The Overall Rating
The number on the dial is the average of exactly two ratings — nothing else is mixed in.
Illustrative example
Enjoyment and Career Prospectsare the only two dimensions where higher is unambiguously better, so they're the only two averaged into a single quality number. Difficulty, workload, math, and writing don't belong here — a heavy math load isn't bad, it's just heavy. That's what the Academic Fingerprint is for.
Would Recommend
The share of students who'd choose this major again — reported on its own, never folded into the Overall Rating.
Illustrative example
We keep it separate for two reasons. First, it's a yes/no proportion, not a 1–5 rating, so averaging it in would mix scales. Second, it's the single most persuasive number a review produces, and blending it away would waste that. It gets the same small-sample treatment as every other score — see Small Samples below.
The Academic Fingerprint
Difficulty, workload, math, and writing intensity — shown as a profile, never averaged into a score.
Illustrative example
None of these have a “good” direction. Heavy math is a selling point to one student and a dealbreaker to another, so instead of scoring it, we show it. These four dimensions power the radar chart, interactive positioning maps, and school-vs-school comparisons — but they never touch the Overall Rating.
Percentiles
“Top 25%” badges compare a major only against others with enough reviews to rank fairly — and we only ever show the flattering half.
Illustrative example — a major at the 75th percentile
Ranked pool
Only majors with at least 5 reviews are ranked. Fewer than that and there isn't enough signal to place fairly.
Flattering only
A badge only shows when a major beats at least half the field. We'd rather show nothing than a discouraging number.
Per-dimension
Enjoyment, Career Prospects, and Would Recommend are each ranked separately — a major can lead on one and trail on another.
Keeping Small Samples Honest
A 5.0 from two reviews at one school isn't better than a 4.6 from two hundred. We blend small samples toward a steadier baseline instead of trusting them outright.
displayed score = (n × school score + 10 × major average) ÷ (n + 10)n = reviews for that major at that specific school
Illustrative example — a major averaging 3.6 overall, where one school's reviews so far are a perfect 5.0
Blends toward the major
Every school-specific score starts anchored to that major's overall average, then shifts toward its own reviews as they accumulate.
“Early data” label
Cells under 10 reviews are marked so you know the score still leans heavily on the major-wide average.
Review counts, always
Every score is shown next to how many reviews back it, so you can weigh the evidence yourself.