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.

It has been a minute since I wrote one of these posts. So what’s been happening in student success? Quite a lot, apparently: institutions are trying to figure out what to do with AI, predictive analytics may be changing the outcomes they are supposed to predict, a free-college program made graduation rates worse by some measures while producing more graduates, and we have 14 years of evidence on ASAP.

The more things change

EAB has a new report on student success in an era of constraint, based on a survey of 459 student success leaders. The title is well chosen. Institutions are under growing pressure to improve retention and completion while simultaneously dealing with budget constraints, staffing shortages and difficulties coordinating student support across the institution.

The data show a field that is fairly clear about what it wants to achieve, has been trying to do much of it for years, and still struggles with the organizational capacity to do it consistently. AI adoption is slow, but to the extent it is being adopted, it is being layered onto that existing operating model rather than prompting much rethinking of the model itself.

Take institutional priorities. Retention remains overwhelmingly at the top, followed by familiar goals around persistence, academic progress and student engagement.

The fact that AI support for student success is last is notable. These aren't entirely comparable categories—retention is an objective while AI is a means of achieving one—but the fact that it is so low a priority is still striking given the enormous amount of attention AI is receiving elsewhere in higher education.

And when institutions are using AI for student success, they aren't generally doing anything particularly revolutionary with it.

The most common uses include things like drafting student communications, chatbots and automating repetitive tasks. The field has discovered generative AI and apparently decided: “Excellent. Now we can write that advising email faster and with fewer typos!”

There is nothing inherently wrong with that. Saving advisers time is useful, and some of these applications may ultimately prove quite valuable. But it also tells us something about how institutions are approaching AI. So far, they appear much more interested in using it to make the existing student-success operating model more efficient than in asking whether AI makes a different student success system possible.

That matters because the existing model isn't exactly firing on all cylinders.

Despite the enormous amount of attention institutions have devoted to retention and student success, a substantial share still report falling short of the goals they set for themselves. And this is where I think the survey gets particularly interesting. Asked what is preventing greater progress, respondents point to the budget, but surprisingly they don’t seem to be that worried about inadequate technology.

Instead, they point to problems such as coordination across departments, resistance to change, leadership alignment, staffing capacity and difficulty engaging students.

These are the same sorts of organizational problems that have bedeviled student-success efforts for years. Better technology and better data may help institutions address some of them, but they don't make them disappear.

And that seems particularly important as we start talking about agentic AI in student success. There is an understandable tendency to focus on what the agents will be capable of doing: identifying students who need help, answering questions, coordinating services, making recommendations and eventually taking actions on students' behalf.

But many of the hardest problems identified in this survey aren't really problems of technological capability. They are problems of institutional operating capacity.

If an institution struggles to get advising, financial aid, faculty, student affairs and other units to coordinate around a student today, adding an AI agent doesn't automatically solve that problem. In fact, giving an agent the ability to act across those same organizational boundaries may make the unresolved problems even worse. Who has authority to act? Which unit owns the workflow? What happens when policies conflict? Who is accountable when the agent gets it wrong?

The interesting question may therefore be less “What can AI do for student success?” and more “Is the institution organized to make useful use of what AI can do?”

The problem with predictive analytics

I have plenty of complaints about predictive analytics in student success. They are one of those “shiny” things that people seem irrationally to love, causing them to sometimes treat prediction as the goal rather than asking what happened and why, as well as jumping over the descriptive and diagnostic work that ought to precede it. The familiar Gartner analytics hierarchy is old but still useful on this point.

But a LinkedIn post this week illustrated an additional problem with predictive analytics.

I think the image makes two important points about how we use predictive analytics in student success.

The first is relatively simple: a prediction is not evidence that the predicted outcome will occur. An at-risk flag, for example, is an estimate based on patterns in existing data. However accurate the model may be, the prediction itself is not proof of anything about an individual student.

The second point is more complicated. Once we act on a prediction, we create what the image describes as a loop: prediction → expectation → decision → outcome. The initial prediction can shape how we think about the student, which can shape the decisions we make, which in turn can shape the student's eventual outcome, which may then appear to confirm the original prediction.

The danger comes when we then treat that outcome as confirmation of the original prediction or of the intervention that followed it. As the image puts it, “the system may seem right because people acted like it was.” The outcome has not occurred independently of the prediction; it has emerged from a process partly set in motion and shaped by that prediction.

That doesn't make the outcome meaningless. But it means we need to question every step in the loop rather than treating the eventual outcome as validation of the whole process. Was the prediction accurate? What expectations did it create? What decisions followed? How did those decisions affect the outcome?

Prediction is not proof, and an outcome is not necessarily validation.

When worse graduation rates mean better student success

A new NBER working paper looks at the long-term effects of Tennessee Promise, the state's free community college program that fills in any gap between full-time tuition and fees and a student’s other sources of grant aid. Promise is often discussed as a college affordability policy, but the researchers find that it also substantially changed where—and whether—students enrolled in college.

The program increased postsecondary enrollment at two-year institutions by about 5.4 percentage points. Importantly, the authors estimate that roughly three-quarters of the increase came from what they call “New Access Enrollees”—students who otherwise would not have enrolled in college immediately after high school.

That is an important finding in itself. But the finding I find much more interesting is what happened to one of the standard ways we measure whether community colleges are successful. Their graduation rates got worse.

That sounds like a fairly damning student-success result. Except that at the same time, the colleges were producing more graduates. Associate-degree production increased by roughly 15%. The number of students transferring out increased by approximately 24%.

And when the researchers follow individual students rather than looking at institutional graduation rates, they find that Promise increased associate-degree attainment. By ages 23–24, students exposed to Promise were 4 percentage points more likely to have earned an associate degree or higher. There is also little evidence that this came at the expense of bachelor's-degree attainment.

So we have the wonderfully counterintuitive result that community colleges had lower graduation rates while producing more community-college graduates. The explanation makes sense once you look at who Promise brought into the system.

Many of the additional students were students who otherwise wouldn't have attended college at all. Those marginal students were less likely to complete than the students community colleges had previously enrolled, putting downward pressure on the institutional graduation rate. Other students transferred to four-year institutions without first earning an associate degree—another potentially successful student outcome that can hurt the originating college's graduation metric. The authors explicitly identify both mechanisms as explanations for the apparent contradiction.

And this is why I think this paper matters beyond Tennessee Promise.

We use graduation rates constantly as measures of institutional student success. But a graduation rate is a ratio, and changes in whom an institution serves can change that ratio in ways that have surprisingly little to do with whether the institution is producing more successful outcomes.

Expanding access changes the denominator. Successful transfer can hurt the numerator. An institution can therefore produce more graduates, more transfers and greater individual educational attainment while simultaneously reporting a worse graduation rate. Neither measure is necessarily wrong. They are answering different questions.

• The graduation rate asks something like: What proportion of a particular group of students entering this institution completed a credential here within a specified period?

• Degree production asks: How many students actually earned degrees?

• And the student-level analysis asks the question I ultimately care about more: Did the policy increase the likelihood that an individual student attained a credential?

In this case, those questions produce very different answers. That is a useful warning whenever we use institutional performance metrics to judge student success.

Quick, adopt the model ASAP

SUNY recently announced another expansion of its student success programs Advancing Success in Associate Pathways (ASAP) and its four-year counterpart, Advancing Completion through Engagement (ACE). The programs will now operate at forty-four SUNY institutions, with a goal of serving 10,000 students this year.

ASAP originated at CUNY and has since been replicated at institutions in Ohio, Colorado, Arizona and California. The basic model combines a fairly intensive package of supports: dedicated advising and career guidance, financial assistance for expenses such as textbooks and transportation, priority registration and access to high-demand courses, and sometimes block scheduling. Students, in turn, generally commit to full-time enrollment and other program requirements.

And there really isn't much reason left to argue about whether ASAP improves degree completion, because it does.

The Ohio replication, for example, found that after six years 44% of ASAP students had earned a degree compared with 29% of control-group students. Fourteen percent had earned a bachelor's degree compared with 9% of the control group.

But we now have something even more valuable: Fourteen years of follow-up data from the original randomized CUNY evaluation.

The chart is fascinating because it shows both the power and the limits of looking at short-term graduation rates. At the end of the three-year ASAP program, 39.5% of the program group had graduated compared with 22.4% of the control group—a huge 17.1 percentage-point effect. The gap then narrowed as some students in the control group eventually graduated. But it never fully disappeared. After 14 years, 57.7% of students offered ASAP had earned a degree compared with 50% of the control group.

The researchers interpret that pattern as evidence that ASAP did two things. It helped some students graduate faster than they otherwise would have, but it also appears to have caused some students to graduate who would never have earned a degree without the program. That is an impressive result.

But there is another finding in the 14-year study that complicates the story. In the final two years for which earnings data were available, ASAP students earned about $36,000 annually compared with about $38,000 for the control group. That difference was not statistically significant, so we shouldn't interpret it as evidence that ASAP reduced earnings. The finding is that researchers could detect no earnings benefit despite a persistent increase in degree attainment. That is particularly interesting because the six-year Ohio study had found something different. By Year 6, Ohio ASAP participants were earning $1,948 more than the control group, an increase of about 11%. The authors attribute this in part to the kinds of majors students chose.

I don't think the new 14-year CUNY finding means ASAP failed. Degree completion is an important outcome in its own right, and ASAP produced a large and remarkably durable improvement in it. But it does complicate the tendency to move casually from more graduates to better economic outcomes. Those are two different claims and require two different pieces of evidence.

It also makes me think again about my longstanding concerns about ASAP.

I have previously criticized the model for its cost, its focus on full-time students, and the difficulty of determining which parts of a large bundle of interventions actually produce the results. Those concerns remain. But increasingly I think the more important issue is scale.

SUNY's latest announcement is a useful illustration. Its goal is to serve 10,000 students across 44 institutions. That sounds like a lot—and for the 10,000 students involved, it certainly is. But it works out to an average of about 227 students per institution in a system enrolling well over 300,000 students.

A targeted intervention can be highly effective without being a scalable operating model for student success. And there is a risk in allowing the success of boutique interventions to distract us from the much harder problem of improving the experience of the entire student population.

There will always be students who benefit from intensive programs and specialized support, and we should absolutely provide those programs when the evidence supports them. But the larger student-success challenge is figuring out how to remove barriers and improve the educational experience for all students, including those who cannot enroll full time or participate in a highly structured program.

This is also why I get nervous when a particularly successful intervention becomes a “model” that institutions are encouraged to replicate. ASAP was developed for particular students, with particular participation requirements, backed by a relatively intensive and expensive bundle of services. Evidence that it works is extremely valuable. But that doesn't mean the answer to student success everywhere is ASAP, any more than it is Guided Pathways, the National Institute for Student Success (NISS), or whatever model comes next.

Evidence that something works is the beginning of the strategy question, not the end of it. We still need to ask: for whom, under what conditions, at what cost, and at what scale?

Musical coda

Since writing the Tennessee Promise section I’ve had this song playing in my head. A very young, spiky-haired Rosanne Cash. And if you ever get a chance to see her live, “slip away and put your jewellery in hock” and do whatever it takes to go see her.

And yes, her version is better than her dad's.

Evidence that sharing this newsletter improves student success remains limited. But why wait for the RCT? Forward it to a colleague.

Thanks for being a subscriber.

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.

What chart and listicle goodness do I have to share this month?

For education to lead to good jobs there need to be good jobs (obligatory John Burn-Murdoch post)

We spend a lot of time asking whether higher education is preparing students well enough for the labor market. The Financial Times’ John Burn-Murdoch turns the question around. Perhaps we should also ask whether the labor market is producing enough good jobs for the people education prepares.

What makes the chart striking for me is that the US advantage isn't confined to highly educated workers. Americans earn more than their British counterparts among degree holders and professional workers, but also among people without degrees, retail workers and factory workers. Education can create human potential, but the economy determines how much of that potential gets translated into good jobs and wages.

Hollowing out the middle

A fascinating analysis in the Journal of Economic Geography looks at how Americans have become geographically sorted by education. The researchers define high-skill workers as those with a bachelor's degree or higher, and middle-skill workers as those with a high school diploma, some college or an associate degree.

The charts take a little explaining. In 1970 there was a wide spread of local labor markets, but places with lots of high-skill workers also tended to have lots of middle-skill workers—the upward slope. By 2019 that had completely reversed. High- and middle-skill workers are now increasingly concentrated in different labor markets—the very strong downward slope.

What does it mean for higher education when people with different levels of education increasingly don't even inhabit the same local labor markets? Among other things, it changes both the populations local institutions serve and the labor markets they are preparing them for.

And geography isn't the only way Americans are becoming more socially separated.

The great social divide

Increasingly it includes how often we hang out with others and whom. In 1990 there was almost no educational divide in friendship. By 2024 there was a large one. Americans without a college degree were much more likely to report having no close friends and much less likely to have six or more.

And this may be becoming more important. A separate analysis shows that young Americans are spending much more time physically alone. Before COVID, only about 3.5% of 23–29-year-olds spent an entire waking day without being physically with another person. By 2025 that was 9.6%—almost three times the pre-pandemic level

Some of that appears related to remote work, but not all of it. Society reopened; young adults haven't completely returned to their old social routines. Put these two charts together and all our conversations about belonging in higher education start to look rather more consequential.

Children of the 1960s

Apparently the 1960s really were far out, man.

Community college enrollment grew nearly five-fold during the decade, a rate of expansion unlike anything before or since. It is a useful reminder of just how much of the higher education system we now take for granted was built during one extraordinary period of expansion.

Too much success?

CUNY has succeeded spectacularly in attracting students to computer science. Now it has a different problem: faculty capacity hasn't remotely kept pace.

And this isn't simply a matter of institutions failing to respond. Recruiting and retaining computer science faculty is difficult when colleges are competing with much higher salaries outside academia.

Growing demand for a program doesn't help students much if institutions can't provide the capacity needed to serve them.

The seven deadly sins of student success

Jobs and the young

Teen employment has fallen enormously over the past several decades. Even after a recent rebound, only 35.5% of 16–19-year-olds worked this summer, compared with well over half for much of the postwar period.

I'm not sure this is entirely good news for student success. Jobs aren't simply a source of income. They are also one of the places young people learn how to work—showing up, dealing with other people, taking responsibility, and acquiring experience.

And work isn't the only part of young people's lives policymakers are worried about.

Bans don’t work, Part 75

It's early days, but Australia's social media restrictions don't appear to have changed young people's actual use very much yet.

Before the restrictions, 85.9% of children surveyed had used an age-restricted platform in the previous four weeks. Three months later: 81.5%. Statistically significant, perhaps. Transformative, not so much.

China to US re demographic cliff: hold my Tsingtao

China's preschool enrollment has fallen by a third since 2020, from 48 million to 32 million, and nearly 30,000 kindergartens and primary schools closed in 2025. The US demographic cliff suddenly looks rather more like a curb.

Things I wish I’d known

I wish someone had given me this list when I first became an analyst. It's nominally aimed at junior strategists, but much of it applies to anyone trying to think critically about a market or practice—which includes most of us working in EdTech and student success.

Envy may be a deadly sin. Sharing isn't. If you enjoyed this post, send it to someone who might enjoy it too.

Thanks for being a subscriber.

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.

