What we use
AdmitScale is built on three public datasets every U.S. college reports to the U.S. Department of Education: IPEDS, the Common Data Set, and the College Scorecard. Each is freely available. Each tells you something different. Together, they describe roughly 6,000 U.S. institutions.
- 01
IPEDS
Integrated Postsecondary Education Data System. Enrollment counts, admission rates, financial aid distribution, completion rates, faculty data, and institutional characteristics. Published by the National Center for Education Statistics. Source of: school size, admit rate, percent of students receiving aid, six-year graduation rate.
- 02
Common Data Set
Each college's own annual self-reported admissions snapshot. Source of: mid-50% test scores (SAT/ACT), high school GPA distribution of admitted students, acceptance rate by demographic where available, yield rate, demonstrated interest tracking.
- 03
College Scorecard
U.S. Department of Education's outcomes database. Source of: post-graduation earnings by major, debt at graduation, repayment rates, completion rates by demographic.
Each number in your Blueprint is tagged to its source so you can audit it.
How we synthesize
Each of the three datasets is public, but they don't talk to each other. IPEDS uses one school identifier system. The Common Data Set is unstructured PDFs colleges post on their own websites. College Scorecard uses a third schema. Synthesizing them so they describe the same institution requires:
- 01
Normalization
We map each dataset to a common institution identifier so "University of Pittsburgh," "Pitt," and IPEDS unit ID 215293 resolve to the same school across all three sources.
- 02
Reconciliation
When two sources disagree (e.g., IPEDS admit rate vs. CDS admit rate), we prefer the more recent and more specific source, but show both when the gap is meaningful (>3 percentage points).
- 03
Profile matching
Once a unified institution record exists, we re-rank it against your profile inputs (GPA, test scores, intended major, financial profile, geographic preferences).
- 04
Re-publication
Updated monthly as new federal data publishes.
How fit scores are calculated
Each school in your Blueprint receives a "Fit Score" from 0–100. This is not an admission probability - it's a composite of how well the school matches your profile across academic, financial, and geographic dimensions. We weight it as follows:
- 45%Academic fit
How your GPA and test scores compare to the school's published mid-50% range. We use the Common Data Set as primary, IPEDS as fallback.
- 25%Financial fit
How the school's net-cost-after-aid estimate compares to your stated family budget.
- 15%Major fit
Strength and outcomes of your intended major's program at this school (faculty count, completion rate, post-graduation earnings).
- 10%Geographic / size fit
Match against your stated location and size preferences.
- 5%Scholarship likelihood
Whether the school is known to offer generous merit or need-based aid to students with profiles like yours.
Weights are fixed across users. We don't adjust them based on what you might want to hear.
How net-cost estimates work
"Sticker price" almost never reflects what your family actually pays. Net cost is closer. We estimate it using:
- The school's published net price by family income bracket (IPEDS reports this)
- The school's percent of students receiving aid and average aid amount (Common Data Set)
- Your stated family income range (from your assessment)
- Whether the school is need-aware vs. need-blind (Common Data Set)
Estimates only. The only authoritative net price is the official Net Price Calculator on each school's website, which we link to in your Blueprint. We help you narrow which calculators are worth running.
How scholarship signals work
A "scholarship signal" doesn't mean you'll get a scholarship. It means the school is statistically more likely to award one to a student with your profile. We derive this from:
- Percent of non-need-based aid awarded (IPEDS)
- Average merit aid amount for students in your academic bracket (CDS)
- Whether the school's published admission rate exceeds 50%, indicating selective-recruitment behavior
- Geographic and demographic preferences disclosed in the CDS
Signal, not guarantee. Always apply if the school is on your list - these scholarships are typically awarded automatically based on application.
Why this is hard to find elsewhere
- 01
Three sources, one view
IPEDS, the Common Data Set, and College Scorecard each tell you something different - admissions selectivity, cost, outcomes. Reading them separately requires hours per school. We unify them.
- 02
Monthly synthesis layer
The federal data refreshes on its own schedule. Our synthesis is updated monthly, so the picture stays current without waiting for the next annual federal release.
- 03
Profile-matched, not generic
Most public data is presented as one-size-fits-all averages. We re-rank everything against your profile, intended major, and financial goals. The dataset is public; the personalized lens is not.
What we don't claim
Plain English about our limits, because hiding them is what makes other tools feel like sales pitches.
- 01
Essay quality
We can't evaluate your writing. A great essay can move you across a school's mid-50%. We don't see it.
- 02
Recommendations
We can't see your teachers' or counselors' letters. They matter.
- 03
Demonstrated interest
Some schools track campus visits, email opens, and event attendance. We can't.
- 04
Demographic context
Admissions readers consider context (first-gen, underrepresented backgrounds, school context). We don't model this.
- 05
Institutional priorities
Each year, every school has shifting internal priorities (geographic balance, donor relationships, athletic recruits). They're invisible to outsiders.
- 06
The reader's mood at 11pm in February
Admissions decisions are made by humans under fatigue. We can't model this either.
Estimates only. Not a guarantee of admission.
When the data is wrong
Federal data has lag. A school's true admit rate this year may differ from what's been reported. The Common Data Set is self-reported and occasionally inflated. We mitigate this by:
- Refreshing monthly so we catch updates as they publish
- Showing both IPEDS and CDS values when they disagree by more than 3 percentage points
- Flagging schools where data is more than 2 years old
- Linking directly to each school's official source so you can verify
If you spot data that looks wrong, email counselor@admitscale.com - we update within 1 business day.
Refresh cadence
Our synthesis layer updates monthly. Underlying federal data follows its own schedule:
- IPEDS: annually, typically published in spring
- Common Data Set: annually, colleges typically publish theirs in late fall
- College Scorecard: annually, usually winter
Your Blueprint reflects the most recent data we have on the day you build it.