Falling completions do not mean a dying field
87% of US programs with falling completion counts face growing occupational demand. A falling award count is not evidence of a dying field.
Academic program assessment • a True Heading Collective service
Know which programs are actually heading somewhere.
Tell us a program. We return a written assessment of its viability—assembled from seven public datasets spanning completions, occupational projections, wage records, and earnings outcomes, with every figure traced to its source and vintage.
The premise
The data is the input. The interpretation is the product.
Why this exists
Program decisions are usually made on a hunch and a spreadsheet tab.
The evidence needed to judge a program already exists and is free: federal completions data, occupational projections, earnings records. It is also scattered across three agencies, published in incompatible formats, and revised without notice. So the decision defaults to whoever has the strongest opinion in the room.
This service does the assembly and, more importantly, the reading. A completions decline is not automatically bad news, and growth is not automatically good news. Which of those is true depends on what the occupation is doing and whether the degree is the route into it.
Read the analysis: 87% of programs with falling completions face growing occupational demand →
What the report contains
A written assessment, not a data dump.
Each report runs a few pages and follows the same structure, so two programs can be compared directly.
1. The assessment
A single word, then the reasoning behind it. Reports land in one of four states: healthy, conditional, declining, or insufficient evidence. That last one is a real outcome—when the data cannot support a judgment, the report says so instead of inventing one.
2. Why this assessment
Plain-language reasoning that connects the numbers. For example: completions fell 15.3% over two years while interpreter employment grew 3.5%—the occupation is expanding, but the degree is not the route into it.
3. The evidence table
Every figure that shaped the assessment, each with its source and vintage: national awards conferred, two-year and eight-year change, occupational demand, state-level demand, the share of mapped employment that requires a bachelor’s, and the licensure picture.
4. Where these jobs are
Employment and median wage for the mapped occupations across every metro area in your state, with a location quotient showing whether the area holds these jobs at the national share or more. A quotient of 1.50 means half again as concentrated as the country as a whole. These are current job counts from May 2024, not a forecast, and the report labels them that way.
5. Openings against graduates
Annual openings in the mapped occupations set against the awards conferred nationally by every program that feeds those same occupations. This is the supply-and-demand balance of the field. A field can be growing and still be oversupplied, and this is the figure that shows it.
6. What we did not score
Some figures inform the reading without moving the verdict—state demand, institution counts, metro concentration, the openings ratio, your own enrollment. These are tagged context only, so you can see exactly what did and did not influence the outcome.
7. Your inputs, stated back
Facts only your institution holds—whether the program is a licensure gate in your state, roughly how many students are enrolled—are recorded separately and marked as yours, not inferred.
8. The limits, up front
A frank section on what the assessment cannot tell you, including the model’s measured agreement with expert judgment on programs it had never seen. It is not buried in a footnote.
The four outcomes
Healthy
Conditional
Declining
Insufficient evidence
There is deliberately no 0–100 score. A single number implies a precision the underlying data does not have, and it invites a decision that the number cannot actually support.
The analyses
We also publish national program profiles for twelve of the largest US programs — completions, occupational demand, graduate earnings, and state-level projections, each with its classification stated.
What the national data actually shows.
Three analyses run across every US program where both an IPEDS award series and a BLS occupational projection exist. Each is reproducible from the public sources named at the foot of the page.
Programs are not closing. They are getting smaller.
Of 61 US programs with falling award counts, 32 had the same number of institutions or more. The decline is inside the program, not the map.
What a program review actually needs to know
A program review needs five questions, not a score: the completions trend, whether the occupation is growing, and whether the degree is the route in.
Where the data comes from
One report. Seven public datasets. Every figure traceable.
A single assessment joins records across the National Center for Education Statistics, the Bureau of Labor Statistics, the state projections system those agencies fund, and the Department of Education’s earnings collections. Nothing is estimated, imputed, or filled in. Where a source publishes nothing for a program, the report says not published rather than substituting a plausible number.