Part-time students, even those studying online, do not necessarily struggle to pass their courses. A new national study from Australia, Is the Problem Us? Helping Part-Time Students Succeed in a Full-Time World, from the Australian Centre for Student Equity and Success (ACSES) at Curtin University, combines national sector data with a more detailed analysis of student-level data from four universities. In those four institutions, part-time and full-time online students had virtually identical raw course-success rates. The larger disparity appeared in retention: part-time students could succeed in one course or semester and still struggle to keep moving through the program.

That distinction in where students are coming unstuck matters because it changes where we look for the problem. If part-time students cannot do the academic work, institutions might focus on tutoring or academic preparation. If students are passing courses but not progressing, the problem may instead lie in the student journey: in course availability, program sequencing, compulsory attendance, financial pressure and policies built around uninterrupted full-time study.

The ACSES study is one of two new reports examining part-time student success from markedly different directions. The other, from Ithaka S+R in the United States, begins with the available causal evidence and is consequently drawn toward comprehensive support programs designed to increase academic momentum. The ACSES report begins with national and institutional data alongside part-time students’ experiences, exposing the smaller obstacles that accumulate across their journeys.

Reading the reports together is instructive. They demonstrate how different use of evidence can produce different diagnoses of the same underlying problem, and why the most rigorously evaluated intervention is not necessarily the one most closely aligned with the lives of the students being studied.

Part time is something students do, not who they are

Part-time enrollment is too large to be treated as a marginal category. In Australia, 360,125 domestic university students—35 percent of the total—studied part time in 2024.

In the United States, Ithaka reports that part-time students represented 33 percent of fall 2023 enrollment and 49 percent of students enrolled at some point during the complete 2019–20 academic year. A third of students is not a special population. At that scale, part-time participation is part of the basic operating model of higher education, whether institutions acknowledge it or not.

It is also a population with permeable boundaries. Among US students entering college in 2011–12, 23 percent of those enrolled exclusively full time in their first year moved into part-time or mixed enrollment in their second. Movement also occurred in the other direction: 26 percent of those who began exclusively part time moved into full-time or mixed enrollment.

The Australian institutional data reveal similar movement within the academic year. Across 2022 and 2023, 34 percent of students classified as full time across the year studied part time during at least one principal semester.

The apparent size of the population also depends on when and how it is measured. Ithaka’s fall snapshot places the US part-time share at 33 percent, while its full-year measure captures 49 percent. These figures describe different observation periods rather than a sudden change in student behavior, but the difference shows how much a single census date can conceal.

Australian institutional policies create another form of variation. Some universities calculate load using only the two principal semesters, while others include summer study and aggregate credits across the complete calendar year. At two of the four universities studied, 5,026 students completed what the government considered a full-time annual load but remained classified as part time under institutional policy because their summer credits were excluded. Meanwhile, 87 percent of summer-term enrollments across the four universities were part time, even though many of those students maintained full-time status across the year.

Part-time status is therefore neither a fixed student identity nor a consistently defined condition. It describes a momentary relationship among a student’s circumstances, course load, academic calendar and institutional rules. Part-time status is a shifting category.

Passing courses, losing momentum

Ithaka identifies online enrollment as one of the characteristics associated with part-time study but does not explore it much further. The ACSES report does, revealing an intriguing distinction between success within courses and progression through a program.

Nationally, the course-success rate for part-time online students was only four percentage points below that of full-time online students (Figure 4A). In the student-level data from four universities, the raw course-success rates for part-time and full-time online students were both approximately 79 percent, with no statistically significant difference between them (Figure 4B).

Online study could itself be contributing to poorer outcomes, but this comparison holds delivery mode broadly constant: it compares part-time online students with full-time online students. Within the same mode of study, enrollment intensity made little difference to raw course-success rates in the four-university data.

The difference becomes more pronounced when the report turns from passing courses to retention, defined in the four-university analysis as remaining enrolled at—or graduating from—the same institution the following year. Among online students, the part-time/full-time retention gap was significant even though the raw course-success rates were virtually identical (Figure 7B).

The adjusted analysis tells a similar story. Compared with full-time online undergraduates, part-time online students had 0.77 times the odds of passing but only 0.67 times the odds of being retained. These estimates control for age, disability, socioeconomic status and several other characteristics. Retention, however, does not reliably distinguish among permanent departure, transfer and a temporary pause.

But the distinction between passing courses and retention changes the diagnosis of the problem. Many part-time online students appear capable of passing the courses they take. The larger difficulty arises in maintaining enrollment across semesters and converting a series of successful courses into progress through a program. Work, caregiving and financial pressures matter, but so may course availability, program sequencing, compulsory attendance and policies designed around uninterrupted full-time study.

The distinction between course success and continued progression provides a useful way to read the two reports’ proposed responses. Ithaka looks across the existing research for interventions that have demonstrated effects on persistence and completion. The Australian report looks more closely at what happens as students move through courses and programs. The first approach favors larger-scale and more intensive interventions; the second makes visible the less dramatic institutional frictions that can interrupt an otherwise successful journey. What each report sees as a promising solution follows partly from what it recognizes as persuasive evidence.

The evidence used shapes the solution

Ithaka explicitly recognizes that many students cannot attend full time and criticizes the historical tendency to treat increased course-taking as the principal solution. Yet its commitment to the strong, but also the most widely shared causal evidence pulls it back toward that same logic. Its first and most prominent research priority is the adaptation of comprehensive support programs modeled on CUNY ASAP.

ASAP has produced unusually large improvements in three-year graduation, although the recent 14-year evaluation shows that the permanent completion advantage is considerably smaller than the initial difference.

The difficulty is that the best-supported intervention changes the enrollment behavior that defines the population under discussion. It demonstrates that a comprehensive package can help some students attend more intensively and graduate faster. It tells us much less about how institutions should serve students whose employment, caregiving or health circumstances make sustained full-time study impossible.

Even the part-time adaptations retain some emphasis on enrollment intensity. Nashville Flex participants were 23 percentage points more likely to transition to full-time enrollment, although Ithaka stresses that the program is intended to reduce students’ working hours rather than push them into full-time study. Part-Time SAIL requires at least 18 credits annually, offers larger textbook vouchers for heavier loads and encourages summer enrollment to maintain momentum. Ithaka wants to move beyond ‘make them full time,’ but its strongest available evidence still comes from models that help students increase or sustain their enrollment intensity.

Ithaka’s other priorities include flexible scheduling, recognition of prior learning, financial aid, public benefits and stronger connections between education and employment. These address genuine constraints but operate at very different levels. Some require national or state policy changes; others depend on institutions. The result is a broad research and policy agenda rather than a tightly connected account of where progress through a particular program becomes obstructed and what individual institutions can do about it.

The ACSES report also reaches beyond institutions, recommending changes to income support, scholarship taxation and public-transport eligibility. The difference is one of emphasis. Ithaka organizes its agenda around intervention categories and gaps in causal evidence; the ACSES report devotes much more attention to the institutional machinery through which students must move. Its institutional recommendations fall into three broad groups.

Make each course navigable

An emphasis on course-level changes might seem odd after evidence that part-time online students are generally succeeding in the courses they take. But these recommendations are not primarily about changing academic content or providing remediation. The course is also where students encounter the practical organization of their journey: when materials become available, how assessments are spaced, whether attendance is compulsory and what happens when work or caregiving disrupts a deadline. Friction at that level can make continued enrollment difficult even for students capable of passing the course.

Students therefore asked primarily for predictability and consistency. Ninety-three percent valued well-spaced assessments, 90 percent consistent LMS layouts and 88 percent early access to materials. The report also recommends recordings, clear expectations and routine extensions. None is especially novel, but their cumulative importance is easy to underestimate. Together, they reduce the planning burden placed on students attempting to fit study around relatively fixed work and caregiving responsibilities.

Make the program navigable

The ACSES report is more original when it moves from course design to program architecture. Students wanted explicit part-time pathways showing course sequence, prerequisites, future availability and sensible combinations. Across the four universities, the researchers identified 15 two-course combinations with an average success rate of only 53 percent—approximately 20 points below the rate when the same courses were taken in other combinations. The analysis does not establish that the combinations caused the lower success rates, but it gives institutions a clear set of enrollment patterns to investigate. The report also recommends more reliable course availability and realistic maximum completion periods. These are progression problems: a student can pass each available course and still encounter no workable route to the credential.

Remove full-time assumptions from part-time study

The report also asks universities to reconsider the full-time assumptions embedded in programs that permit part-time enrollment. It recommends fewer compulsory synchronous activities, earlier notice of unavoidable requirements, wider examination windows, part-time placements and recognition of relevant employment. An online program offers limited flexibility if students must attend at a single time, travel to campus repeatedly or leave paid employment for a full-time placement.

The ACSES recommendations are practical, but the evidence supporting them is uneven. Students overwhelmingly prefer consistent LMS structures, early materials and recorded classes, but preferences do not demonstrate that these practices improve retention. The scholarship comparison is observational, while the extension case study cannot fully isolate the policy’s effect. The report provides institutions with a credible map of frictions to investigate, not a proven package guaranteed to increase completion.

Some changes will also be difficult to implement consistently. Universities have encouraged common course templates and recorded classes for years, often with uneven adoption. Reliable part-time course sequences and flexible placements require coordination across academic departments, scheduling, advising and external partners. The recommendations are smaller than a comprehensive program such as ASAP, but they are not necessarily simple.

Rigor, relevance and the missing evidence

Read together, the reports expose an awkward problem in the evidence base. The intervention with the most widely-known causal evidence, ASAP, was designed around full-time enrollment. The interventions most closely aligned with sustained part-time study—predictable course design, flexible attendance, better sequencing, realistic completion periods and part-time placements—have much weaker evidence about their effects on completion.

That gap should not become an excuse for institutional inaction. The ACSES report identifies a succession of smaller institutional barriers that can prevent individually successful courses from adding up to a completed credential. Institutions can examine course combinations, course availability, LMS consistency, compulsory attendance and maximum completion periods without waiting for a new national program or a randomized trial. They can also evaluate the changes they make and begin strengthening the evidence the report currently lacks.

These institutional changes will not, by themselves, overcome financial insecurity, caregiving demands or the consequences of accumulating credits slowly. Ithaka appropriately keeps comprehensive support and public policy in view. Institutional friction is only part of the problem, but it is the part over which universities have the most immediate control.

The challenge is to combine Ithaka’s demand for credible evidence with the ACSES report’s attention to the student journey. Helping students pass their next course matters, and helping some take more credits may matter too. But part-time student success ultimately depends on creating a pathway through which the courses students can realistically take eventually add up to a credential, and on developing better evidence about what keeps that pathway open.

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.

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.

Two apparently unrelated things caught my attention this week, but they share an important feature: both concern the institutional machinery that shapes student success while remaining largely invisible to students. The first is the network of people and services required to bring a student and faculty member together in a classroom. The second is the collection of program choices, registration rules and defaults that influence what students study once they arrive.

Neither is normally described as teaching, but both shape students’ opportunities to learn and succeed.

99% Invisible ¹

A few years ago, I hired a plumber to repair something. When he handed me the bill, he also gave me a sheet explaining why his hourly rate was so high. I was paying not only for his time but also for part of the cost of his truck, tools, license, receptionist and the website through which I had found him.

The sheet made visible the infrastructure behind an apparently simple service. A classroom works much the same way.

I thought of that plumber after encountering yet another post attributing higher-education costs and declining confidence to the growth of administrators and professional staff. The underlying staffing trend deserves examination, although this particular comparison counted only full-time faculty and said little about what the expanding categories of employees actually do. References to lazy rivers supplied the usual rhetorical garnish.

Aggregate staffing figures can identify a question, but they cannot answer it. Before deciding that employees outside the classroom represent excess, we need to understand the work required to make the classroom possible.

To that end, I have mapped the institutional functions required to bring a student and faculty member together in a classroom. The diagram does not estimate the number of employees needed to perform those functions, and it deliberately excludes online learning, now a significant part of US higher education. It also omits some functions I consider important, including centers for teaching and learning.

Even this deliberately limited classroom version contains more than 20 distinct functions. They do not correspond to 20 employees for every faculty member: they may be combined, centralized, shared, outsourced or supported by technology. The exercise shows why a simple comparison between faculty and “everyone else” tells us remarkably little about whether an institution is appropriately staffed or operating efficiently.

Some of these functions may be inefficiently organized, and some institutions may employ more people than the work requires. The diagram is not a defense of every position or every increase in staffing. It is an argument for examining functions before declaring entire categories of employees unnecessary. The classroom is the visible encounter between faculty and students. The systems that admit the student, approve and schedule the course, appoint the instructor, maintain the space and support participation are less visible, but no less real. “Not teaching” is not the same as “not contributing to education.”

The quiet power of program defaults

A new report from the Community College Research Center begins with a troubling finding: although more community college students are earning credentials, more than one-third of those credentials do not lead to jobs paying a living wage or provide a viable pathway to a bachelor’s degree.

In 2023, CCRC partnered with the Aspen Institute College Excellence Program to create the Unlocking Opportunity network. Ten pilot colleges sought to increase enrollment and completion in high-value programs while reducing enrollment in programs associated with weak post-completion outcomes.

Aspen and CCRC define value primarily in economic terms, while also considering transfer and bachelor’s degree completion for programs intended to lead to four-year institutions.

The scale of the enrollment shift is striking, particularly among programs intended to lead directly to employment.

Producing these enrollment shifts required substantial work.

Each college approached the challenge somewhat differently. But across the ten pilot colleges the approaches featured the following actions.

Changing the choice architecture

• Flagging students without a declared major

• Removing General Studies as an easy default

• Strengthening career and program advising

Changing the program portfolio

• Redesigning programs with employers to make them higher value, for example, by adding drone technology to an agronomy program

• Expanding high-value programs

• Creating alternative pathways for students unable to enter capacity-constrained programs

Expanding the capacity to deliver

• Adding clinical placements

• Hiring placement coordinators and other necessary staff

• Reallocating resources toward programs the colleges wanted to grow

The changes to program portfolios and the architecture of student choice are particularly interesting—and underappreciated—as potential student-success interventions. How much apparent student choice is actually produced by application design, program labels, advising rules and institutional defaults?

Students still make choices, but institutions determine which choices appear first, which require additional effort and which are constrained by available capacity. “General Studies” can function as a neutral holding category, but it can also become a path of least resistance that postpones consequential decisions. Removing that default may prompt better planning; it may also steer students more assertively toward institutionally preferred programs. Whether that is beneficial depends on the quality of the advice, the suitability of the alternatives and the students’ own objectives.

These practices are promising and provide a welcome piece of positive news. But we should be precise about what they have achieved and what they represent in terms of student success.

Enrollment in a high-value program is an important intermediate result, but it is not yet a student-success outcome. The report demonstrates changes in program enrollment, not changes in completion, earnings or transfer success. Nor do we yet know the full cost of expanding high-value programs, many of which require specialized faculty, equipment or clinical placements. Finally, economic value is important, but it is not the only legitimate consideration in a student’s program choice.

Nevertheless, this is unusually encouraging work. The colleges did not merely tell students to make better choices. They changed the programs available, the defaults students encountered, the information informing their decisions and the capacity of high-value pathways.

That connects this report to the apparently unrelated question of university staffing. Student success is shaped by a great deal of work that occurs outside the immediate exchange between student and faculty member: designing programs, building systems, arranging placements, advising students and maintaining the infrastructure through which choices become possible. The useful question is not whether that work occurs inside or outside the classroom. It is whether the work is well designed, efficiently performed and demonstrably valuable to students.

