Programs are not closing. They are getting smaller.

When a program's award count falls, the first question a review committee asks is usually whether the field is going away. The second is whether the program is being wound down. Both are reasonable, and the public data can test both — because IPEDS records not only how many credentials were awarded, but how many institutions awarded them.

Those are two different numbers, and they behave very differently. Across the 61 programs whose award counts fell between 2021 and 2023, the number of institutions offering them barely changed. The number of graduates per institution fell sharply.

That distinction matters for what a review should do next. A field losing providers is a question about the map. A field keeping its providers and losing its students is a question about demand, marketing, curriculum, and cost per graduate — a different set of questions with different answers.

What the two counts measure

The award count is the total number of credentials conferred nationally in a program. It is the product of two things: how many institutions offer the program, and how many graduates each one produces.

The institution count is the number of institutions conferring at least one award in that program in that year. It nets new programs against discontinued ones. If the count is flat, entries and exits roughly cancelled — it does not mean no institution closed a program, and it does not mean none opened one.

Dividing the first by the second gives average awards per conferring institution. That is the number that tells you whether a decline is about reach or about depth.

Totals across the 61 programs with falling award counts, 2021 to 2023. Institution counts are summed across programs, so an institution offering three of these programs counts three times.

Decomposed on a log basis, 98% of the average decline across these 61 programs comes from the per-institution term. The institution term is essentially zero.

Where the decline actually sits

Thirty-two of the 61 programs — 52% — had the same number of institutions or more in 2023 than in 2021. In that group, awards fell 5.2% while institutions rose 3.8%. More institutions were teaching the program, and between them they produced fewer graduates.

The eight largest per-institution declines in that group:

Award and institution change: IPEDS completions, all award levels, 2021 to 2023. Per-institution figures are total awards divided by the number of institutions conferring at least one award.

Actuarial Science is the extreme case. It lost 38% of its awards while adding institutions — the average conferring institution went from 18.3 awards to 10.9. Pharmacy lost 22% of its awards and added institutions, going from 109.3 to 84.5 per institution. Neither of those is a program being withdrawn. Both are programs that are still widely offered and are producing roughly a third fewer graduates each.

Four programs in this group lost more than 20% of their per-institution output while the institution count grew: Actuarial Science, Nutrition, Pharmacy, and Chemical Engineering.

Is this a two-year artefact?

Two years is a short window, and 2023 completions are provisional. So the same test was run on three different windows, and the pattern holds in all three:

• 2021 to 2022. Of 59 programs with falling awards, 45 — 76% — had the same number of institutions or more.
• 2021 to 2023. Of 61 programs, 32 — 52% — did.
• 2019 to 2023. Of 49 programs, 22 — 45% — did.

Over the longer eight-year window the same group behaves the same way: across the 32 programs, awards grew 6.2% and institutions grew 7.3%, so per-institution output fell 1.0%. Twenty-six of the 32 had more institutions in 2023 than in 2015. The programs are not being abandoned. They are being spread thinner.

The one place institutions really did leave

There is a counter-case, and it is worth naming because it shows what the opposite pattern looks like. Architecture is the only program in the set where the institution count fell materially faster than the award count — 11.0% against 7.4% over two years, and 30.5% against 55.4% over eight.

That is what a field contracting at the point of provision looks like: institutions stop offering it, and the remaining ones are left carrying a smaller total. Over eight years the contrast is stark — awards down 55.4% and institutions down 30.5%. Per-institution output did not collapse over the two-year window (41.0 to 42.7, a slight rise), so this is not a clean case of consolidation into fewer, larger programs. But the direction of the institution count is the signal that separates it from the other 60.

Two other programs — Theatre and Medical Laboratory Science — also had institution counts fall faster than awards, but by less than two percentage points each. They sit on the boundary rather than clearly across it.

Why this changes the review question

A review that reads only the award count asks "should we keep this program?" That is the wrong question for 52% of these cases, because the program is not being withdrawn anywhere — it is being kept and producing fewer graduates everywhere.

The better question is why the yield per institution fell. That has several possible answers, and they lead to different decisions:

• Demand shifted. Fewer students are choosing the field. This is the case the occupational data speaks to, and it is the one most likely to be a genuine field-level problem.
• The credential stopped being the route in. The occupation may be growing while employers hire through a different credential. This is the pattern the earlier analysis found in Spanish and Journalism.
• Competition inside the market. More institutions offering the same program, splitting a fixed pool of students. The institution count rising while awards fall is exactly this signature.
• Institutional choices. Advising, scheduling, prerequisites, or a shift to a related program can all reduce yield without any change in the field.

Only the first two are arguments about the field. The third is an argument about the local market, and the fourth is an argument about the institution. A review that cannot tell them apart is guessing.

What the data cannot tell you

Four limits bound what any of this can support.

• These are national totals. Your program's own institution count and yield may look nothing like the national pattern. A national figure cannot establish that a specific program is weak or strong.
• The institution count nets entries against exits. A flat count is consistent with several closures offset by several new programs. It is not evidence that no institution discontinued the program, and it cannot tell you which ones did.
• Per-institution output is an average, not a distribution. A program whose national average fell may contain institutions that grew. The average says nothing about any single case.
• None of this measures quality, cost, or mission. A program can be small, thinning, and essential. Nothing here speaks to whether it should exist.

This is decision support, not a decision. It is a way to ask a better question at the start of a review, not a way to end one.

For the wider method behind this — the five questions a review needs, and why we do not compress them into a single score — see what a program review actually needs to know.

Check your own program

We produce a written viability assessment for a single program: completions against occupational demand, the institution count and per-institution yield, state-level projections, graduate earnings where they exist, and every figure traced to its source and vintage. The first one is free.

Request an assessment for your program →