’s academic model results in many of its students being classified as full-time. Some of the apparent geographic difference is therefore a difference in institutional structure and reporting conventions.

That matters because classifications such as part-time and full-time are often used as proxies for student circumstances and support needs. At institutions organized around nontraditional calendars and competency-based models, those categories may tell us surprisingly little about how students actually experience college.

“It’s tough to make predictions, especially about the future.”

The wisdom in this line, often attributed to Yogi Berra, is even more important now than ever. I am a Paul Krugman fan, but his predictions about the internet have not aged well.

The point is not that Krugman was uniquely bad at forecasting. Smart people routinely underestimate changes that initially look marginal, especially when those changes depend on infrastructure, complementary innovations and altered behavior that do not yet exist. This seems worth remembering as we encounter confident predictions about AI, the disappearance of universities, the end of degrees and whatever else is scheduled to happen by next Thursday.

The post from which I nicked borrowed this is interesting and well worth a read.

Something is amiss

From time to time, I rail against the moral panic about cheating. But for the two of you who have not yet seen it, this chart from a Brown University professor, comparing scores on a take-home midterm with scores on an in-class final, certainly suggests that something is amiss.

The orange dots show students’ performance on the midterm and the gray dots their performance on the final. The gap between the two is often striking and is consistent with unauthorized assistance on the take-home exam, including possible use of AI. It is not conclusive evidence of either AI use or cheating. The assessments differed in format, timing, and conditions, and some students simply perform much worse under in-person exam pressure.

Even with those caveats, the size and consistency of the gaps make this difficult to dismiss. The chart also points toward a larger assessment problem. If institutions want to make claims about cheating, they will need stronger evidence than detection software and anecdotes. They may also need assessment designs that allow instructors to compare what students can produce with assistance and what they can independently explain or apply.

Transfer reform, allegedly

This is a small and informal poll about progress in sorting out the mess that is student transfer, so I would not treat the percentages as a national estimate. Still, only 12 percent of respondents believed their state had made significant progress, while 60 percent reported no progress or were unsure.

The “not sure” response may be almost as revealing as “no progress.” Credit-transfer reform has generated years of legislation, articulation agreements, common course-numbering initiatives and technology projects. If people working in the field cannot see whether those efforts have produced meaningful change, that is itself a measure of the problem.

Questions before answers

I have a love-hate relationship with lists, but this one is useful because most of the questions are designed to interrupt the rush from data to action. What assumptions are embedded in the measure? What alternative explanation fits the evidence? What would change my mind? Those are especially important questions in student success, where institutions have become very good at producing dashboards and rather less consistent at converting them into understanding.

I borrowed this from a branding and marketing newsletter (in the same way that raccoons borrow unattended food) so the questions tend in that direction. Next month, I’ll post my own list oriented toward research and student success.

The administrator danger zone

But I love 2×2s.

This Bad Administrator Matrix is especially useful, although experienced university administrators know that the real danger is someone moving among all four quadrants during a single committee meeting.

The main On Student Success newsletter is free to share in part or in whole. All we ask is attribution.

Thanks for being a subscriber.