Every college on this site shows a UCD Score between 30 and 100. It sits at the top of every college profile, drives the default sort on the colleges archive, and appears as a column in the Compare tool. The score is built from federal data that the US Department of Education already collects on every Title IV-accredited US college, and the full formula is published in code that anyone can read. This guide explains how the score is calculated, why it works the way it does, and what it deliberately does not measure. The purpose is not to defend the score but to make it legible, so a visitor can decide for themselves whether a UCD Score of 78 at one school means more than a 73 at another.
Definition
UCD Score
A number between 30 and 100 representing a college's relative performance on four federal outcome metrics within its peer group. The score is a transformed percentile: a school at the 50th percentile of its peer group scores 65, the 90th percentile scores 93, the 10th percentile scores 37. Five label tiers summarize the range: Excellent (85+), Strong (70–84), Good (55–69), Fair (40–54), Average (below 40).
Why a Single College Score Is Hard to Build Honestly
Building a single score that works honestly across all 3,839 accredited US colleges requires solving one structural problem first.
The Problem
Comparing a selective research university to an open-admission community college on a single linear scale produces numbers that reward prestige and resources, not outcomes. Most published scoring systems ignore this or hide their methodology behind paywalls.
The Approach
Three peer groups, each scored separately. Each college is ranked by percentile within its own group, then mapped to the 30–100 scale. A 75 in Peer Group A means the school outperforms 75% of similar four-year selective colleges on the four sub-scores.
The Data
Five federal sources: College Scorecard (earnings + completion rates), IPEDS (institutional characteristics), Bureau of Labor Statistics (job growth), O*NET (career attributes), and the CIP-SOC crosswalk linking programs to occupations. Full citations on the data sources page.
The Result
A single score with a public formula, built entirely from federal data, calibrated within peer group rather than against all colleges. The methodology page documents every calculation step for anyone who wants to verify the numbers.
The Three Peer Groups and Why They Exist
The UCD Score uses three peer groups, labeled A, B, and C, to separate institutions whose populations and economics are different enough that a single comparison would be misleading.
A
Four-year selective
Bachelor's degrees as the primary credential, acceptance rate below 100%. Covers highly selective privates, mid-size public flagships, and small liberal arts colleges. All four sub-scores apply. The largest of the three groups.
B
Four-year open or online
Accept rate of 100% or admissions data not reported. Includes most for-profit colleges, online-first universities, and some open-admission publics. Higher proportion of part-time and working-adult students. Selectivity sub-score not included.
C
Two-year
Community colleges, technical colleges, and institutions whose primary credential is an associate degree or certificate. The IPEDS six-year graduation rate undercounts success here because it was designed for four-year cohorts. Selectivity sub-score not included.
The peer-group structure means a UCD Score is always a comparison within group, never across group. A 75 in Peer Group A means the school outperforms 75 percent of similar four-year selective institutions on the four sub-scores. A 75 in Peer Group C means the school outperforms 75 percent of similar two-year institutions on the four sub-scores. The two numbers communicate similar relative position, but they are not directly comparable in absolute terms. The cluster spoke on peer groups and why we score within them covers the peer group definitions in further detail and explains the edge cases.
The Four Sub-Scores That Build the Total
The UCD Score is built from four sub-scores. Each reflects one dimension of institutional performance that the federal data lets us measure consistently across schools within a peer group.
| Sub-score | What it measures | Applies to |
|---|---|---|
| Outcomes | Completion rate + first-year retention | Peer Groups A, B, C |
| Value | Earnings-to-cost ratio at 10 years | Peer Groups A, B, C |
| Affordability | Average net price relative to peer group | Peer Groups A, B, C |
| Selectivity | Admission rate within peer group | Peer Group A only |
This sub-score is built primarily from the federal completion rate (the percentage of first-time full-time students who complete a credential within 150 percent of normal time) and the federal retention rate (the percentage of first-time students who return for a second year). Outcomes is the most direct signal of whether students who enroll at an institution actually finish. The score is calibrated to the peer group, so a 70th-percentile Outcomes score in Peer Group C reflects strong performance against other two-year colleges, even though the absolute completion rate is lower than the equivalent percentile in Peer Group A.
Value is the ratio between earnings outcomes ten years after entry and the cost of attendance during the student's time at the institution. The earnings figure comes from IRS tax records linked to federal student-aid data, so it reflects what graduates actually earn rather than what surveys claim. The cost figure includes sticker price for full-pay students and net price across income tiers. A high Value score means graduates earn back the cost of attendance quickly relative to peers. A low Value score does not necessarily mean the institution is bad, but it does mean the financial return on the credential is weaker than peers within the same group.
Affordability is calculated from the average net price across all income tiers, weighted to reflect typical student family income distribution at the institution. This is distinct from Value (which compares cost to earnings) in that Affordability is purely about the up-front sticker. A school with high Affordability is inexpensive relative to peer institutions, regardless of the earnings outcome. Some schools score high on Affordability and lower on Value, which usually means they are inexpensive but the earnings outcomes lag the peer group. Some schools score lower on Affordability and high on Value, which usually means they are expensive but the earnings premium more than recovers the cost.