Musical coda

My spouse and I have been assiduously working our way through lists of the top pop and rock songs for each year, beginning in 1965. I had forgotten how much I enjoy Creedence Clearwater Revival.

The default option should be to forward this post to someone who might find it interesting. Please don’t let it remain 99% invisible.

Thanks for being a subscriber.

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.

Higher education has spent decades treating enrollment growth as evidence of institutional success. Presidents are praised for adding students, programs, buildings, and geographic reach. “If we’re not growing, we’re dying,” one university president told researchers from the National Center for Higher Education Management Systems (NCHEMS).

But what happens when the number of prospective students stops growing? In a stagnant or shrinking market, institutions cannot all maintain their enrollment by capturing a larger share. Growth becomes increasingly redistributive: one institution’s gain is more likely to produce losses elsewhere.

Two recent reports—one on American public higher education from NCHEMS and another on English universities from the Higher Education Policy Institute (HEPI)—show what this looks like in practice. Higher education systems are, like it or not, stratified by prestige, selectivity, wealth, and recruiting power. That is a description of institutional position, not a judgment about educational quality or public value.

Better-resourced institutions expand into markets traditionally served by other colleges and universities. Institutions with less power in the hierarchy respond by adding programs, recruiting new student populations, or moving into the territory of still more vulnerable institutions.

Each decision may be rational for the institution that makes it. Collectively, however, these decisions can leave higher education systems more hierarchical, less stable, and less capable of serving students who depend on affordable and geographically accessible options.

The US: Enrollment moves up the hierarchy

NCHEMS released a policy brief with the Lumina Foundation describing how enrollment changes in the United States have shifted the roles played by different kinds of public institutions.

NCHEMS argues that the long period of enrollment growth allowed states to rely on competition and student choice to shape public higher education. When there were enough students to go around, expansion at one institution did not necessarily threaten others.

That assumption no longer holds. Public research universities generally possess greater visibility, stronger finances, and a wider recruiting reach than regional universities and community colleges. As demand softens, these advantages allow them to capture a growing share of students and public resources.

We see this most clearly in undergraduate enrollment. Enrollment at public research universities grew in 36 states—but in only six of those states did the public comprehensive sector also grow. Utah was the only state in which enrollment at comprehensive universities grew more than enrollment at research universities.

This was not, therefore, a shift confined to a few unusually competitive state systems. It was a widespread movement of enrollment toward public research universities.

Some of this change reflects the reclassification of 54 institutions from the comprehensive to the research category between 2013 and 2023. But NCHEMS argues that movement between the categories is itself evidence of mission convergence: institutions are not simply responding to enrollment pressures but also repositioning themselves within the hierarchy.

The problem is not that public research universities are growing or that students should be prevented from choosing them. It is that few states have mechanisms for determining whether that growth advances public priorities or strips enrollment and capacity from institutions with different but equally important missions.

Institutions facing these pressures seek new students and sources of revenue. Regional universities expand graduate education, recruit farther from home, or pursue research status.

Program portfolios are also being remade. Institutions across most sectors are simultaneously adding and eliminating large numbers of programs. Some of this activity reflects legitimate changes in student and workforce demand. But creating programs is also a strategy for attracting new student populations and entering markets historically served by other institutions.

The chart’s most striking feature is not simply the number of new programs but the scale of the churn. Institutions are committing resources to new offerings while eliminating others, even though research cited by NCHEMS suggests that relatively few newly launched programs become sustainable over the long term.

Taken together across a state, these individual program decisions can substantially alter institutional missions without anyone first determining whether the resulting mix of programs effectively serves students, communities, or workforce needs. Student choice may determine which institutions grow, but it cannot by itself ensure geographic access, an appropriate distribution of workforce programs, or adequate educational opportunities for populations whose participation in college is least assured.

The UK: England’s recruitment cascade

England provides a clearer view of how enrollment redistribution works and shows that it can begin even before demographic decline arrives. England’s population of 18-year-olds is expected to grow modestly until around 2030 and then fall substantially during the following decade. Yet institutions lower in the hierarchy are already losing ground.

A HEPI report, which I previously covered in On EdTech, shows that between 2016 and 2025, acceptances at higher-tariff universities—essentially, more selective institutions—grew by approximately 27 percent, compared with growth of about 7 percent across the sector. HEPI estimates that roughly 70 percent of the growth at higher-tariff institutions resulted not from expansion of the overall system but from the redistribution of students who, under earlier enrollment patterns, would probably have attended other universities.

This produces a cascade. Higher-tariff universities recruit further into the attainment range historically served by mid-tariff institutions. Those institutions respond by reaching further into the pool traditionally served by lower-tariff universities. Institutions at the bottom of the hierarchy experience the cumulative losses.

Demography helps determine the overall size of the student pool, but it does not determine which institutions bear the losses. Institutional power, recruitment behavior, and public policy help determine how those losses are distributed.

What does redistribution mean for student success?

Both reports recognize that system-level dynamics can affect students, but neither follows the argument very far into students’ experiences and outcomes. Neither provides direct evidence that redistribution has already worsened access, affordability, support, or completion. The implications discussed below should therefore be understood as risks that policymakers and institutions need to investigate, not as conclusively established effects.

What might happen when students and resources move up the institutional hierarchy?

Access is about more than the number of available seats

Institutions are not interchangeable suppliers of college places. NCHEMS finds that community colleges and regional universities draw far more heavily from nearby populations than research universities do. The median student travels 10 miles to attend a community college and 13 miles to attend a public bachelor’s or master’s institution, compared with 39 miles to attend a public research university.

Many community college students face constraints related to employment, income, caregiving, and transportation. If a rural or small-town community college weakens or closes, growth at a flagship hundreds of miles away does not replace the opportunity that was lost. A technically available college place is not meaningfully accessible if a student cannot afford to relocate, leave a job, or set aside family responsibilities.

Redistribution up the institutional hierarchy may also direct students toward more expensive options. Even when financial aid offsets some tuition differences, relocation, housing, and forgone earnings can make the total cost substantially higher, limiting access or imposing heavier debt burdens on students.

The support squeeze at both ends

Greater access to a more selective institution can benefit students, and admitting a broader range of applicants should not automatically be characterized as predatory. But recruiting students is not the same as supporting them through completion.

If an institution rapidly recruits students whose academic, financial, employment, or family circumstances differ from those of the populations it has traditionally served, its advising, scheduling, financial aid, and academic support systems must change accordingly. Without that investment, expanded access can create a gap between the students an institution admits and those it is designed to support.

At the same time, institutions that lose students also lose tuition revenue. They may reduce course sections, advising, tutoring, and student services precisely when their remaining students have substantial support needs. Redistribution therefore creates student-success risks at both ends: capacity strain at growing institutions and resource erosion at shrinking ones.

Program churn can create student risk

Institutions under pressure frequently respond by creating new programs and entering new markets. But expansion is not free. New programs consume startup funding, marketing capacity, faculty time, and leadership attention before they generate revenue—and some will never enroll enough students to cover their costs.

This can create a destructive cycle: enrollment loss produces a search for growth; the search for growth diverts resources from existing students; weaker services make retention more difficult; and declining retention adds to the original financial pressure.

Program churn also creates direct risks for students. When institutions close programs, students can face discontinued course sequences, changes in degree requirements, credits that no longer apply, delayed graduation, or pressure to transfer. Institutions may provide formal teach-out plans while still leaving students with fewer course options and less predictable routes to completion.

The student-success question is therefore not only whether a new program attracts students. It is also whether the institution can sustain that program long enough for its students to complete it—and protect them if the program closes.

Conventional success metrics can conceal system failure

Redistribution can also make institutional metrics misleading. Suppose students with stronger academic preparation move from regional universities to a flagship university. The flagship’s completion rate may remain high. But the regional university loses both enrollment and revenue while continuing to serve students with greater support needs, and its completion rate may fall.

Those numbers might appear to show that students have chosen the more successful institution. But some of the difference may reflect student sorting and resource redistribution rather than changes in educational quality.

We measure success institution by institution, even when the forces shaping it operate across an entire system.

Two reports, two policy responses

If redistribution threatens access, student support, and program continuity, allowing every institution to pursue growth independently is not an adequate response. NCHEMS and HEPI both argue for greater system-level intervention, but they differ over how far that intervention should go.

NCHEMS calls for states to manage the missions of public institutions more deliberately:

HEPI goes further, recommending institution-level recruitment caps intended to prevent higher-tariff universities from protecting their enrollment at the expense of the rest of the sector.

That proposal makes the underlying trade-off unusually visible. Protecting institutional diversity may require limiting some forms of expansion. But doing so could also restrict opportunities for students who would benefit from attending a more selective university.

A system problem needs a system response

Neither unrestricted competition nor rigid enrollment caps provide a satisfactory answer. The first allows individually rational decisions to produce damaging collective effects. The second risks becoming paternalistic—protecting institutions from competition, restricting student choice, and assuming that policymakers can determine in advance where students should enroll.

Individual institutions nevertheless have responsibilities. Leaders and governing boards should understand how their missions are changing, whether expansion is based on genuine new demand or the redistribution of existing students, and whether their support systems can serve the populations they are recruiting. Institutions losing enrollment must also examine how contraction, program closures, and service reductions affect the students who remain.

But institutional action is unlikely to be enough. Institutions are being asked to compete for students and revenue, and it is unrealistic to expect them to voluntarily sacrifice their own interests to preserve the health of the wider system. The central problem described by NCHEMS and HEPI is created by the interaction of institutional decisions, not simply by the failure of individual institutions to make better ones.

Nor is the problem necessarily a lack of policy tools. States and systems already conduct mission reviews, approve programs, develop strategic plans, collect extensive data, and make funding decisions. The problem is that these activities are often fragmented, weakly enforced, or focused on individual proposals rather than their cumulative effects. A new program may make sense for one institution while contributing to unnecessary duplication across a state. Enrollment growth may strengthen one university while undermining access and capacity elsewhere.

A system response does not require freezing institutional missions or assigning students to particular colleges. It does require making changes in enrollment, programs, and missions more explicit; evaluating their combined effects on students and communities; and judging institutional growth by more than the number of students recruited. Institutions pursuing new populations should be expected to demonstrate that they can support those students, while systems should pay closer attention to what is lost when other institutions weaken.

Meaningful student choice requires more than allowing students to apply anywhere. It also requires preserving viable, affordable options for students who cannot relocate or absorb higher costs. Governments cannot treat every enrollment decision as an isolated transaction and then act surprised when the combined result undermines access, institutional stability, and student success.

The answer is not to preserve every institution in its current form. It is to recognize that colleges and universities perform different public functions and that market competition alone will not necessarily preserve the functions students and communities need.

The enrollment cliff will not simply reduce the number of students in higher education. It will redistribute students, resources, and educational opportunity. Whether that redistribution advances student success—or leaves the most vulnerable students with fewer viable choices—will depend on whether policymakers continue to let the market drift or begin treating the system as a system.

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.

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.

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.

The Postsecondary Commission recently published a fascinating report on measuring the value-added of higher education credentials at scale. Most headlines focused on a single message: college pays. The Arizona Board of Regents, for example, summarized the report by claiming that "no matter your major, earning a college degree pays off." Yet that is almost the opposite of what makes the report important. Its contribution is not simply demonstrating that higher education generates positive economic returns. Rather, it demonstrates how misleading it is to reduce educational value to a single earnings figure. Once educational costs, opportunity costs, student characteristics, and non-completion are incorporated, the real questions become where students study, what they study, whether they complete, and how long it takes for returns to emerge.

So what did the report actually say?

The method changes the story

The report's most significant innovation is methodological. Rather than comparing graduates with non-graduates using earnings alone (as many previous studies do), the researchers, in collaboration with Mathematica, estimate Value-Added Earnings (VAE) from the point of enrollment. This approach allows them to account for tuition costs, foregone earnings while studying (opportunity costs), students who never complete their credential, and compare participants with similar individuals who did not attend college. In doing so, the report measures the economic value of higher education as an investment rather than simply comparing graduate salaries.

This methodological shift fundamentally changes the question being asked. Instead of asking whether graduates earn more than non-graduates, it asks whether enrolling in higher education represents a worthwhile investment once educational costs, opportunity costs, student characteristics, and non-completion are taken into account.

The methodology differs from previous studies in three important ways:

• It includes all entrants, not just those who complete a credential.

• It measures outcomes from the point of enrollment, allowing educational costs and foregone earnings to be incorporated.

• It compares statistically matched groups, balancing demographics, household income, and prior academic achievement to isolate the effect of postsecondary education on earnings.

Every credential pays off - but not equally

The report shows that all three types of credential generate a positive economic return. However, both the time it takes for that return to become positive and the magnitude of the return vary considerably by credential type.

At first glance, the findings are reassuring: every type of credential generates a positive value-added return over time. But averages conceal enormous variation. Once the data are dis-aggregated, institution, program of study, and completion emerge as critical determinants of educational value.

Program choice matters more than institution

Conventional discussions of educational quality tend to focus on institutional prestige and rankings. The Texas evidence suggests that program choice has a greater influence on economic returns than institutional choice.

Institution still matters. For all credential types, the institution attended influences the value-added return, and in some cases students experience a negative return overall.

Among bachelor's degree providers, value-added returns vary substantially across institutions. Two public bachelor's degree institutions in Texas produced a negative value-added return, meaning students who enrolled would, on average, have been financially better off had they not enrolled.

Institution also matters for associate degrees and certificates. Among associate degree providers, returns again vary across institutions, with one institution producing a negative value-added return.

Variation is even more striking among certificate providers. Students attending 14 institutions did not experience a positive value-added return. Although returns vary less than for bachelor's degrees—likely reflecting the lower cost of certificate programs—the differences between institutions remain substantial.

But major has an even bigger impact.

For all credential types, program of study has a substantial influence on value-added returns. Among bachelor's degrees, the choice of major affects the overall level of return, with a wide spread across disciplines. Engineering programs generate some of the highest value-added returns, while fields such as the liberal arts generate much lower returns. Even so, the average value-added return remains positive across all majors.

For associate degrees and certificates, however, some fields of study produced a negative value-added return on average.

Comparing bachelor's and associate degree programs side by side illustrates the variation across both institutions and fields of study. The trend lines, rather than the individual data points, tell the story.

This finding is significant because discussions of educational quality have long emphasized institutional prestige, reinforced by rankings. The Texas evidence suggests that field of study has a much greater influence on economic returns than institutional choice.

There is, however, an important limitation. The analysis includes only public institutions in Texas. Had the study also included private institutions—including highly selective universities such as Rice, smaller private colleges, and for-profit institutions—it is possible that the variation across institutions would have been even greater. Whether that would alter the broader conclusion that field of study matters more than institution remains an open question.

Ignoring non-completers distorts the results

The report makes a compelling case for the importance of completion. Most studies examining the economic impact of postsecondary credentials exclude students who never complete their program. Doing so introduces a significant source of bias. Looking only at those who complete a credential is naturally more likely to produce larger and faster economic returns than following all students from the point of enrollment.

This is particularly problematic given the scale of non-completion in U.S. higher education.

Excluding such a large proportion of students introduces substantial measurement bias, and this study directly addresses that problem.

The report also helps quantify the value of completion. It is hardly surprising that students who complete their credential experience greater economic returns, but the report shows just how large that effect is. For bachelor's degrees, for example, each one-percentage-point increase in the completion rate is associated with an additional $2,000 in value-added earnings.

