Where Graduates Work Is Not the Same as Where Jobs Are

Rural workforce planning often collapses three different maps into one: where jobs exist, where graduates work, and where learners can realistically reach training. Census data can illuminate the first two. It cannot answer the third without local evidence. In our view, a real Iowa example shows why the distinction matters—and how a college can use it to make a better program decision.

The planning mistake: treating one map as the whole market

A county can have a large healthcare workforce while graduates from a nearby college disperse across a much larger region. A college can retain many graduates in-state while still placing a program beyond the practical reach of working adults in its rural service area. Each fact can be true at the same time.

In our view, starting with a broad demand question is too blunt. Program teams should separate at least three decisions: whether an industry has enough regional employment to deserve attention, whether graduates from a relevant field enter and remain in that industry, and whether prospective learners can access the proposed schedule and location. The datasets are inputs to those decisions, not substitutes for them.

A real example: Kirkwood health professions graduates

The Census Bureau’s Post-Secondary Employment Outcomes program publishes experimental earnings and employment-flow statistics for graduates of participating institutions. The data can be filtered by institution, degree level, field of study, graduation cohort, destination industry, and employment geography.[1]The current Iowa institution file includes Kirkwood Community College.[2]

Consider one bounded question: Where did Kirkwood associate-degree graduates in the health professions field work one year after graduation? In the current Iowa PSEO employment-flow file, the all-cohort record reports 4,853 employed graduates. Of those, 3,824 worked in healthcare and social assistance, 4,395 worked in Iowa, and 4,497 worked in the West North Central Census division.[3]

These ratios are calculated from the published counts above. They are descriptive, not causal. The all-cohort code pools the available five-year associate-degree graduation cohorts; it does not describe one graduating class.[4]The release covers Iowa graduates from 2001–2021 and was produced for 2026 Q2.[5]

The result is still useful. It shows a strong relationship between the broad field of study and the healthcare industry, along with substantial in-state retention among the graduates represented. In our assessment, it does not show that every program within CIP family 51 performs alike, that employers currently have openings, or that a new rural training location would enroll enough learners.

Pair graduate destinations with the regional job base

Quarterly Workforce Indicators can supply the other half of the picture. QWI includes 32 indicators covering employment, job creation and destruction, wages, hires, and other employment flows. It can be tabulated by county, industry, worker demographics, and employer characteristics.[6]

For an Eastern Iowa planning exercise, a college could compare the same healthcare industry measure across the counties it intends to serve. Keep the quarter, industry code, ownership category, and indicator constant. That comparison can show whether the employment base is concentrated in one county or distributed across the region. We advise treating it as regional context, not as a count of open jobs or a forecast of student demand.

The combination sharpens the question. We read the PSEO counts as evidence that the represented health professions graduates commonly entered healthcare and remained in Iowa. QWI can show where healthcare employment currently sits across the service region. The next decision is operational: which employers, learners, schedules, clinical placements, and travel patterns would make a particular offering viable?

A 90-minute implementation exercise

1. Define one field and one decision. We recommend narrowing the decision to whether a rural nursing site is warranted instead of beginning with every possible regional program.
2. Choose a consistent QWI view. Select the service-area counties, one industry code, one quarter, and the employment or earnings indicators relevant to the decision.
3. Choose a comparable PSEO view. Select the institution, degree level, field, cohort window, and one- or five-year outcome. We recommend recording whether the institution is represented.
4. Write down what each dataset cannot establish. QWI does not prove openings or employer commitments. PSEO does not prove that a new delivery model will reproduce historical outcomes.
5. Turn the gaps into fieldwork. Interview employers about roles and clinical capacity. Test schedules, transportation, and price with prospective learners. Review faculty and facility constraints internally.
6. We recommend a decision rule before collecting more data. Define what evidence would support a pilot, a redesign, or a stop.

What the data should change

The point is not to produce a more impressive dashboard. It is to change the next conversation. If graduates remain in-state but employment is concentrated far from the proposed site, the college may need a partnership or hybrid-delivery strategy rather than another fixed location. If the regional employment base is strong but graduate flows leave the field, curriculum alignment, job quality, or employer practice may deserve attention. If both signals are weak, the responsible decision may be to investigate a different field.

In our view, program teams can be tempted to make one dataset carry the entire case. Resist it. Regional jobs, graduate destinations, and learner access are connected, but they are not interchangeable. A credible proposal makes each claim at the level the evidence can support—and names the local evidence still needed before commitment.