This sub-score applies only to Peer Group A, where acceptance rate is meaningful. It reflects the school's position on the admit-rate distribution within the four-year selective peer group. Selectivity is included not as a measure of institutional quality (a low admit rate does not mean a school is better) but as a signal of student-population characteristics that interact with the other three sub-scores. A highly selective institution will tend to have higher Outcomes and Earnings simply because the students admitted started with stronger academic preparation. Including Selectivity in the score makes this visible rather than hiding it inside the other dimensions. Peer Groups B and C do not have a Selectivity sub-score because the underlying data is not meaningful for those populations.
The four sub-scores are combined with weighted averaging to produce the total UCD Score. The exact weights are published on the methodology page and can be reweighted by the visitor on the Compare tool when comparing schools head-to-head.
How the 30 to 100 Scale Is Calibrated
The raw output of the percentile calculation is a number between 0 and 100, where 50 means median performance within the peer group. The UCD Score remaps this raw percentile into a 30 to 100 scale to produce a more readable headline number.
The transformation is straightforward. The score is calculated as 30 plus 0.7 times the raw percentile. So a school at the 50th percentile of its peer group on the combined sub-scores produces a UCD Score of 65. A school at the 90th percentile produces a 93. A school at the 10th percentile produces a 37. The transformation compresses the bottom of the distribution and stretches the top, which makes meaningful differences in the high-performing range more visible than they would be on a raw 0 to 100 scale.
The five label tiers attach human-readable interpretations to score bands and appear on every college profile alongside the numeric score.
| Label | Score range | Approx peer percentile |
|---|---|---|
| Excellent | 85–100 | Top 20% within peer group |
| Strong | 70–84 | 60th–80th percentile |
| Good | 55–69 | 35th–60th percentile |
| Fair | 40–54 | 15th–35th percentile |
| Average | Below 40 | Bottom 15% |
Colleges with fewer than 500 enrolled students are capped at a UCD Score of 80, regardless of where their percentile calculation lands. This cap exists because federal data for very small institutions is statistically noisy. A school with 200 students that happens to have one cohort with strong outcomes can produce a percentile score that does not reflect institutional performance. The cap acknowledges the noise while still rewarding small institutions with genuinely strong sub-scores. Without the cap, a handful of micro-institutions would dominate the high end of the score distribution in ways that would mislead users.
How to Read the Score on Any College Profile
The UCD Score appears in three places on every college profile. The top pill in the hero shows the headline score and the label. The sidebar shows the four sub-scores so visitors can see which dimensions drive the total. The full breakdown section in the body of the profile lists the percentile position within peer group for each sub-score, along with the raw underlying data (cost, completion rate, earnings).
-
Check the peer group first. A 78 in Peer Group A and a 78 in Peer Group C communicate similar relative position (around the 70th percentile within group) but absolute comparison between them is not meaningful. The peer group label appears next to the score on every profile.
-
Check the sub-score balance. A college with a high Outcomes score and a low Value score is performing well on completion but charging a premium the earnings do not recover. High Value with low Outcomes means strong earnings for the students who finish but too many dropping before they get there. High Affordability with low Outcomes is inexpensive but underperforming on completion. The sub-score pattern often communicates more than the total.
-
Check the underlying data. The score is a percentile mapping, so it tells you the school's position relative to peers, not the absolute level. A 65 in Outcomes in Peer Group C might correspond to a 35 percent completion rate, which is median for two-year colleges but would be a yellow flag at a four-year institution. The completion rate, earnings, and net price are all visible on the profile below the score.
Three sub-score patterns come up often enough to have a short decision rule attached to each:
- Strong Outcomes, weak Value: the school completes students at a high rate but the cost-to-earnings ratio lags the peer group. Worth attending if the net price for your family is within cap and the major is strong. Run the Cost Calculator for your income tier before deciding.
- High Affordability, weak Outcomes: inexpensive relative to peers but a higher dropout risk. Check the raw completion rate in the underlying data before shortlisting. If it sits below 50 percent, treat that as a yellow flag regardless of the total score.
- Strong Value, lower Affordability: expensive relative to peers, but graduates recover the cost faster. A reasonable bet if the family can absorb the net price; the ROI Calculator can model the long-term math.
Reading the score this way takes 90 seconds per profile after the first one. The cluster spoke on reading a College Scorecard page covers how to interpret the federal data that feeds the score and where the federal numbers themselves have caveats.
See a worked example
Representative example. Ridgeline State University, Peer Group A (four-year selective). UCD Score: 72, label: Strong. Sub-scores: Outcomes 68, Value 76, Affordability 63, Selectivity 54. Underlying data: six-year completion rate 69%, earnings ten years after entry $62,000, average net price $21,000.
Step 1, peer group check. Peer Group A: this school is compared against other four-year selective colleges. A score of 72 means it outperforms roughly 60 percent of that peer group on the combined sub-scores.