This positive relationship is consistent across bachelor's, associate's, and certificate programs. Institutions with higher completion rates tend to generate greater long-term economic returns for their students.

Finally, accountability systems based solely on graduate outcomes—that is, on students who complete a credential—are likely to overestimate institutional performance. By excluding non-completers, they omit precisely those students who incur the costs of higher education without realizing the economic benefits of a credential.

Returns take time

One of the report's most striking findings is how long it takes for postsecondary education to generate positive economic returns. Students pursuing bachelor's degrees do not break even until nine years after enrollment and do not exceed $80,000 in value-added earnings until 15 years after enrollment. The break-even point is seven years for associate degrees and four years for certificate programs.

These findings add to the growing body of evidence questioning accountability systems that assess institutions too soon after students leave college. For example, the accountability provisions in the One Big Beautiful Bill Act (OBBB) evaluate graduate earnings just four years after graduation. The Texas evidence suggests that this is far too early to assess the long-term economic value of many higher education credentials.

Economic context shapes returns

The report also hints at the influence of broader macroeconomic conditions on educational returns. For example, value-added earnings for bachelor's degree students vary by year of entry. Students who enrolled in 2008–09, at the height of the Great Recession, experienced substantially higher long-term returns than those who enrolled in other years. While the report does not explore this relationship in detail, the pattern suggests that the economic context in which students begin their studies may influence the long-term value of their credential.

One plausible explanation is that the opportunity cost of attending college was lower during the recession because unemployment was high and wages were depressed. Students who enrolled during this period also entered the labor market as the economy recovered, potentially allowing them to benefit from improving economic conditions.

Although the report does not explore this issue in depth, it highlights an important consideration. The economic returns to higher education are shaped not only by the credential itself but also by the broader economic context in which students enroll and graduate. That context should play a larger role in discussions of educational value and in the design of accountability systems.

The student matters, too

The researchers also uncover several interesting patterns in the individual characteristics associated with value-added returns. For example, at the bachelor's and associate degree levels, students with stronger high school math achievement tended to realize higher cumulative value-added returns. This is unsurprising, as these students are more likely to enroll in fields of study associated with higher economic returns.

More surprising is the relationship with household income. For bachelor's degrees, value-added returns vary little by household income, suggesting that the long-term economic benefits of a bachelor's degree are relatively consistent across income groups.

The picture is different for shorter credentials. Household income does influence value-added returns for associate degrees. For certificate programs, the relationship is actually reversed: students from lower-income households tend to realize higher value-added returns than those from higher-income households.

Taken together, these findings challenge the way higher education is often discussed. Public debate tends to ask whether college "pays," assuming that a single average figure can capture educational value. The Postsecondary Commission instead demonstrates that value is conditional. It depends on where students enroll, what they study, whether they complete their credential, when they enter the labor market, and the broader economic context. Measuring educational value therefore requires moving beyond averages toward a richer understanding of variation.

Why this matters

The Postsecondary Commission's report does more than confirm that higher education generally produces positive economic returns. Its real contribution is methodological. By measuring value from the point of enrollment rather than graduation, incorporating educational costs, foregone earnings, and non-completion, it reveals a much more nuanced picture of educational value than conventional earnings studies.

That richer picture has important implications. It suggests that debates about institutional quality should pay greater attention to fields of study; that completion should be treated as central, rather than peripheral, to measures of educational value; and that accountability systems built around early graduate earnings risk producing misleading conclusions. Rather than asking simply whether college "pays," policymakers should instead ask a more useful question: Under what conditions does higher education create value, for whom, and over what time horizon?

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.

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.

It’s still hot, the hills around me are still burning and I am suffering from cricket withdrawal. But what is happening with student success?

This week everything I read seemed to be about misaligned assumptions.

The confidence gap

One of my mid-year resolutions is to stop writing about what employers want in graduates.

Not because the question isn't important, but because it has become too dominant. It reduces higher education to workforce preparation, assumes employers always know what they need (they often don't), and too easily shifts responsibility for every perceived skills gap onto colleges and universities.

That said, a new report by Pearson and AWS highlights some interesting misalignments between higher education leaders, learners (hereafter referred to as students), and employers.

The biggest misalignment for me concerns the perceived value of higher education. Students are the least convinced that university education is becoming more important in the age of AI. They were less likely than either employers or higher education leaders to say that university education is becoming more essential, and more likely to believe it is becoming less essential.

The chart below is worth a closer look. The gap isn't huge—62% versus 67%—but it's psychologically fascinating.

The people universities are trying to convince are the least convinced. Meanwhile, employers—the people doing the hiring—are the most convinced. That suggests a messaging failure.

The UK results are particularly striking, where students are substantially more likely to believe that higher education has become less important in the age of AI.

A second interesting finding concerns hybrid skills, that is, the combination of technical and human skills. Higher education leaders may be underestimating the importance of technical skills slightly, but once again it is students who are the most interesting group.

Universities spend enormous amounts of time trying to understand what employers want. Perhaps the more urgent question is whether students understand what universities actually offer. If students are less convinced of the value of higher education than employers themselves, we don't have an employability problem; we have a confidence problem.

Revisiting my mid-year resolution, I don't want to stop writing about the graduate labor market or the misalignment between higher education and employers. What I do want to stop is treating employers as the sole authority on educational value. That already happens far too often in the literature.

The myth of independent job applications

Graduates finding employment is an increasingly important aspect of student success. But we tend to assume two things about job applications. First, that every employer evaluates applicants independently. Second, that sending more applications improves your chances. A fascinating new article suggests both assumptions may be wrong.

Using data from more than four million job applications, researchers examined hiring decisions made through a widely used AI screening platform. Most research on AI in hiring asks a straightforward question: Is the algorithm biased? This study asks a different one: What happens when hundreds of employers rely on the same algorithm?

Rather than looking only at aggregate results, the researchers examined hiring decisions job by job. They found that around 10% of jobs showed practically and statistically significant disparities in selection rates for Black applicants. Although employers used different hiring models, many were built on the same underlying platform. As a result, applicants were more likely to receive similar decisions across employers.

The consequence is striking. The researchers estimate that applicants may need to submit roughly 25 applications instead of 10 to achieve the same chance of receiving at least one recommendation. The reason is simple: many hiring decisions are no longer fully independent.

That's a profound shift in how we think about job searching. The challenge isn't just competing for jobs—it's navigating an increasingly interconnected ecosystem of AI-mediated hiring.

There are important limitations. The study examines one vendor that uses game-based assessments rather than the broader universe of AI résumé screening tools, so we shouldn't assume the same pattern exists everywhere. But it does raise an important question: what happens as more employers adopt increasingly similar hiring technology?

Every application feels like a new opportunity. Increasingly, it may simply be another opinion from the same AI. As AI becomes more deeply embedded in hiring, we may need to rethink some of our oldest career advice. Just keep applying assumes every application is a fresh opportunity. Increasingly, that may no longer be true.

Overlaps

I love a good Venn diagram. This one makes me chuckle even though the British now swear loyalty to a King not a Queen.

Though I feel obliged to point out that historically the British took things from the colony not to it.

When saying less helps more

I loved this post about the practice of purposeful restraint in coaching. Whether we're coaching, advising, teaching, managing, or mentoring, it's often tempting to jump in with feedback to help a player, student, mentee, or employee avoid making the same mistake.

The secret to all this is timing, understanding the player or student or employee and in having their trust.

Which makes it sound hard—and it is. It's also easy to mistake passivity for purposeful restraint in any of the roles I've mentioned.

To me, there's an obvious student success lesson here. Student success isn't just about providing support; it's about providing support when students are ready to use it. Timing may be just as important as the intervention itself.

Kentucky - the grass is always more blue

Some good news from Kentucky

Musical Coda

Staying in Kentucky.

Know someone whose assumptions could use a little realignment? Forward them this newsletter.

Thanks for being a subscriber.

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.

What happened and what did I read this week on student success?

This week offered four reminders that choosing the wrong measure, or measuring the wrong thing entirely, can easily lead us to the wrong conclusions.

The part-time comeback

Buried deep in the latest National Student Clearinghouse report is a surprisingly hopeful trend. Persistence and retention for part-time students have improved year-over year after almost a decade of decline.

But across all institution types, persistence and retention are noticeably up for part-time students for all institution types but especially for 4-year institutions of all types.

4-Year public institutions saw gains this year of 3.2% and 2.9% for second fall persistence and retention.

4-Year private institution part-time persistence and enrollment rates were up a similar amount. Part-time persistence and retention rates even improved year-over-year at 4-year for-profit institutions, albeit from a lower base level.

By contrast the uptick in persistence and retention for students at public 2-year institutions was more modest.

In all institution types these upticks in persistence and retention rates for part-time students reverse a long trend of declining rates since 2015. For years the story of part-time students has been one of slow decline. This is the first evidence that story may finally be changing.

Accidental pedagogy

Most OER debates ask a simple question: do open textbooks impact student finances? A new AAC&U study asks a much more interesting one: under what circumstances do they improve student success?

It is a really good study, based on a huge dataset and using mixed methods. The authors conclude OER should qualify as a new High Impact Practice. I came away convinced of something different: the real intervention isn't open textbooks, it's better teaching.

OERs and student success

Initially they seem to prove their point. They find that OERs are correlated with student success measured in terms of withdrawal rates which fall from 8.6% to 7.1% with the use of OERs. They examine these in terms of institution type but also learner type, using the concept they call Composite Learner Complexity (CLC)—a measure incorporating part-time status, Pell eligibility, first-generation status, age over 25, and race/ethnicity.

There is no single OER effect. It varies substantially by institution type.

Implementation beats adoption

The heart of the report is about implementation. To measure this, the researchers separated implementation into three basic modes: no OER, simple adoption and more complex adoption and creation where OERs were revised, remixed, or created by the instructor.

Higher-level implementation generally produces better outcomes than merely swapping textbooks, particularly in community colleges and regional public universities.

Unfortunately the report does not always fully explore the interesting issues that the chart raises.

Doctoral universities. There is a perverse result where higher-level implementation of OERs leads to higher withdrawal rates than more simple implementations. They explain this as a consequence of a poor sample and an especially tough class. Highlights mine

Once the authors realized a single gateway mathematics course dominated the higher-level implementation category, I would have expected them to rerun the analysis without it. Without that, the doctoral findings are difficult to interpret.

Regional public universities. Adopting an OER textbook increased withdrawal rates at least for low complexity learners. The researchers dug into the data and found that the courses where withdrawal rates went up were predominantly non-STEM courses. They speculate that this is because of cost and because STEM textbooks are generally higher cost than non-STEM textbooks.

This is pure speculation and suggests to me that the authors have not recently for example been in the business section of a college bookstore.

The wrong intervention

These issues and frustrations aside, the findings that in general more complex implementation leads to lower withdrawal rates at least in community colleges and regional public universities suggests to me that something else is going on rather than OERs. Different pedagogy appears to lead to lower withdrawal rates.

The authors describe this process themselves. Faculty with more complex implementations of OER made many more changes to their courses than faculty who simply adopted an OER textbook.

Better pedagogy, better outcomes. You could knock me over with a feather. What if OER isn't the intervention at all? What if OERs are a form of accidental pedagogy, simply prompting instructors to redesign their courses? The real high-impact practice may not be open textbooks. The real HIP may be course redesign.

Wirecutter headlines from the revolutionary war

On Saturday it is of course American Independence Day. To celebrate the occasion, McSweeneys wrote a list of Wirecutter headlines from the American Revolutionary War. For the uninitiated, Wirecutter is a New York Times service that has an often cult-like following. They rate and review consumer items using a very distinctive vocabulary. In my household, we aren’t allowed to buy anything without consulting it.

Here are my favorites from their list.

Beyond the player with the ball

While I must confess that I haven’t actually watched any of the World Cup football (all my sports time has been taken up by cricket), I have been fascinated to read about the use of data and analytics by football teams in the World Cup. Football as a sport has for a long time relied on data to improve team tactics and play (and there are some great books about this, far more interesting than Moneyball IMHO).

Footballers regularly wear trackers, there is a tracker in the ball and of course cameras tracking players, leading to all kinds of insights. But use of data has reached some new heights during this World Cup in part due to FIFA making a meaningful amount of that data public through the Enhanced Football Intelligence (EFI) platform. This public data layer gives all teams (and others) access to the data as well as the ability to query it directly rather than relying on experts.

The data goes beyond traditional stats and instead uses the numbers to create visualizations of how teams play. The system combines event data with tracking data from all 22 players, allowing analysts to understand the shape of the game rather than simply what happened to the player with the ball.

There are many lessons this use of data has for student success, but for now I want to focus on three.

Lesson 1: Context matters more than individuals.

Student success data still focuses overwhelmingly on the student. Football increasingly focuses on the interactions surrounding the player. A broader set of data about what is happening outside of the learner seems a fruitful avenue to explore.

Lesson 2: Flow matters more than snapshots.

Football analytics is shifting from counting events to understanding systems. Again this seems like a great evolution for student success data to make.

Lesson 3: Democratization has hidden labor.

Many of the accounts of World Cup data use make the argument that the public availability of the data has enabled a democratization of analytics in soccer, making it available and easily understandable. It turns out that a lot depends on your definition of democracy.

While the sensors produce millions of data points, real people are often required to annotate that data before it can be useful. This work often happens in developing countries.

We hear a lot of claims about data democratization in student success. It too is often more complex than it seems.

What gets measured gets exaggerated

Every few weeks another article appears claiming that students should skip university because some non-degree jobs can pay six figures. But the arguments about salaries in non-degree fields often obscure the fact that there is a wide variance in salaries, and most non-degree job earnings cluster at the lower end of the spectrum.

A post in Chalkbeat analyzed salaries for jobs where workers typically don’t have a degree (or low degree attainment jobs), those where roughly half of workers hold degrees (medium degree attainment) and where a majority of workers are credentialed (high degree attainment).

This is a useful corrective to the kind of overblown claims we sometimes see such as the one they quote from the New York Times “Job demand in fields like construction, along with the allure of potential six-figure salaries, have some high schools investing in hands-on classes that are redefining what success looks like for the Class of 2026.”

These discussions also tend to gloss over how physically demanding and sometimes dangerous many of these occupations are.

The analysis also lumps together occupations that probably shouldn't be treated as a single category. The post mentions some of this passing but never digs into the fact that child care workers and home health aides (which are usually the lowest paid non-degree workers) are different from skilled trades. The latter require extensive training and often a period of apprenticeship while the former (unfortunately) do not. Salaries reflect these differences. A good bit of this stems from the fact that they are looking only at whether a degree is required or not. Looking at actual occupations would be much more useful as these dictate salaries considerably more than the simple binary of whether a degree is required or not.

Whether a job requires a degree turns out to be a much weaker predictor of earnings than the occupation itself.

Musical coda

I saw the Australian band the Audreys in a very small venue once in 2007. The lead singer noticed a song had made me sad, reached out and squeezed my knee with her hand and reassured me it was all going to be OK. True story. They are also amazing.

The population of people who might enjoy this newsletter is almost certainly immeasurable. Rather than attempting a full census, I'll settle for a small random sample. Please forward it to a colleague who likes thinking about higher education.

Thanks for being a subscriber.

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.

The state and the city where I live seem to be on fire. This picture was taken at noon.

Picture by glycineglutamate on Reddit

Among other bad consequences of this the fires are driving more rattlesnakes down into living areas.