6,096official program-to-occupation mappings
2,325six-digit program codes searched per request
393metro areas checked for concentration
868occupations with employment and wage detail
The source ledger
Program supply — who is awarding what
- IPEDS Completions Awards conferred by program, award level, and institution, 1983–2023. Supplies the program’s national size, its two-year and eight-year trajectory, and the count of institutions offering it. The 2023 collection is provisional and labelled as such. National Center for Education Statistics
- CIP Code Taxonomy The full 2,325-code Classification of Instructional Programs, with titles, definitions, and cross-references. This is what turns a typed program name into a defensible code rather than a guess. National Center for Education Statistics
Occupational demand — where the graduates would go
- Employment Projections Ten-year occupational growth, annual openings, and — the decisive field — the typical entry-level education each occupation requires. Growth is weighted by employment across every occupation a program maps to. Bureau of Labor Statistics, 2025–2035
- Occupational Employment & Wage Statistics Current employment, median and mean wages, and location quotient at national and metro resolution. The location quotient is what reveals whether an occupation is more concentrated near a given campus than it is nationally. When you give a state, the report shows the metro areas within it. Bureau of Labor Statistics, May 2024
- State Occupational Projections Long-term projections for 55 state and territorial areas across 863 occupations — 70,546 records. Reported as a separate dimension so a program’s national trend is never mistaken for its local one. Projections Managing Partnership / Projections Central, 2024–2034
The join — how a degree connects to a job
- CIP–SOC Crosswalk The official mapping between instructional programs and occupations: 6,096 mappings spanning 2,143 programs and 868 occupations. This is the mechanism that lets the report say whether a degree is actually the route into its own mapped occupations. National Center for Education Statistics
Earnings — what graduates actually make
- College Scorecard Median earnings by field of study at one, four, and five years after entry, with the middle fifty percent reported alongside. Shown as context, not scored — and only where enough institutions report to make the figure meaningful. U.S. Department of Education
What no dataset can tell us
- Your two answers Whether the program is a licensure requirement in your state, and roughly how many students are enrolled. No federal dataset encodes either one, and the licensure answer materially changes whether falling completions mean lost demand or a constrained pipeline. Both are recorded as yours, never inferred. Supplied by you
Every figure in the report carries its source and its vintage, because these datasets are not the same age. Completions run through 2023; employment actuals are May 2024; national projections cover 2025–2035 while state projections cover 2024–2034. Where two sources disagree — and they do, on median wages — the report shows both rather than silently picking one.
How it works
Three steps, and one of them is yours.
01
You name the program
A plain name is enough—“Spanish,” “Cybersecurity,” “Nursing.” You do not need a CIP code. Where a name is ambiguous, we show you the candidate programs and you pick rather than guessing on your behalf.
02
We assemble and read the evidence
Seven public datasets are queried and joined: federal completions records, the full program taxonomy, national and state occupational projections, employment and wage statistics down to metro level, the official program-to-occupation crosswalk, and field-of-study earnings. The result is interpreted against a scoring model tuned on expert judgment.
03
You get a written report
Delivered by email as a document you can forward, print, or attach to a curriculum committee packet—with sources cited so any figure can be checked independently.
What this is not
Decision support. Not a verdict.
This matters enough to state plainly, and it is repeated in every report we send:
- It is not a recommendation to close, expand, or retain a program. It is evidence assembled and interpreted, so that the people accountable for the decision can argue with the reasoning instead of with each other.
- The model is a screening rule, not a validated predictor. Its measured agreement with expert judgment on programs it had never seen is stated in each report rather than implied.
- Completions data are national. Your program’s local situation may differ substantially, and state-level demand is reported separately for that reason.
- Occupational projections are not graduate outcomes. A growing occupation does not guarantee that your graduates enter it, or that they enter it with this credential.
- No live job-posting data. Current openings are not part of the assessment. That is a known gap, and we would rather name it than paper over it.
- AI is used to generate these reports, and no person reviews them before they are sent. The figures are pulled and computed from the public datasets listed above, and every one is traceable to its source. AI assistance is used to assemble and draft the written interpretation around those figures. Reports are not checked by a person before delivery, so verifying any figure or claim before you rely on it is your responsibility, not ours. Treat the prose as a starting point for your own reading of the evidence, not as an authority.
Request a report
Tell us the program you are weighing.
Leave your email and the program name. We send a confirmation link, then generate the report and email it to you as a PDF. The two optional questions below change the reading when you answer them—no national dataset carries either one—but the report is produced without them.
The first report is free. After that a report is $49 at the introductory price, paid by card. No account, no subscription, no procurement. The price will rise as coverage grows.
Prefer email?
bt@trueheadingcollective.com