← Back to the Proportional / Ratio Rescaling calculator
Proportional / Ratio Rescaling
“The Fair-to-Effort Curve”
Scores are scaled proportionally so the highest reaches the maximum. A common variant restores part of the points each student missed: New = Old + (100 − Old) × 0.5.
curved = score × (max / highestScore) Best for: Keeping relative gaps The scenario
- The quiz was too difficult; several strong students only hit 65–70%.
- Struggling students scored 40–50%, showing some understanding.
- The highest score was 72%.
- You decide to give back 50% of the points they missed.
How to justify it
- This method rewards effort proportionally.
- A student who got 50% right gets back 25% of the lost points → 75%, reflecting more learning than the raw score shows.
- A student who got 80% right gets back 10% → a small boost that still reflects stronger performance.
- The top performer stays the top performer.
Data to show your administrator
- The formula in plain English: "I am restoring half of the points they missed."
- An anonymized before/after list: Student A (40% → 70%), Student B (65% → 82.5%), Student C (72% → 86%).
- The equity angle: it rewards students who showed understanding without handing full points to those who guessed.
When to use this justification
- You have a mix of weak and strong students and the test was too hard for the whole class.
- The administrator values equity (fairness to effort, not equality).
- You can articulate the formula clearly.
Likely pushback — and how to answer it
- “Why 50% of missed points and not 75%?”
- This percentage is based on how much the test underestimated their knowledge — or it follows our district standard.
Tone to strike
Growth-minded, equity-focused — "I’m restoring points to fairly represent effort and understanding…"
Try the Proportional / Ratio Rescaling curve →
Frequently Asked Questions
What is the core principle behind ratio curving?
It makes the highest score 100%, then scales all other scores proportionally — answering, "if the best possible score was 85%, what does that mean for everyone?"
Can you walk through a ratio curving example?
Highest score 80% → ratio 100 ÷ 80 = 1.25. A 80% → 100%, a 64% → 80%, a 48% → 60%. Rankings stay the same while the spread grows proportionally.
How is ratio curving different from linear curving?
Linear adds fixed points (everyone +10); ratio multiplies proportionally (everyone ×1.25). Ratio preserves the proportional spread between scores.
When should I use ratio instead of square root?
Use ratio when you want the highest score to hit exactly 100% and everything to scale accordingly. Use square root when you want larger boosts specifically for lower scores.
Is ratio curving fair to all students?
Yes. It treats the highest score as a reference and scales everyone proportionally, so no student benefits more relative to their starting position.
What if the highest score is already 100%?
Then the ratio equals 1.0 and nothing changes. Ratio curving only adjusts scores when the highest is below 100%.
Can I combine ratio curving with adjusted grade boundaries?
Yes. Apply the ratio curve first to adjust raw scores, then apply your grade boundaries based on your institution’s standards.
How do I calculate the ratio adjustment manually in Excel?
Use =A2 / MAX(A:A) * 100 — divide each score by the maximum score, then multiply by 100 for the new percentage.
What are the limitations of ratio curving?
It’s less intuitive than linear, assumes the highest score is a valid reference (not luck), isn’t based on statistics, and can cause large adjustments when the top score is far below 100%.
Can I use ratio curving for partial-credit or cumulative grades?
Yes. Apply it to any numeric scale — points, percentages, or weighted grades — as long as you’re scaling to a known maximum.
Still have questions? Get in touch.