This seems a good time to distract myself with some charts.

The forgotten success story

The progress the US has made in increasing post-secondary attainment is extraordinary, and something that I know I at least sometimes forget.

In 1940 an incredibly small proportion of students - 5% - attained bachelor’s degrees. That percentage varied quite a lot by state.

By 2025 that percentage had improved considerably, although the state variation remained large.

Rates increased rapidly between 1940 and 2025 though at different rates by ethnic group.

It is easy to focus on the persistence of attainment gaps. Those gaps matter. But it is also worth remembering that mass higher education is a relatively recent invention. The United States went from a society where a bachelor's degree was genuinely rare to one where it is increasingly the norm.

When navigation becomes inequality

A fascinating post from HEPI looks at how the university choices of UK students from different family income quintiles are affected by clearing. The author makes the argument that the centralized application process (UCAS) is not just sorting students. It is rewarding students who can better interpret and navigate the system. For my US readers this represents a centralized application process. A student’s Firm choice represents their own top selection, an Insurance choice is a backup selection (or safety school) and Clearing is the UK's late-stage admissions marketplace where institutions fill remaining places.

It relies heavily on the concept of matching—whether the grades of the student (in this case A’ Levels) are similar to others who have been made offers by the same institution.

The two most disadvantaged groups of students start out being slightly over-matched but as they go through the process become worse and worse off until by clearing they are being seriously under-matched. The two most advantaged groups actually do slightly better from Insurance to clearing.

This is another example of a theme I keep returning to. Opportunity is not enough. Students also need the ability to interpret opportunity. The same admissions system appears to produce different outcomes depending on how well students understand and navigate it.

We are all getting older

The decline in the male labor force participation in the US is not as bad as we might think. Data from AIBM shows that some of the decline in labor force participation is due to population ageing. Controlling for that means that men’s labor force participation rate declined from 75% in 2000 to 73% today. Not controlling for age shows a decline from 75% to 68%. Pay attention to the pesky details.

Aging explains much of the decline, but not all of it. That distinction matters because the policy solutions are very different.

The Nigerian football engine

Readers have complained that I don’t have enough World Cup soccerball coverage

Okay, maybe only Neil.

“Immigrants: We get the job done!”

I’d love to come up with a way to similarly show the impact of immigrants on their home economies.

Escaping mum and dad

Applicants to higher education in the UK are being driven more by a need to live apart from their families compared to the recent past.

China’s hidden research engine

Discussions of Chinese research often focus on universities and government labs. But the chart shows that enterprise R&D now dominates the system. The state is still heavily involved, but increasingly through incentives, tax policy, and strategic direction rather than direct spending.

That said, there is still a lot of government involvement in the form of tax incentives etc.

The inexperienced need not apply

A chart from the Strada Institute for the Future of Work shows employers ranking of recent college graduates.

Employers routinely say they want graduates with experience. The question is whether employers themselves are creating enough opportunities for students to get that experience. That would be a much smaller chart.

Education and happiness

Higher education doesn’t solve everything and outcomes vary a lot by major but having a bachelor’s degree leads to far higher perceptions of financial well-being.

Based on my own experience I suspect that having a PhD puts you below the 40% mark :-)

Looking back at these charts, what strikes me is how often the most interesting story is hiding underneath the obvious one.

Chaser: It would have been the best day of mine too, TBH

From New Yorker cartoonist, Emily Bernstein

If you enjoyed this post I have two things to ask. One, send it to everyone you know who may also enjoy it or find it useful. Two, subscribe to Adam Tooze’s Chartbook, from whom I nicked the inspiration and a couple of charts.

Thanks for being a subscriber.

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.

It has been a minute since I last posted. What have I been reading, and what has happened in student success in the interim?

I've been on a translation kick recently, and a lot of my reading this week confirmed my thinking on the issue. It is not so much that we have a lack of information, but rather that there is a gap between what institutions think they are providing and what students perceive they are receiving.

Do students see the things that faculty believe that they are doing?

As much as it might portray me as a geek, I must confess to experiencing a frisson of excitement every time I see a new Tyton Partners survey report land in my inbox. Yes, they cover topics that are right in my wheelhouse. But what I really love is that they usually include student, faculty, and administrator perspectives on the same topic and have a knack for presenting the data in ways that highlight internal contradictions.

The latest Time for Class report is no exception. The report focuses heavily on AI and especially on the ways it has created challenges for faculty and forced some of them to rethink how they teach and, in particular, how they design and run assessments.

But for me, the most interesting thing in the report may have nothing to do with AI at all. Instead, I found myself fascinated by the disjuncture (Phil Hill argues that I am disproportionately attached to that word. I'm not. I just keep seeing them everywhere.) between what faculty say they are doing and what they are actually doing—or at least how those practices are being experienced by students.

Take this question about what practices faculty report using often or in every class. It sounds like someone swallowed the training schedule for the Center for Teaching Excellence.

I am sorry, but I am really not buying that 80-odd percent of faculty are using rubrics and active learning. Not at all, let alone often or in every class. Maybe 56% are using a dashboard if you count a quick glance at the LMS, but I'm not sure that really counts.

To me, this chart paints an overly sanguine picture of innovative teaching. Contrast it, for example, with the answer to this question about the major instructional challenges identified by faculty.

If most instructors are routinely setting clear expectations, using active learning, providing regular feedback, assigning authentic work, and designing courses around student engagement, why are cheating and attendance still the two biggest challenges?

That does not necessarily mean faculty are being dishonest. It may mean that faculty and students define these practices differently. It may mean that instructors are using them occasionally rather than consistently. Or it may mean that many of these practices are being implemented in ways that students do not experience as particularly meaningful. Whatever the explanation, the two charts sit uneasily beside one another.

But a little side comment attached to a later discussion of workforce readiness suggests what might be going on.

When asked whether they were incorporating real-world projects into their courses, between 58% and 73% of faculty said they were, depending on whether workforce readiness was a high priority for them. Yet only 26% of students reported completing a real-world project in a course.

That is not a small gap. Maybe faculty and students define "real-world project" differently. Maybe these projects are concentrated in a subset of courses. Maybe faculty are counting assignments that students do not perceive as particularly authentic or connected to the real world.

I don't know the answer. But when nearly two thirds of faculty say they are assigning real-world projects and only a quarter of students say they have completed one, it suggests that something is getting lost in translation.

Does this disjuncture extend to other teaching practices, such as active learning? If students do not recognize the things institutions believe they are providing, then measuring implementation may tell us very little about actual student experience.

We may have reached the point where most faculty know the language of good teaching: rubrics, active learning, authentic assessment, career relevance. The question is whether students are actually experiencing those things in meaningful ways.

I wish Tyton had explored this disjuncture in more detail because I suspect it lies at the heart of much of the engagement challenge in higher education and, by extension, student success itself.

Girls just want to get paid

A new report from the Strada Education Foundation digs into data from its 2025 State Opportunity Index to examine who gets paid for work-based learning. The results are disappointing.

Work-based learning—and student employment more broadly, including internships and apprenticeships—is becoming increasingly important in helping graduates secure meaningful employment. That point was reinforced for me by a recent report from ZipRecruiter that also landed in my inbox. Looking at recent graduates, ZipRecruiter found significant differences in employment outcomes based on prior work experience.

Which makes Strada's findings all the more troubling.

But a lot of internships and work-based learning is unpaid, severely limiting access for those for whom working on an unpaid basis is simply out of the question. Strada found a significant difference in the number of paid opportunities between four-year schools and two-year institutions. But within these broader categories, women, first-generation students, and students from underrepresented racial and ethnic groups were far less likely to be paid for work-based learning opportunities.

The disproportionate impact at two year institutions is particularly concerning. We increasingly tell students that work-based learning is essential to career success while simultaneously making many of those opportunities financially inaccessible, especially for those who need it most. In other words, we have translated the importance of work-based learning but not yet translated access to it.

Birds

I found a newsletter listing of 100 of the greatest bird names. I found it useful as a potential source of new nicknames for annoying collaborators or colleagues.

Who amongst us, for example, has not had to deal with a Screaming Cowbird?

Not to mention an Oleaginous Hemispingus (I once worked with several of those).

There are many others (Bananaquit, Obscure Berrypecker, etc., etc.). But you, dear readers, you are the Charming Hummingbirds of my life.

In defense of progress

The latest U.S. Department of Education Condition of Education 2026 report contains some good news for those of us focused on student success. Given how much of the conversation centers on crisis and decline, it is worth occasionally remembering that many of the metrics we care about have actually improved over the past decade.

Some of the highlights for me include the following.

At two-year institutions, completion rates were higher across every student category for the 2016–17 entering cohort than for the 2009–10 entering cohort. First-time students, returning students, full-time students, and part-time students all saw gains.

Completion rates are also higher for the 2016-17 cohort than the 2009-10 cohort across all time frames in both 2-year and 4 year institutions.

Sometimes the translation problem works in the opposite direction. We have become so accustomed to narratives of decline that we overlook evidence of improvement. But we shouldn’t do that, instead we should celebrate this progress.

Distrustful, skeptical, still interested

Public Agenda has a subtly nuanced and useful report examining the experiences of young men and their engagement with civic institutions, including higher education. The authors map respondents along two dimensions—trait orientation and political alienation—and use those dimensions to create a set of personas that help explain how different groups of young men view institutions, opportunity, and their place in society.

In terms of higher education they describe a situation where the majority of men are distrusting of higher education or simply don’t see the ROI.

Young men especially question the extent to which higher education will prepare them for a career.

Despite this, a large percentage of young men without a degree indicate that they would like to obtain one.

And those who did complete a degree see it as paying off in multiple ways.

I find the contrast between the lack of trust and negative perceptions of ROI and yet the continued desire to get a degree fascinating. One of the things this suggests to me is that when it comes to men we have a translation problem in higher education. The challenge may not be convincing young men that education matters. It may be helping them translate higher education into the futures they want.

It looks like they are hosting a webinar on this report on July 9th.

Musical coda

Something to hold on to

f you found this issue useful, feel free to forward it to the Charming Hummingbirds in your own life.

Thanks for being a subscriber.

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.

Before I jump into this week’s post, three things to be aware of.

Tomorrow I will be participating in a webinar sponsored by Mentor Collective that, as it happens touches on some of the themes I am writing about today. Check it out and maybe join us.

Next week I will be attending most of the Open edX conference happening in my backyard of Salt Lake City. Last year it was in Paris; this year it’s in the Paris of the Wasatch Valley. If you will be there, let me know. I also have a highly opinionated and non-comprehensive food guide to Salt Lake City, which I have linked to at the bottom of this post for anyone who is interested.

I will also be heading to Norway the following week and, again, the offer to meet and chat stands. I will alas be missing a catch-up with my regular and favorite Norwegian co-conspirator. But Roger won’t be able to escape me forever. On to the post.

Preparing students to thrive in the job market is an increasingly critical part of student success. Not only because it is the right thing to do, or because students increasingly judge the value of higher education in terms of its ability to prepare them for what comes later, but also because universities are increasingly being held accountable for students’ post-graduation earnings.

And the task facing students in the labor market is increasingly daunting.

Navigating entry into and flourishing within the job market is often framed as an information challenge. But in many ways students are awash in career information: labor market data, career centers and counselors, advice networks, and increasingly AI tools that help them identify opportunities and prepare for them.

Yet students still struggle to navigate careers and make decisions about their futures. One report after another this week circled the same underlying issue from different directions. Together, they convinced me that students increasingly suffer not from an information deficit, but from a translation deficit.

Career navigation as translation

A new report from Harvard’s Kennedy School Project on Workforce provides some fascinating insights into how students and workers think about and navigate careers. Its central argument is that careers are no longer linear. Instead, most people’s working lives are characterized by a constant need to adapt, pivot, and potentially exit and re-enter the workforce. The report treats career navigation as a problem of interpreting fragmented labor-market signals in an unstable economy.

What I found particularly interesting were the findings about how students and workers navigate this increasingly complex environment. The results reveal a growing divide between institutional authority and practical utility.

Traditional institutional sources remain trusted, but students increasingly turn elsewhere for guidance that feels more immediate, actionable, and personalized.

The contrast between traditional sources of career information — teachers, counselors, and career service centers — and newer, less formal sources was particularly striking. Traditional sources ranked higher on reliability than helpfulness, while networks, online tools, and social media ranked surprisingly high on helpfulness despite being viewed as less reliable.

What this chart suggests to me is that students and workers are finding value in:

• low-friction,

• ambient,

• conversational,

• always-on systems.

Students and workers increasingly appear willing to trade some degree of reliability for information that feels more immediate, personalized, conversational, and actionable.

The challenge is that the kinds of translation students increasingly prefer — low-friction, always-on, AI-mediated — may not be the kinds of translation best equipped to help students navigate uncertainty, identity, and long-term career development.

Many low-friction guidance systems are optimized for immediate decisions. Relational translation, by contrast, is often about helping students navigate uncertainty over time.

Which partly explains why students are still holding onto people and institutions that provide guidance. Students appear to be constructing hybrid guidance ecosystems made up of both traditional and non-traditional sources. But what is still missing is what the authors refer to as “interpretive infrastructure” to help people make sense of the labor market and navigate pivots, exits, and re-entries.

The role of interpretation, or translation is critical. As they explain:

Career readiness as curricular translation

While the Harvard report focuses on labor-market navigation, a report from Every Learner Everywhere (ELE), The Career Readiness Imperative in Gateway Courses, treats career readiness primarily as a curricular translation problem.

The ELE report is based on the premise that too few students engage with traditional career supports such as career centers or co-curricular activities (for example, career-oriented clubs). Institutions already teach students many skills and competencies that are enormously valuable in the workforce, including in the liberal arts. But what they have traditionally failed to do is help students recognize, narrate, articulate, and connect their learning to what employers need.

To address this, the report proposes embedding career readiness throughout the curriculum. Drawing on frameworks such as those from NACE and AAC&U, it emphasizes integrating career discussions and career-oriented content across courses, especially gateway courses such as required quantitative, writing, or introductory disciplinary courses.

Including career content in these kinds of courses means that students are exposed to career translation early and often, particularly when they may need it most. Faculty help provide students with ways of translating between the curriculum and the workplace.

I like this approach, especially in how it moves away from the narrow vocationalism that characterizes many efforts to incorporate career support into the curriculum. But it is only a partial solution.

Even apart from its deeply institutional focus, the model relies heavily on faculty to do the translation work. We have seen this become a problem in many other “across the curriculum” initiatives, where responsibility often gets pushed downward and informally assigned to the most junior faculty member, or to the one person in the department willing or interested enough to take it on.

But the Harvard report suggests the challenge is messier and far more ongoing. Translation is not simply a one-shot matter of helping students connect coursework to careers inside a class. Labor markets themselves are fragmented, unstable, and difficult to interpret. Students pivot, stop out, re-enter, relocate, and continually reassess opportunities and risks.

Translation in that context is not an episodic curricular intervention but an ongoing process of interpretation, navigation, and adaptation. In unstable labor markets, translation increasingly becomes lifelong rather than episodic work.

Mentorship as relational translation

What both reports never fully solve for me is who actually performs this translation over time, especially as students pivot, reassess, stop out, re-enter, and encounter changing labor markets. A third report, this time from Mentor Collective, partially answers that question by reframing translation as relational rather than merely informational. (In the interests of full disclosure, I am participating in a Mentor Collective webinar tomorrow — you should come — but my opinions, as always, are my own.)

