Spike Rankings 2027 · Version 1.0
How Spike Ranks Colleges
Everything that decides a rank is on this page: the parts, the weights, the sources, and the rules for missing data.
Who is ranked.
Spike ranks the 233 colleges in its public guide, not every college in America. 232 have enough data for the Best Colleges list. A college missing too much data for a list is left off it.
Every number comes from federal reporting, read at one data year per table so no list compares two years. Source years differ by dataset and are listed below. The published data snapshot is rebuilt when the site is deployed. The method changes only with a new version on this page.
Which data years are used?
Each dataset uses the year with the most college coverage. The guide’s update date is separate from these reporting years.
| Dataset | Reporting year |
|---|---|
| Graduation, retention and faculty | 2024–25 |
| Cost, aid and debt | 2024–25 |
| Earnings | 2024–25 |
| Student background | 2024–25 |
The five parts of Best Colleges.
Each input is turned into a percentile among the ranked colleges, so a college at 0.9 beats 90% of them. For inputs where lower is better, like price, the scale is flipped. A part is the weighted average of its inputs. If an input isn’t reported, its weight is shared among the others, and a part needs at least half its inputs to count. A college needs parts worth at least 80% of the weight to be ranked.
| Part and input | Weight | Source | Reported by |
|---|---|---|---|
| OutcomesDo students finish, and how do they do after? | 35% | ||
| Graduation rate | 14% | IPEDS Graduation Rates, via College Scorecard | 233 of 233 |
| Return for year two | 7% | IPEDS Fall Enrollment, via College Scorecard | 233 of 233 |
| Earnings after 10 years | 8.8% | College Scorecard (Treasury earnings data) | 231 of 233 |
| Earnings after 6 years | 5.3% | College Scorecard (Treasury earnings data) | 232 of 233 |
| LiftHow students do compared with what their starting point predicts. | 20% | ||
| Graduation vs expected | 10% | Spike’s model, from the inputs above | 232 of 233 |
| Earnings vs expected | 7% | Spike’s model, from the inputs above | 231 of 233 |
| Return rate vs expected | 3% | Spike’s model, from the inputs above | 232 of 233 |
| AffordabilityWhat families pay after grants, and what students borrow. | 20% | ||
| Average net price | 7% | IPEDS Student Financial Aid, via College Scorecard | 232 of 233 |
| Net price, income under $30k | 6% | IPEDS Student Financial Aid, via College Scorecard | 224 of 233 |
| Typical federal debt | 4% | College Scorecard (federal loan data) | 231 of 233 |
| Meets full need | 3% | The college’s own financial aid pages | 73 of 233 |
| AcademicsSmall classes, money spent on each student, and strong classmates. | 15% | ||
| Students per faculty | 5.3% | IPEDS, via College Scorecard | 232 of 233 |
| Endowment per student | 5.3% | IPEDS Finance | 188 of 233 |
| Median SAT | 2.3% | Common Data Set or IPEDS Admissions | 178 of 233 |
| Acceptance rate | 2.3% | IPEDS Admissions, via College Scorecard | 233 of 233 |
| OpportunityWhether students from every background are enrolled. | 10% | ||
| Students with Pell Grants | 6% | IPEDS Student Financial Aid, via College Scorecard | 232 of 233 |
| First-generation students | 4% | College Scorecard | 228 of 233 |
What Lift measures.
Some colleges enroll students who would likely graduate anywhere. Lift asks how students do compared with what their starting point predicts. It’s the same idea as the graduation-rate performance measure used by other guides, extended to earnings.
Spike fits a regression across the ranked colleges. It predicts each outcome from the share of students with Pell Grants, family income, acceptance rate, public or private control and enrollment. Earnings also account for the share of degrees in STEM fields, so an engineering school isn’t credited for its major mix. Lift is the gap between the real number and the prediction.
| Outcome | Colleges | Share explained (R²) | Predicted from |
|---|---|---|---|
| Graduation rate | 232 | 0.82 | Pell Grant share, family income, acceptance rate, public or private, enrollment |
| Return for year two | 232 | 0.76 | Pell Grant share, family income, acceptance rate, public or private, enrollment |
| Earnings after 10 years | 231 | 0.65 | Pell Grant share, family income, acceptance rate, public or private, enrollment, share of STEM degrees |
A positive Lift doesn’t prove a college caused the gain. It shows students there do better than similar students elsewhere.
What Spike leaves out, and why.
- Reputation surveys. They measure fame, and fame follows old rankings. Spike uses none.
- Data colleges submit to Spike. Every input is federal reporting or the college’s public pages.
- Most of selectivity. Acceptance rate and SAT together are 4.5% of the score. Rejecting more students shouldn’t raise a rank.
- Money from colleges. Colleges can’t pay to be ranked, to move, or to be left off. Sponsored content never touches these lists.
How sure a rank is.
Any set of weights is a choice. So Spike re-runs every multi-part list 200 times, each time shifting the weights at random around its published ones. The range under a rank is where the college landed in 80% of those runs. Lists that rank on one number, like net price, are exact and show no range.
Equal scores share a rank (1, 1, 3). Names only set the order between tied colleges.
Every list.
- Best Colleges. Outcomes 35%, Lift 20%, Affordability 20%, Academics 15%, Opportunity 10%.
- Best Value Colleges. Months of pay to cover four years 45%, graduation 25%, debt 15%, average net price 15%.
- Lowest-Cost Colleges. Ordered by reported average net price, lowest first. Equal prices share a rank.
- Best Colleges That Admit Most Applicants. The Best Colleges order, kept to colleges with an acceptance rate of 50% or higher. Includes colleges with: acceptance rate of 50% or higher.
- Best Colleges for Financial Aid. Price for families under $30k 35%, grants from the college 25%, debt 15%, average net price 15%, meets full need 10%.
- Best Colleges for Outcomes. The Outcomes part on its own: graduation 40%, earnings after 10 years 25%, return for year two 20%, earnings after 6 years 15%.
- Colleges That Beat Expectations. Graduation, earnings and return rates compared with what the college’s intake predicts. Graduation 50%, earnings 35%, return rate 15%.
- Best Colleges for First-Generation and Low-Income Students. Pell Grant students 25%, price for families under $30k 25%, graduation 20%, first-generation students 15%, graduation vs expected 15%.
- Best colleges by field of study. The Best Colleges score 75%, program size 25%. A college needs at least 25 bachelor’s degrees a year in the field.
Filtered views (public, liberal arts, a state) keep the national order. A college’s place in a state list matches its national rank among the colleges in that state. A filtered view publishes only with eight or more ranked colleges.
What a rank is not.
A rank describes a college, not your chances there or whether it fits you. Averages hide a lot: your net price depends on your family, and earnings depend on your major. Use a ranking to find colleges worth a look, then check each one against your own numbers.
Corrections.
Found a number that looks wrong? Email help@get-spike.com with the college and the figure. Spike checks it against the federal source and fixes errors in the next update of the rankings.