How Scores Work

Every school is scored across six fundamental dimensions of the student experience.

When you rate a school, you answer branching slider questions. Each answer routes you to deeper follow-up questions based on your response. If you rate teaching poorly, we ask what specifically is wrong. If you rate it highly, we ask what makes it great.

This branching system means we collect specific, diagnostic data — not just “4 out of 5 stars.” Every answer maps to an atomic scoring dimension. We have 84 such dimensions for schools and 80 for teachers.

Scores are computed bottom-up through four layers:

Recent ratings matter more. We apply an 18-month half-life decay — a rating from 18 months ago counts half as much as one from today. This means scores reflect the current state of the school, not ancient history.

Every score has a confidence level based on four factors: sample size, freshness, diversity (different year groups and time periods), and consistency (how much raters agree).

We use device fingerprinting and behavioural analysis to prevent fake ratings. Duplicate submissions are blocked. Suspicious patterns — uniform answers, robotic timing, burst activity — are flagged and down-weighted. Structured sliders are inherently harder to game than text reviews.

Teachers are scored on a separate set of six dimensions: Clarity, Engagement, Fairness, Support, Control, and Impact. Teacher scores are only shown after 10+ ratings to protect against small-sample bias.

School profile data (Ofsted ratings, exam results, pupil numbers) is displayed alongside student scores for context. This external data never affects the Student Score. The two are completely separate.

All ratings are anonymous. We don't collect names, emails, or any personal information. Device fingerprints are stored as one-way hashes and automatically deleted after 90 days.