Unlike many institutional supports, mentoring relationships are adaptive. Mentors can contextualize advice, personalize guidance, revise interpretations over time, and help students navigate ambiguity rather than simply deliver information. That matters in labor markets characterized by uncertainty, pivots, and unstable pathways.

The report describes this relational translation in several ways.

What is striking about these examples is that mentors are not simply providing career information. They are helping students interpret unfamiliar systems, decode professional expectations, imagine possible futures, and convert aspiration into action. Much of this work involves navigating tacit knowledge and hidden norms that are difficult to fully codify into websites, career modules, or competency frameworks.

Mentors are not only interpreters. They are also cultural brokers, navigators, and translators of institutional and professional norms. And they do this on an ongoing basis. In institutions where support systems are often fragmented and siloed, mentors can also function as continuity figures, helping students connect experiences and navigate across institutional boundaries.

Taken together, these reports suggest that translation is doing much more than simply conveying information about jobs. It is about explaining worlds, navigating systems, and helping students construct identities and possible futures.

Translation does more than convey information

Across these reports, it becomes clear that translation in career success — and the translation being done by mentors, faculty, advisors, and institutions — is performing at least six different kinds of work:

Cognitive work: Helping students make sense of complexity.

Interpretive work: Helping students connect signals to decisions.

Narrative work: Helping students tell coherent stories about themselves.

Emotional work: Reducing intimidation, uncertainty, and paralysis.

Social capital work: Helping students access norms, networks, and hidden rules.

Institutional work: Bridging the gap between formal institutional systems and lived student experience.

Higher education institutions have traditionally been organized around content delivery and information provision. But increasingly, student success depends not just on helping students acquire information, but on helping them construct coherent professional identities and navigate uncertain systems.

Increasingly, student success — especially in career terms — depends on interpretation, navigation, identity construction, contextualization, sense-making, and relationship-building.

What these reports collectively suggest is that translation itself is increasingly becoming institutional infrastructure. Not a supplemental service or optional co-curricular support, but a core institutional function.

Universities historically organized themselves around content delivery, credentialing, and information provision. But in labor markets characterized by uncertainty, pivots, and fragmented pathways, students increasingly need systems that help them interpret, contextualize, and navigate complexity over time. That is a very different institutional challenge.

It is also a huge shift, and AI complicates this further because it dramatically lowers the cost of information access while potentially increasing the need for interpretation and judgment. Higher education increasingly suffers not from an information deficit, but from a translation deficit.

But once translation becomes infrastructure, questions about power, norms, and inequality become unavoidable.

The challenges of translation

Translation may be increasingly critical, but it is important to recognize that it is not always an unalloyed good. Translation in any context is complex, contested, unequal, partial, and political. In student success and career navigation, translators inevitably carry assumptions about what success looks like, what kinds of careers matter, and how professionalism should be performed.

Translators carry their own biases and blind spots

Mentors, faculty, advisors, networks, and AI systems all interpret the world through partial experience. And social capital has a tendency to reproduce itself. The Harvard report hints at this when discussing networks that are narrow and self-reinforcing.

Mentors as translators may steer students toward familiar paths. Faculty may overvalue academic trajectories. Employers may privilege dominant norms. Networks often reproduce class and cultural assumptions.

Translation is never neutral.

Translation risks becoming assimilation

A great deal of translation work in career readiness implicitly involves:

• teaching students how to speak professional-managerial language,

• teaching them institutional norms,

• teaching them how to perform employability.

But this raises difficult questions: whose norms, whose professionalism, and whose communication styles?

This especially matters for:

• first-generation students,

• neurodivergent students,

• working-class students,

• students of color,

• international students.

Sometimes “translation” quietly becomes less about helping students navigate systems and more about helping them adapt to unequal systems rather than changing them.

When organizations outsource translation

Organizations increasingly rely on mentors to provide interpretive and developmental labor that formal systems either cannot or will not provide directly.

When I worked at Gartner, we were assigned mentors — usually from outside our immediate unit — to help us find our feet as analysts. My mentor Kristin was the best mentor imaginable, and she remains so to this day. In her words, “there is no endpoint to mentoring.”

At first, I thought the system was amazing — and to be fair, I got enormous value from it. But over time I also realized it partly allowed the company to absolve itself of the responsibility to provide more formal professional development and institutional support.

Gartner-style mentoring systems can be enormously valuable. But they should not become replacements for broader investments in training, development, and institutional support.

Translation as infrastructure

Higher education has spent years building information systems. But increasingly, student success may depend on building interpretive systems instead.

Those systems need to be intentionally developed and supported. They include:

• mentoring networks of all kinds,

• embedded advising,

• peer interpretation,

• labor market contextualization,

• alumni guidance,

• career translation infrastructure,

• adaptive coaching.

But building interpretive systems also requires understanding how translation actually functions within student success work. If translation is increasingly becoming infrastructure for student success, then institutions need to think much more carefully about who does that translation, whose norms shape it, and who gets left out.

A Highly Opinionated Guide to Eating in Salt Lake City

Morgan’s completely quirky, personal and incomplete food guide to Salt Lake City, for those attending Open edX or just visiting.

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.

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.

Somehow it’s the end of April already. Time for a few charts that have stuck with me this month, and what they might mean for student success.

Humor is harder than it looks

I use humor a lot in presentations (which I am pretty great at, and available for). So I know how hard it is.

So I wasn’t surprised to read that in a study done on humor in scientific conference presentations a small minority of efforts at humor landed well enough to get a laugh.

The cost of college, measured in time

We’ve all been saying this for years. Now we have the numbers. Current students would have to work so much longer to cover the costs of annual tuition. At a public institution in the US, for in-state tuition, they would need to work 26 hours per week. At a private institution, it is 102 hours a week.

Employment while at university has morphed into a structural barrier to completion rather than a pathway to affordability.

From cost pressures to research quality

How much of this should we trust?

The reproducibility of research results is a problem in many fields. But I was more than a little dismayed to see the poor showing of education on this chart on the reproducibility of research in the social sciences.

We should probably do fewer randomized controlled trials (RCTs) in education anyway, which would at least reduce the number of fragile findings. But if we can’t reliably reproduce results in education research, we should be cautious about building entire student success interventions on them.

Florence Nightingale & W.E.B. Du Bois, data nerds

I love a good data visualization (though I limit myself to admiring them rather than producing them). I recently came across a post sharing some great historical visualizations I hadn’t ever seen before.

I knew that Florence Nightingale was a big data geek and was reminded of this by her depiction of causes of death among British soldiers in the Crimean War.

But I didn’t know that W.E.B. Du Bois was also a superb visualizer of data. Du Bois had studied under William James at Harvard.

Part of me is cringing at the concept of an “American Negro Exhibit.” But the visualizations are delightful and more of them are available at the Library of Congress.

This is what great data visualization does: it doesn’t just show data, it makes an argument.

Students, as always

First, a disclosure: my spouse works at the University of Utah Library.

I have been having a rough time with all the national news and negativity. Unsurprisingly, students provide a great antidote.

Yesterday this Instagram reel from the University of Utah library won an award from the Association for Research Libraries. It’s called Whisper Something Good.

Instagram post

The internship bottleneck

Brandon Busteed shared data from the National Association of Colleges and Employers (NACE) showing the rate at which interns convert to full-time employees. Internships are hugely important and there aren’t enough of them for the number of students seeking them.

We’ve quietly made internships a gatekeeper to opportunity, but without building enough capacity to support that expectation. It has also become an equity problem.

When the metric misses the point

Hannah Ritchie (whose writing on data I adore) recently wrote about national rates of train electrification, and why they didn’t quite make sense.

The numbers didn’t feel right to her, so she had a closer look. It turns out the data used in this and other similar charts are based on the proportion of the railway lines that are electrified, with no reference to how busy the lines may be.

She reran the numbers looking at train traffic, but not passenger numbers, and got some quite different results.

This is exactly the kind of measurement problem we run into constantly in student success.

AI and the new divide

It wouldn’t be a student success charts post without a John Burn-Murdoch chart. This one has been doing the rounds this week, understandably so given the sharp relationships it shows between earnings and AI use.

I keep using AI more and more in the hope it shows up in my earnings. So far, no luck. I guess the causality runs the other way. But if that ever happens, one of the things I would do is pay for an actual subscription to the Financial Times so that I could get more John Burn-Murdoch charts.

But this chart is a useful reminder: AI adoption is not evenly distributed. It’s clustering among higher earners, which risks widening gaps rather than closing them.

Musical coda

Great songwriting and great data visualization do the same thing: they convey a lot with very little.

“Busted flat in Baton Rouge, waiting for a train.”

Nine words. Character, situation, mood.

That’s the bar.

H/T Steven Mintz

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.

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.

It has been a hellish week of allergies, exhaustion, and frustration with datasets, balanced by a birthday celebration in the OSS household. But apart from that, what is the news in student success?

This week’s readings all point to a familiar pattern: we keep building better ways to pull students out of the river, but we are far less willing to redesign the system that keeps pushing them in.

It’s a design problem, again

The Trellis Strategies Some College, No Credential (SCNC) Survey Report describes a landscape familiar to many of us. There are 43.1 million people with some college but no credential, including 37.6 million working-age adults. Trellis’s large survey of former undergraduate students produces a set of insights that are both encouraging and frustrating.

On the encouraging side, a majority believe that higher education is a good investment, and nearly 73% say that returning to earn a credential would improve their career and earning potential. Sixty-three percent of respondents report intending to return to finish their studies.

On the frustrating side, the report presents a familiar list of reasons why students stop out: family responsibilities, finances, health issues, and employment demands.

The report reinforces something we have seen repeatedly: students are not leaving because they cannot succeed academically. They are leaving because higher education is structured around assumptions of time and financial stability that no longer hold.

I found some of the differences between two- and four-year schools striking. On one level, the greater reported financial stress at four-year institutions makes some sense, given the often lower cost of two-year schools. Differences in the role of campus life also make sense, since more students at two-year institutions are commuters and may have less time for campus engagement. But the divergence in other areas—particularly the much higher share of students at four-year institutions citing health and academic reasons for leaving—is cause for concern.

But I was mostly frustrated by the implicit (and sometimes explicit) assumption that institutions can solve these issues simply by identifying which barriers affect their students and layering on more services. That assumption becomes even clearer when you look at the toolkit.

The recommended actions are largely reactive. For example, the report states that 71% of SCNC students never spoke to anyone before stopping out. The toolkit responds with recommendations around referral channels and scripts for advisors. Similarly, it recommends instituting exit surveys, tracking students who leave, and monitoring payment plans. These are symptom-detection mechanisms, not system redesign. Institutions are not redesigning the system to accommodate student reality; they are building increasingly complex processes to manage the consequences of a system that does not fit that reality.

This is less a critique of Trellis than a reflection of how the sector as a whole approaches student success. It is less a student problem than a system design problem. The result is a system that becomes increasingly complex without becoming more effective.

Managing around constraints

Reading the Trellis report, I kept thinking about this video of a man going to extraordinary lengths to park in an impossibly narrow garage.

Students are optimizing for reality

That mismatch between system design and student reality doesn’t just show up in retention. It may also explain another trend. The National Student Clearinghouse Research Center recently released its Undergraduate Degree Earners Report. Some coverage focused on the fact that, for the first time, students ages 18–20 comprise the largest share of first-time associate degree earners. But what caught my eye was the increase in students earning certificates rather than associate (two-year) or bachelor’s degrees.

Certificate completers are up 5.7% year over year, compared to more modest gains for bachelor’s degrees (2.8%) and associate degrees (2.6%).

This year’s increase in certificate completers is less dramatic than last year’s—5.6% compared to 16.0%. But over the two-year period, certificates have grown by more than 21%, while bachelor’s and associate degrees have seen much more modest growth in the 2%–4% range. This may be less about increased demand for certificates and more about students seeking faster, more flexible pathways that better fit their lives.

Same problem, same playbook

That same pattern shows up in institutional responses as well. A news article in The Philadelphia Inquirer paints a picture of significant student success challenges at Temple. Based on an internal university report obtained by the newspaper (and not shared with readers), the university is experiencing notable declines in retention—not only from freshman to sophomore year, but also from fall to spring. The consequences are clear in terms of students not staying enrolled, but the report also frames the issue as a substantial loss of revenue.

The news report does not give us a lot of information about why the drop in retention happened. Officials attribute the decline in enrollment to financial pressure as well as some academic challenges. In seeking to address the challenges Temple is following a tried and true playbook, according to the President John Fry.

The response is familiar: more tools, more tracking, more interventions. But note what is missing—any indication that the underlying structure might be part of the problem.

The story also surfaced something else that is easy to overlook but deeply related. The report included a study by NACUBO showing that the majority of graduates are concentrated in a small number of programs (the Pareto principle at work). At first glance, this might appear to have little to do with the retention challenge.

This raises an uncomfortable question: how much of the curriculum actually supports student pathways, and how much exists because of internal structures and incentives? Again, the response is to manage symptoms rather than redesign the system.

It’s a design problem—assessment edition

This question about what we are actually designed to deliver shows up in another domain as well: assessment. There has been no shortage of bad takes on higher ed this week, but there was at least one strong piece from Mark Bassett, a faculty member at Charles Sturt University (which I always think of as the Johns Hopkins of Australia—always one letter away from predictability).

Bassett writes about the impact of AI on assessment. Many of us have been stumbling toward the conclusions he reaches for some time, but he frames the issue in an especially cogent way, using Stephen Covey’s circles—control, influence, and concern.

I am going to quote from the piece more than I usually would, because he frames the issue so well. He begins by arguing that GenAI has thrown the question of assessment into sharp relief, laying bare a set of assumptions and weaknesses that were always there. The emphases are mine.

But GenAI also raises questions about professional identity that are embedded in those practices and the role assessment plays in higher education.

All faculty face these challenges in assessment from GenAI, but how they respond depends on whether they operate within the circle of concern or the circle of control. Those in the concern camp see the impact of GenAI as something whose root causes they cannot influence. They respond by emphasizing detection and punishment.

For faculty or institutions in the control camp, Gen AI is seen as something where the root causes of impact can be understood and action taken.

It is not a perfect piece. It does not address how to solve these problems. But as an analysis, it surfaces issues I haven’t seen framed this clearly before and will, I believe, ultimately help us address one of the core challenges facing higher ed today.

What GenAI exposes in assessment is the same problem we see everywhere else: we have built systems that assume stability, and we are trying to patch them in a world where that assumption no longer holds.

I started with a quote from Desmond Tutu, whom I deeply admired. I have a funny story about the time I got to meet him. For the cost of a fake beer or a real coffee, I will share it with you the next time we meet.

Musical coda

To the firecracker, on the occasion of her birthday.

If you know someone who spends a lot of time pulling students out of the river, feel free to send this newsletter along to them. Or better yet, someone who might be willing to walk upstream.

Thanks for being a subscriber.

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.

EdTech has a recurring habit of announcing the next transformative breakthrough with enormous fanfare and little introspection. A new technology. A new model. A new price point that promises to reshape higher education, reduce costs, and finally achieve scale.

Often, these innovations are not entirely new. They are familiar ideas repackaged with updated language and some contemporary technology. But they tend to rest on a persistent set of misunderstandings — about how students actually engage, how learning unfolds, and how difficult scale really is.

