College Salary and Graduation Data: What the Numbers Leave Out
College earnings and graduation figures are useful only when you know whom they describe. Keep the population, measurement period, and definition beside the number before comparing colleges.
Use this page if outcomes numbers are shaping a college choice; leave with a comparison row that names each figure’s population, measure, period, and missing information.
A higher reported salary does not prove that attending one institution would increase a particular student's earnings by the same amount. Differences in fields of study, student backgrounds, location, and the people included in a measure can affect the comparison.
Identify the level of the data
The federal College Scorecard documentation distinguishes institution-level data from field-of-study data and explains the timing of its measures. Read the definition for the specific field you use; do not assume every figure includes the same students or the same period after college.
The Scorecard's field-of-study documentation describes a different level of aggregation and its own limitations. The institution-level documentation explains that some earnings measures concern federally aided students who are working and not enrolled at the measurement time. That is narrower than “all graduates,” and the field-level data should not be silently substituted for an institution-level result. Write the variable definition beside the value before drawing a conclusion. If two figures refer to different populations, the right output is two labeled observations, not a winner.
For a college's own outcomes survey, look for the response rate and whether graduate study, employment, service, and people with unknown outcomes are counted separately. “Successful outcomes” needs a definition before it can be compared with another school's employment rate.
Build a fair comparison row
| Record | Question to answer |
|---|---|
| Program or institution | Is this the student's intended field or the whole college? |
| Cohort | Who was included, and who was excluded? |
| Time period | How long after entering or completing college was the outcome measured? |
| Measure | Is this a median, average, percentage, or survey response? |
| Missing information | Are values unavailable, suppressed, or based on a small group? |
Leave an unavailable number blank. It is not a zero, and it is not evidence that a college performed poorly.
For a worked comparison, imagine College A publishes institution-wide median earnings six years after entry, while College B advertises a survey of employed graduates one year after graduation. Both may be accurately reported, but the populations and clocks differ. Put them in separate rows and mark “not directly comparable.” Ask whether either school publishes the same measure for the student's intended field and whether the group is large enough for a reported figure. This method prevents a compelling headline from becoming a false numerical contest.
Read earnings alongside cost and access
Consider a hypothetical comparison: College A reports higher institution-wide earnings, while College B offers the student's intended field at a substantially lower family cost. Before deciding, check whether College A's mix of majors explains part of the difference and whether comparable program data exist.
Then investigate the mechanisms the student could use: access to required courses, internships, advising, and support when progress slows. A salary statistic does not tell you whether a first-year student can enter the desired program.
Add the family's likely net price and the student's program-access path to the same decision sheet, but do not convert all three into a homemade score. In the example, College A's higher institution-wide earnings might reflect its mix of majors, while College B may offer a lower estimated cost and a clearer route into the field the student wants. The next useful questions are whether that field is open to incoming students, what prerequisites apply, and what help exists if a required course is full. Outcomes data can flag a concern; it cannot answer those access questions.
Treat completion rates as a prompt for questions
Check whether a graduation measure follows first-time full-time students, another cohort, or a particular time horizon. Ask about degree planning and bottlenecks instead of interpreting a percentage as a promise about your own graduation date.
You can use outcomes data to identify a concern, but resolving that concern may require a department or advising office to explain the actual student path.
A completion-rate edge case illustrates the limit. A student who enters part time, transfers, or pauses enrollment may be counted differently from the cohort behind a prominent four-year figure. Before using a rate as a forecast, ask which students are counted, when the clock starts, and whether transfers or other paths are visible in that measure. Then ask the college how students in the intended major sequence their requirements. The family is looking for a plausible support and degree path, not a guarantee from an aggregate percentage.
Attach a definition to every number you save
A usable note says more than "earnings: higher." Record the source, year, population, field or institution level, measure, and time after enrollment or completion. Put a question mark beside a value whose cohort or reporting period you cannot identify. If one college reports a survey of employed respondents and another reports an administrative median for a different group, leave the figures in separate rows until you understand the difference. Do the same for graduation rates: write down who entered the cohort and what counts as completion. This keeps an attractive number from silently becoming a claim about the student's own likely outcome.
Ask your student to spend 15 minutes making one outcomes row for a college: source, report year, population or cohort, field of study, measure, period, and missing-data caveat. Leave any unknown field blank and label it as a question, rather than silently treating the numbers as comparable.
Combine the evidence with the college research template. If your family is weighing a significant cost difference, bring the definitions and the unresolved questions to an IvyReady consultation, alongside the student's academic priorities.