Step 2, sub-score balance. Value is the strongest sub-score at 76. This means earnings-to-cost ratio is in the top 24 percent of four-year selective schools, so the school earns back its cost faster than most peers. Outcomes at 68 is above median but not a standout. Affordability at 63 is slightly above median, not a bargain but not expensive either. Selectivity at 54 is near the peer-group median; the school admits roughly half of applicants.
Step 3, underlying data check. 69 percent six-year completion rate is solid for Peer Group A, where the peer median sits around 68 percent. $62,000 in earnings is healthy. $21,000 net price at a four-year selective institution is relatively affordable. The strong Value sub-score is driven by this combination: moderate cost + solid earnings = fast payback.
Pattern match. This is the "Strong Value, moderate Affordability" pattern from the decision rules above. The school is not the cheapest option but it delivers strong returns on what it charges. A good candidate for a student whose major is available here and whose family net price estimate from the Cost Calculator lands within the 10–15 percent income rule.
Note: these numbers are representative, not pulled from a specific institution in the database.
Common Mistakes in Interpretation
The five mistakes below appear most often in conversations about the score with students and parents who are encountering it for the first time.
Comparing scores across peer groups. A 75 in Peer Group A and a 75 in Peer Group C are not equivalent absolute numbers. They communicate similar relative position within different reference populations. Comparing them as if they were identical is a category error.
Treating the score as a quality rating. The score is a relative position metric within a peer group. It is not a single answer to "is this a good college." A college with a UCD Score of 60 in Peer Group A might be the right fit for a specific student because of major availability, location, or financial considerations that the score does not capture.
Ignoring the sub-score breakdown. Two schools with the same total UCD Score can have very different sub-score profiles, and the right school for a specific student often comes down to which sub-scores are stronger. The Compare Colleges tool shows the four sub-scores side by side, which makes the differences visible.
Confusing Selectivity with quality. Selectivity is a sub-score for Peer Group A only and reflects admit rate. A low admit rate does not mean a school is better. It often means the school has a strong brand that attracts more applicants than it can admit. The cluster spoke on why acceptance rate is overrated covers what acceptance rate actually predicts and what it does not.
Assuming the score is the final answer. The score is a summary metric. It is built from federal data, but it does not capture every relevant factor in the college decision. Major availability, geographic preference, campus culture, religious affiliation, athletic programs, and many other factors live outside the score. The pillar guide on building your college list covers how to use the score as one filter among several.
Treating the UCD Score like a US News ranking. US News weights inputs: acceptance rate, peer reputation surveys, faculty resources, financial resources. Those factors reward institutional prestige and wealth. The UCD Score weights outputs: completion rates, earnings relative to cost, and affordability. A school that ranks 200th in US News can score Excellent on UCD if it produces strong outcomes for the students it actually admits. The reverse is also common: a school with a strong brand and high US News rank can have a modest UCD Score because the earnings outcomes relative to the premium price disappoint. The two metrics are measuring different things, and using them interchangeably produces wrong conclusions.
Avoiding these six mistakes makes the score useful rather than misleading. The score is a starting point for comparison, not a ranking of college quality.
What the Score Doesn't Measure (and Why)
The UCD Score is built from the federal data that exists for every accredited US college. That data has its own known limits, and the score inherits them. The IPEDS completion rate counts only first-time full-time students, which means institutions serving larger proportions of transfer students and part-timers will tend to show lower Outcomes scores than their actual student success rates would support. The College Scorecard earnings figure covers only students who received federal financial aid, which skews the sample toward middle-income graduates. These limits are disclosed in full on the data sources page. They are worth keeping in mind especially for Peer Group B and C institutions, where non-traditional enrollment is higher.
The score also does not include factors that are not measured consistently across all institutions, factors that are subjective, or factors that would introduce bias into a peer-group comparison.
| Not measured | Why excluded | Where to find it |
|---|---|---|
| Student satisfaction / campus culture | No consistent federal measure across all schools; partial data would penalize non-participants | Student reviews on individual college profiles |
| Specific major or program quality | Score is institution-level only; within-school program variation is not captured | Program profiles + college profile "What You Can Study" section |
| Selectivity vs rankings-publisher weights | Admit rate is used as a population signal, not a quality weight | Acceptance rate on every college profile |
| Athletic, artistic, extracurricular strength | Not in federal data | Fit factors on college profile; school websites |
| Recent institutional changes | Federal data has a 1–2 year reporting lag | "Last reviewed" date on every college profile |
The exclusions are deliberate. The score is built to be a relative position metric on factors that can be measured the same way for every accredited college. Including subjective or partial measures would make the score less reliable, not more comprehensive. The factors above matter for the college decision. They just live outside this particular metric, and the rest of the data on the site is built to surface them where they apply.
The UCD Score works as one filter among several. The pillar guide on building a college list covers how to combine the score with cost, fit, and major-specific data to produce a defensible shortlist. The score is the starting point. The rest of the analysis is the work.