The recent sequence of events involving Sal Khan offers a particularly instructive example. First came the quiet recalibration around Khanmigo, the AI tutoring tool that was initially framed as transformative but appears to have seen limited sustained use. Almost immediately afterward came the announcement of a new joint venture with ETS and TED to launch a low-cost online degree. The new venture was touted as.

Individually, neither development is remarkable. But together, they illustrate something more revealing: the rapid pivot from one ambitious claim to another, without fully reckoning with the structural constraints that limited the first.

The 5% problem

The first signal of this latest cycle was the Chalkbeat article revisiting the performance of Khanmigo, Khan Academy’s AI tutoring tool. In it, Sal Khan appeared to temper some of his earlier claims about how transformative AI tutoring would be. The article noted that student uptake had been limited and that sustained usage was concentrated among a relatively small subset of students.

The piece quickly spawned a familiar secondary wave of commentary. But the most interesting detail in the article was a familiar one. Only a small fraction of students regularly used Khanmigo; roughly 5 percent in some implementations.

Khan and other company representatives expressed frustration at students’ lack of engagement. The students weren’t using it “correctly.” They weren’t self-initiating.

This is where the pattern reasserts itself.

When only 5 percent of students engage with a voluntary educational tool, that is not primarily a student problem. It is a design and integration problem. The hardest challenge in EdTech is not building a capable tool. It is embedding that tool into systems that support motivation, accountability, and sustained use — especially for students who are juggling many demands and uneven academic preparation.

Teaching the already motivated 5 percent is not transformation. It is amplification. Which makes the rapid pivot — from limited engagement with an AI tutor to the announcement of a fully AI-mediated, $10,000 degree — especially striking.

From tutor to degree

Within a week of tempering expectations about Khanmigo, Sal Khan pivoted to something new, this time in higher education. He announced a collaboration with testing giant ETS and TED (of TED Talks fame) to create an accredited university offering low-cost degrees.

The rhetoric is expansive. The ambition is unmistakable.

The details are still emerging. But the scale aspirations are clear: tens of thousands of students, rivaling flagship state universities.

Blue-chip corporate names lend credibility, and many of these same companies are already involved in creating and offering certificates and content on other platforms as well as their own.

None of this is unreasonable. But none of it is new. And none of it resolves the core challenges that have faced other “revolutionary” online low-cost degrees: sustained engagement, student persistence, and the economics of large-scale asynchronous delivery.

If voluntary AI tutoring struggled to engage more than a small fraction of students, what exactly changes when the same logic is applied to a fully asynchronous degree? To understand why this matters, it helps to remember what happened the last time we declared scale solved.

Reinventing what already exists

The Khan TED Institute is presented as a reimagining of higher education for the AI age. Strip away the rhetoric, however, and the underlying structure is familiar: asynchronous lessons, simulations, peer dialogue, remote faculty guidance.

Fully online higher education is not an untested frontier. Nearly 30 percent of U.S. higher education enrollments now include online coursework. Entire institutions operate at scale in fully asynchronous formats.

And yet the proposed design is framed as if it marks a departure:

There is nothing radical here. Virtual lessons, asynchronous assignments, group projects, simulations, peer dialogue sessions— essentially discussion boards or tutorials —have been core elements of online pedagogy for two decades. They are not breakthroughs. They are baseline design choices.

This pattern is familiar. A decade and a half ago, MOOCs promised to democratize higher education with nearly identical rhetoric: lower cost, global reach, scalable delivery, brand-name partners.

The MOOC model evolved into online degrees, many priced below traditional programs and delivered through platforms like Coursera — the most serious and well-resourced attempt to operationalize this approach at scale. Coursera combines brand-name university partners, a massive global learner base, and what Phil Hill has called an enrollment “flywheel.” And yet, even with those advantages, the degree business has proven harder than early projections suggested, as recent earnings reports make clear.

Recent data show that degree enrollments have plateaued and revenue per degree has declined. Coursera has increasingly shifted emphasis toward shorter credentials and enterprise offerings rather than continued degree expansion. This is the best-resourced version of the model. If scale were easy, it would be visible here.

The lesson is not that online degrees cannot succeed. It is that scale requires sustained marketing investment, institutional credibility, student support infrastructure, and retention strategies that go well beyond content delivery.

The proposed Khan TED Institute would enter a more crowded and mature market than Coursera did in 2017. It would do so without an existing institutional brand, and with an undergraduate target population that is typically more brand-sensitive and retention-challenged than graduate learners. A $10,000 price point is rhetorically powerful. But price alone does not solve acquisition costs, persistence challenges, or the economics of sustained student support.

What makes this different from earlier Khan Academy expansions is that the degree market is not a philanthropic distribution problem. It is a competitive acquisition problem. In the past, Khan Academy and Khanmigo benefited from large institutional adoptions, foundation support, and government partnerships. They did not have to fight for individual tuition-paying undergraduates in a saturated, brand-sensitive market with high marketing costs and low retention margins.

Competing in the undergraduate degree market requires sustained marketing investment, enrollment operations, student support infrastructure, and regulatory compliance — not simply compelling rhetoric and strong corporate brand associations.

Scale is not frictionless

The proposed curriculum for the new Khan TED Institute is fairly predictable and not unreasonable, if a bit STEM and jargon heavy.

On paper, this looks coherent. It blends some liberal arts with technical fluency, and applied collaboration. But what is striking is how frictionless the model sounds.

Content can be delivered asynchronously. Skills can be practiced in simulations. Dialogue sessions can be scheduled. But learning — especially for under-prepared or time-constrained students — is not a smooth pipeline from exposure to mastery. Learning requires feedback loops, accountability, structured practice, and often sustained human intervention. AI can assist with parts of this. It does not replace the systems and practices that make engagement durable.

If a voluntary tutoring tool struggled to move beyond a small, self-motivated minority, it is reasonable to ask what mechanisms will ensure persistence in a largely asynchronous degree, especially when the underlying logic still seems to be that access to a tool or content is enough and that students themselves bear responsibility for engagement.

This fact comes out in the recent Chalkbeat article from Sal Khan.

And from Kristen DiCerbo the Chief Learning Officer.

And yet the need for deliberate and designed engagement is even more important in an online degree than in a tutoring app. The students most drawn to low-cost, asynchronous degrees are often balancing work, care-giving, and uneven academic preparation. Designing for their success requires more than access to content and AI tools. It requires structured support systems, good pedagogy and instructional design, integration into institutional processes, and sustained human accountability. And nothing in the current framing suggests that these structural engagement challenges are being treated as primary design requirements rather than downstream engineering problems.

The interval is shrinking

The debate over Khanmigo also prompted a sharp observation from one of the most perceptive critics of technology-mediated learning at scale. Before the announcement of the Khan TED Institute, Justin Reich proposed what he called the “time-to-TED-talk-renunciation” metric — the interval between a bold claim about technological transformation and the subsequent retreat to a more modest position.

Reich’s framing is humorous. But it captures something real. Each new technological promise arrives with expansive rhetoric. Then implementation collides with student behavior, institutional inertia, and economic reality. The recalibration follows.

Reich even sketched a trend line.

The interval appears to be shrinking.

And then, within days of Reich posting that chart, Sal Khan announced the Khan TED Institute — from the stage of a TED talk.

The humor works because the pattern is familiar. But the stakes are not trivial. Each cycle absorbs institutional attention, philanthropic dollars, public imagination — and some students who will enroll in the new experiment and struggle.

The deeper issue is not ambition. It is repetition. We keep repackaging familiar models without grappling with the structural constraints that limited them in the first place. In this case, that constraint is engagement — how to design for sustained participation among students who are balancing work, caregiving, uneven preparation, and financial risk.

The interval between declaration and re-calibration may be shrinking. The underlying mistakes remain unchanged — and they will produce the same outcomes unless the design logic changes.

If this was helpful or interesting to you, forward this post to others who may be interested. All we ask is attribution.

Thanks for being a subscriber.

 

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.

So I may have spent most of yesterday trying to construct a student success version of the NCAA women’s basketball tournament after being annoyed by an alternative bracket someone shared on LinkedIn (ok, if you insist, it was this one). I’m not going to share my version with you because I ultimately learned that coming up with good measures is really hard—something you would think I would know by now.

The things we count (and the things we don’t)

I love Sketchplanations, but this week I really love Sketchplanations. Jono Hey has a wonderful graphic on the McNamara Fallacy.

He goes on to cover some related measurement challenges, such as Goodhart’s Law (which I thought about on an hourly basis during my last couple of years at Gartner), Campbell’s Law, and the Cobra Effect.

We wouldn’t know anything about that in student success, now would we? No, never.

And as it turns out, this week offered several reminders of what happens when we measure what’s easy rather than what matters. One place this shows up very clearly is in how we define completion.

The students who almost finished—and why that matters

One of the things I tried to use to construct my scores for the alternative bracket was completion. Oddly enough, one of the things I read this week was a fascinating report from About Texas 2036 on the need to grant recognition to students for credits they have already earned.

The central thrust of the report is that by 2031, 61% of jobs in Texas will require some form of post-secondary education, but the state is falling well short of that number. Yet a hidden opportunity exists: between 2012 and 2022, more than 54,000 Texans earned 60+ credit hours but left higher education without a credential. The report dubs these students “Potential Completers.”

But I found two things about the report striking. The first is the scale at which we are losing students from the post-secondary system—and how late we are losing them. Almost half (47%) of those who entered the post-secondary system in Texas completed either a bachelor’s degree, an associate degree, or a certificate, which is less than ideal but not as bad as I might have predicted. But tens of thousands left after earning 60+ credits—i.e., really close to completion. As the report argues:

This does beg the question of why these students are falling through the cracks—and whether the factors causing them to stop out are different from those affecting the almost half of students who leave earlier.

The second thing that struck me in the report was the shift in how we talk about these students. Instead of framing them as students who didn’t finish, or even as failures, there is a move toward thinking of them as near-completers whose progress has gone unrecognized.

That is a subtle but powerful move, and one I hope we see more of. It’s also a reminder that sometimes the problem isn’t that students aren’t succeeding. It’s that we’re not very good at recognizing the ways they already have. But measurement isn’t just about credentials. It’s also about how we understand students’ day-to-day lives.

This is not what “affording college” looks like

Trellis Strategies just released their Student Financial Wellness Survey report, and it makes for some interesting—and at times uncomfortable—reading. On one level, the report can be read as a description of the financial hardships students face. But it quickly becomes clear that this isn’t about students being stretched. It’s about structural instability and students living on the edge.

• 54% can’t access $500

• 65% ran out of money

• 54% face basic needs insecurity

This is not exactly the financial cushion most of us would hope for.

On one level, this reads like financial stress. But it quickly becomes clear that this isn’t about students being stretched. It’s about structural instability.

Financial instability is the baseline environment in which college happens, and it affects a majority of students in both sectors. And it’s clear that this financial stress is directly interfering with learning. Financial and financially related disruptions are widespread among students responding to this survey, with around a fifth reporting that they have had to miss class because of a problem.

Of the students experiencing financial difficulties, nearly half said their financial stress made it hard to concentrate on their coursework. This isn’t just about whether students can enroll. It’s about whether they can actually learn once they get there. None of these are unsolvable problems. But they do require us to design around students’ actual lives rather than the ones we imagine they have.

What’s also striking is that students are financing college with consumer financial tools, including credit cards, buy-now-pay-later services, and payday loans.

Which, when you think about it, makes a certain kind of sense. If your financial aid doesn’t cover your rent or your groceries, you don’t have a lot of options. As coping strategies go, it’s a pretty rational one.

What is also clear is that the mental health challenges we keep talking about are not separate from these issues, but are embedded in the financial and structural challenges students face.

Using validated scales to measure student well-being, the survey found that:

Increasingly, we see a chain of conditions: financial stress → time pressure → isolation → mental health → academic disruption. And it is a chain, not a set of separate issues.

But the most surprising thing in the report is that, despite all of this, students remain remarkably optimistic about higher education. A large majority still believe it is a good investment. Even more believe it will lead to a better quality of life. That matters. It means the core value proposition hasn’t collapsed. We haven’t lost them.

But it also raises a harder question: students’ continuing belief in higher education, alongside their debt anxiety, instability, missed classes, and depression, shows how decoupled their beliefs have become from their actual experience of college.

The report doesn’t offer easy solutions. But it does point in a direction. Financial wellbeing needs to be a bigger part of student success, and if we are serious about lowering the cost of college, we also need to pay attention to the ancillary costs that shape students’ day-to-day lives. It also suggests that improving student success may require spending a little less time refining our measures and a little more time paying attention to how students are actually living.

We spend a lot of time trying to measure and manage student success. But weeks like this are a useful reminder that students are navigating systems that don’t always line up with their lives. And when that happens, they adapt.

What happens when systems don’t fit

I’m going to leave you as I started, talking about the NCAA college basketball championships. There is a growing issue with the transfer portal in U.S. college sports, and no, I don’t think an executive order forcing students to stay put is either serious or likely to succeed.

But I am struck by the impact of the urge to transfer to a new program at the end of a season on college athletes. For example, just two days after the end of the season, 557 women’s Division I basketball players had entered the transfer portal, and most will be playing for new teams next year. That number will only increase over the next few days.

For a not inconsiderable number of players, this is not the first time they have entered the portal, as this example shows.

The vast majority of these players will never play professionally. What havoc is all this movement having on their careers and intellectual development, especially given how hard we make transfer when they do try to settle somewhere? It does make you wonder what this level of churn is doing to students’ academic lives.

And I will get off my soapbox by sharing the buzzer-beater that ended LSU’s season.

Musical coda

And thinking about measurement—and the growing distance between B1G campuses—here’s a very old rendition of Peter, Paul and Mary’s “500 Miles.”

If you found this useful, consider forwarding it to your teammates. If not, I will assume I am measuring the wrong thing.

Thanks for being a subscriber.

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.

This Very Good Boy turns 13 today.

But what happened this week in student success?

A number of things I read this week point to a similar underlying issue. Across very different domains—and in quite different ways—AI and workforce credentials, as well as career services and the use of adjuncts, reveal the same pattern: systems that optimize for short-term performance, efficiency, or provision rather than long-term learning, mobility, and effectiveness.

When performance isn’t learning

A new research report from Jason Lodge and Leslie Loble, part of the Australian Network for Quality Digital Education project, throws into stark relief the dangers of unwise AI use. Although the report is technically aimed at K–12 education, it is highly applicable to higher education. Lodge and Loble describe how AI, when used poorly, can undermine student learning.

This outsourcing of knowledge, or cognitive offloading, is increasingly being identified as a risk in the use of AI in learning, but this report advances our understanding of the issue.

The authors usefully distinguish between beneficial offloading—where AI is used to help with something peripheral to a task (for example, checking grammar)—and detrimental offloading, “when a learner uses AI to bypass this intrinsic cognitive effort (the desirable difficulties) required to build long-term knowledge schemas.”

Their argument relies heavily on the idea of a performance paradox, which I think is central to the challenge AI poses to learning.

This is a key observation. AI can be used to improve short-term performance while harming long-term learning. A core problem in higher education—especially in conventional methods of assessment—is that we are optimized to measure and reward performance rather than learning. This problem predates AI, but AI exposes it and makes it something we can no longer ignore. AI doesn’t create this problem. It exposes it.

A second insight from the report is that one consequence of cognitive offloading is that students never fully engage with domain knowledge. The authors usefully push back against the idea that skills like critical thinking can be decoupled from domain knowledge.

They further argue that in the absence of domain knowledge, problems arising from cognitive offloading such as the performance paradox become that much worse.

This is a compelling argument against the claim—sometimes made (and which I won’t link to)—that the best use of AI is in teaching Gen Ed courses.

Finally, their identification of the real “cheating” risk is spot on. The issue is not AI doing the homework, but cognitive offloading itself—something far more difficult to detect and address.

Early wins, long-term tradeoffs

An important report from the Fordham Institute examines the impact of industry credentials earned by high school students in Ohio. These credentials have grown immensely in popularity in recent years.

In the report, the author examines the longer-term effects—seven years after high school—of earning an industry credential, drawing on data from 1.3 million students who entered high school in Ohio between 2011–12 and 2019–20.

The top-line findings are striking.

Students who earn industry credentials are more likely to finish high school (87% versus 81%) but far less likely to enroll in formal post-secondary education (40.6% versus 49.3%).

Credential earners also earn substantially more—about 21%—than non-credential earners immediately after high school, but this advantage fades over time. By their mid-twenties (seven years after graduation), the premium has shrunk to just 5%. As the report notes, this points to a very short-term benefit.

Not all credentials are created equal. There are vast differences in the wage return of credentials in different sectors.

The wage gains not only depend on sectors, but are highly dependent on gender as well, though this is likely a function of the sector in which the credential was obtained.

Even setting aside those gender differences, the data point to the risks of focusing too heavily on getting students into jobs and optimizing for earnings too early. Short-term gains may be short-lived. We are systematically steering students toward decisions that look successful early—and may limit their long-term options. These credentials are less about building pathways than replacing them.

The research also leaves a number of questions unanswered. The finding that earning a high school credential is associated with lower college or university attendance is critical. But it remains an argument about correlation. The author controls for demographics, prior achievement, and course-taking patterns, comparing students within the same schools and cohorts.

Students who pursue credentials may have always been less interested in—and less likely to attend—college, and the author explicitly acknowledges this. He also suggests that credential pathways themselves may “nudge” students toward immediate work, but stops short of unpacking what that might mean or how it operates. Even here, we see the same issue: we are quick to interpret outcomes as effects when they may reflect underlying choices and structures. Students are responding rationally to the signals we give them. The problem is that those signals are often misaligned with long-term success.

Seen together, these examples point to a deeper pattern. We are not just measuring the wrong things—we are structuring systems around those measurements.

Is your university financially sustainable?

This is a bit of an inside joke, as it relies on some knowledge of what happened at the University of Dundee last year. I wrote about the report authored by Pamela Gillies, which dissected what happened there, over at On EdTech.

The report makes for fascinating, if grim, reading. I would strongly recommend it to anyone as obsessed with interested in university politics as I am. It outlines how, although there were some external stressors (such as declining international enrollments), the real cause of the problems lay in the university’s culture, including a lack of tolerance for dissent or differing opinions.

What stands out is not just the financial mismanagement, but the absence of challenge. Systems optimized for smooth operation and consensus often suppress the very signals that would reveal underlying problems.

Against this backdrop, someone writing under the nom de plume Prof Serious has created a quiz to help people determine whether their institution is financially sustainable. We could probably come up with an equivalent scoring system for the U.S.—I’m open to suggestions.

A design problem, not a motivation problem

Although students increasingly identify getting a job as a key goal, yet few of them use the career center.

Students want career outcomes, but they don’t use career services. That’s not a contradiction—it’s a design failure. We are offering support at the wrong time, in the wrong format, and often in the wrong place.

The systems we don’t talk about

A growing share of teaching in U.S. higher education is done by adjuncts. Much of this is wonderful: inspired instruction from people with a real passion for their subject, many of whom work in the field or bring a valuable real-world perspective to their teaching.

But it is also a real challenge. Many adjuncts are low-paid and must string together multiple teaching roles to make ends meet. They can be only loosely connected to the institution and, as a result, less likely to be integrated into student success structures and practices.

The proportion of adjuncts teaching students also varies significantly by institution type, as a new CUPA-HR report makes clear. Adjuncts teach at two-year institutions at nearly twice the rate as at doctoral institutions.

It also varies a great deal by topic. Most adjuncts are concentrated in the humanities, security/protective services, multidisciplinary studies and education.

This both intensifies the need for—and complicates—the delivery of student supports in these institutions and in the areas they serve. We design student success systems as if institutions are stable, cohesive environments, when in reality large portions of instruction are delivered by people structurally disconnected from those systems.

Across all of these examples, the pattern is similar. We focus on what is easiest to measure—performance, early earnings, initial placement. But these are not the same as learning, long-term mobility, or meaningful outcomes. We are confusing what is visible now with what matters over time.

Musical coda

I love this video (and this song).

In keeping with this week’s theme, this post is optimized for short-term performance—specifically, how widely it gets forwarded. Please don’t disappoint.

Thanks for being a subscriber.

 

As always, a pattern runs through this month’s charts. Student behaviors and needs are changing. Markets are changing. But the way we design and measure student success is not keeping up.

Students Are Not Who We Think They Are

I keep turning this chart from John Burn-Murdoch over in my head. It would be easy to dismiss this as a story about youth. But even at the upper end of that range, these are not just “young people.”

The shift is clear: younger adults report lower levels of persistence and follow-through, and higher levels of distraction and carelessness. That has direct implications for how students navigate college and what it takes to succeed.

According to Brandon Busteed, demand for internships is rising sharply even as supply is falling.

Adapted from Brandon Busteed

These are not small shifts. They fundamentally change what it takes for students to succeed and quietly invalidate many of the assumptions built into our advising models, course structures, and support systems.

And even if students themselves were not changing, the environment they are entering would still make success harder.

The Ground Is Shifting Beneath Them

The world our students are entering is becoming more competitive, more constrained, and more uneven. Some of this is good, much of it is not. But all of these changes create challenges.

Ranking institutions is, at best, an imperfect science. But the improvement in STEM rankings by Asian institutions over the past decade is striking.

Another great graphic from John Burn-Murdoch (anyone else want to join me in the fan club?).

If these numbers hold, this does not augur well for the affordability of higher education institutions in London in the future.

Seen this way, the outlook for UK higher education is increasingly fragile.

In the US, some institutions are doing better than others at enrollment. For many this is not news, but this graphic from Stevens Strategy is striking

In the day to day it can be hard to remember that US college costs have actually been decreasing. And, at least at highly selective private institutions, that decrease varies quite a lot by parents’ income.

And yet, despite all of this, our systems are not adapting in response to either of these shifts.

Designed for the “Average” Student

According to the New York Times.

The reason: car safety features are designed around crash test dummies modeled on young to middle age men weighing 170 lbs.

Adapted from the New York Times

This is what happens when you design for a “default” user that doesn’t reflect reality. We do the same thing in student success. We often establish baselines around a specific type of student. These averages don’t just “fail to work” for many students. They actively mislead us about where the real problems are and who is most at risk.

And when we measure, we often confuse precision with accuracy. Both are important concepts in data analytics and student success, but in many cases we are becoming more precise about the wrong things. From the incomparable Jono Hey.

Put these together, and the picture is clear. Students are changing. The environment is changing. But we are still designing systems and measures for a world that no longer exists.

We are often precise about the wrong things and less accurate about what matters. And then we call for more data.

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.

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.

Happy St. Patrick’s Day! May your bacon and cabbage be delicious.

What is new in student success? A theme ran through several of the things I read this week: how easy it is to talk about student success while quietly ignoring the structural forces shaping student outcomes.

When “college quality” really means “student wealth”

Scrolling through LinkedIn this weekend, a trenchant comment on a post by the American Enterprise Institute’s (AEI) Preston Cooper caught my eye.

It turns out that this was hardly the spiciest comment on the post. Although DRTC (Don’t Read the Comments) is generally great advice, in this case I encourage you—in fact—to go read the comments. Do not pass go, do not collect $200. Go read the comments.

But what caused all this criticism and spicy consternation? Preston Cooper’s post, on the surface, seems perfectly reasonable and something that we, as proponents of student success, would favor.

He used the post to reiterate arguments he had made previously: that enrollment declines at what he calls “low quality” institutions were a market correction. But in this post he laments the fact that they are showing signs of a rebound.

He starts off with the correction.

And then laments the fact that enrollments at “low quality” institutions are growing again

Cooper defines quality in terms of a composite index made up of four factors: the share of students who complete within 150 percent of normal time, the rate at which borrowers pay down principal, the loan default rate, and former students’ median earnings after leaving school. He weights each factor equally. At first glance, this seems like a reasonable way to define institutional quality. And who could possibly be opposed to quality? But the measures Cooper uses are not really measures of institutional quality at all. They are mostly measures of who institutions enroll.

The measures have little to say about the quality of education students receive. Not only are they more about who institutions enroll, but they are also strongly indexed to economic factors, making them even more reflective of the conditions shaping students’ lives before they enter college and after they leave.

Let’s look again at the four variables he uses: completion rate, loan repayment rate, loan default rate, and earnings. Each of these is strongly correlated with who the students are. For example, completion rates are heavily affected by:

• part-time enrollment

• transfer behavior

• adult learners

• academic preparation

Community colleges therefore appear “low quality” not because they fail their mission but because they serve students whose educational pathways often include part-time enrollment, transfer, and stop-outs—patterns that the metric treats as failure.

Loan repayment and default rates depend heavily on:

• family income

• student debt levels

• post-college earnings

Institutions serving low-income students will almost always have worse repayment metrics, even if the institution is performing well.

Earnings are strongly influenced by:

• major

• regional labor markets

• student demographics

• occupational sorting

A nursing program will always produce higher earnings than early childhood education, regardless of institutional quality.

What Cooper’s index largely measures is not institutional quality but the socioeconomic composition of the student body. Institutions that enroll wealthier, better-prepared students will almost always look “high quality” by these metrics, regardless of what actually happens inside the classroom.

Hence Kemi Jonas’s comment.

But wait, it gets worse

Even though Preston Cooper’s argument is flawed and revealing, and was rightly roasted by the commenters, it is nevertheless an example of two related things I see far too often in EdTech and student success research and commentary.

The first is the simple ignoring of externalities. In the student success and online learning literature that I follow most closely, I repeatedly see research or analysis that does not acknowledge several basic realities:

• Some students are far less well prepared than others because they attended less well-resourced schools.

• Many of these same students have far less to draw on in terms of family wealth and social networks.

• Some students have to attend part-time, work, and take care of family while attending college or university.

• And the kinds of returns students receive from their educations are heavily shaped by gender, race, and whether they enter the workforce in a high-cost-of-living area or a region where wages are lower.

Instead, student outcomes are often treated as a function simply of whether the institution was “low” or “high” quality, or of the sector or modality of instruction.

Should we hold institutions to account for poor outcomes? Absolutely. I do that frequently in this newsletter, and I try to highlight and broadcast other research and analysis that does the same. But if that analysis does not take externalities into account, it is not being honest—either with itself or with its readers.

A second and related tendency I see is what I refer to as the “caveat-and-proceed maneuver.” Analysts and researchers acknowledge a limitation, signal methodological awareness, and then continue as if the limitation does not materially affect the conclusion.

In Cooper’s original post he acknowledges that student preparedness may play a role. The emphasis is mine.

This is a very small and rote admission. But he waves off even this fleeting and partial acknowledgement.

Admitting that important factors are being left out and then using that admission as a get-out-of-jail-free card to ignore those factors while drawing bold and sweeping conclusions has become far too common in student success, online learning, and EdTech research.

These two tendencies are insidious, and should be called out every time.

A reminder to widen the lens

Related to this larger point, I was reminded recently of the need to check our assumptions.

I grew up in a pretty lower–lower-middle-class household; neither of my parents attended school beyond the age of 14. So I think of myself as fairly attuned to noticing economic and class assumptions in higher education. But a recent conversation with a friend and her husband has made me rethink that.

She and her husband both grew up in trailer parks in different parts of the Midwest, and over coffee with me recently they started talking about how there were always a couple of teachers in each school who “didn’t like poor kids.”

I asked what this meant, and they said that, for example, if you missed a day and asked the teacher if there was work to catch up on, the teachers who didn’t like poor kids would answer no. But there was indeed work, and so you would get a zero on that assignment.

Both of them ended up going to college and working in higher education. But I was shaken by this story.

Stories like this are a reminder that the forces shaping educational outcomes begin long before students arrive on campus. They are also a reminder to all of us that we need to constantly widen the lens through which we look at these issues.

The five unwritten rules of student success product names

After all that seriousness, let’s take a brief detour into something much lighter. Normally, by this point in the newsletter I’ve covered so many serious—and honestly often depressing—topics that I like to include a palate-cleansing funny post. Today, in that slot, I want to talk about student success product names.

I spend a lot of time these days looking at student success tools (and there are a lot of them). I’ve noticed that they tend to follow five key—but unwritten—rules when it comes to naming. I’ve left out actual examples, but I’m sure you can all think of some.

This is all meant in fun. I feel vendors’ pain in trying to come up with a name that hasn’t already been snagged and that will resonate with the intended audience.

Belonging vs connectedness

One place where the structural nature of student success shows up very clearly is in students’ social lives. A new survey by the College Student Mental Wellness Advocacy Coalition and the Hi, How Are You Project, highlights the importance of belonging, but the more striking signal may actually be connectedness. Thriving students report dramatically higher levels of social interaction with friends and family (81% vs. 34%), suggesting that belonging may be less a psychological state and more the result of being embedded in real social networks.

The report also reveals a large gap between awareness and help-seeking. Students overwhelmingly recognize the importance of mental health and know where to find support, yet roughly four in ten say they are uncomfortable discussing it, and nearly half fear being judged for doing so. The barrier appears to be social rather than informational.

That pattern shows up across many areas of student success. Students often avoid tutoring, advising, and other support services not because they do not know about them, but because using them signals struggle and invites stigma. If that is the case, the challenge may not simply be expanding services but normalizing help-seeking behavior.

When student success disappears from the conversation

And maybe we also need to normalize college and university presidents thinking about student success. The new Inside Higher Ed 2026 Survey of College and University Presidents is out, and it is well worth a read. But something that struck me is how oddly absent the concept of student success is. Presidents talk at length about demographic pressures, enrollment challenges, changes to DEI, institutional finances, public perceptions of higher education, and the potential of AI.

But student success itself—retention, progression, completion—rarely appears as a central concern. This is striking because many of the issues presidents highlight are tightly connected to whether students succeed once they enroll. Financial health, enrollment stability, and public confidence in higher education all depend, in part, on student outcomes. Student success shows up only indirectly—in discussions of AI tools that might support learning or in concerns about student wellbeing—but it is rarely framed as the underlying issue tying many of these concerns together.

In other words, we spend a lot of time talking about student success in higher education. But when leaders talk about institutional risk, student success itself often disappears from the conversation.

Taken together, these pieces highlight a recurring problem in higher education discourse. We talk about student success constantly, but we often define it using measures that obscure the structural realities students face, the social networks that support them, and the institutional practices that actually shape outcomes.

Until we grapple with those underlying forces, much of the student success conversation risks becoming a debate about metrics rather than about students.

Musical coda

In honor of St Patrick’s day, some Sharon Shannon, of whom I am a giant fan. The Galway Girl.

Also in honor of St. Patrick’s Day, imagine this newsletter is a snake and your job is to drive it out of Ireland.

Or at least out of your inbox, and into the inboxes of anyone who might enjoy it.

Thanks for being a subscriber.