Grade Curve Calculator - Curve Test Scores

Paste a set of test scores, pick a curving method, and see the before/after statistics and every adjusted score. Free and browser-based, no scores leave your device.

8 valid scores (0-150)

Curving method

Class statistics before and after the curve

Mean

75.5

86.7

Median

74.5

86.3

Min

58.0

76.2

Max

91.0

95.4

Per-score mapping

OriginalCurvedChange
72.084.9+12.9
85.092.2+7.2
64.080.0+16.0
91.095.4+4.4
58.076.2+18.2
77.087.7+10.7
69.083.1+14.1
88.093.8+5.8

Three Common Ways to Curve a Test

A flat add gives every student the same number of extra points, often used when one question was flawed and everyone deserves its value back. Adding 5 points moves a 58 to 63 and a 91 to 96: the whole distribution shifts up but the gaps between students stay identical. It is transparent and easy to defend, which is why many instructors default to it.

The square root curve (curved = 10 x the square root of the original score) helps weaker scores far more than strong ones. A 36 becomes a 60, a 64 becomes an 80, and a 100 stays a 100. It compresses the distribution from below, which makes it popular after a test that turned out much harder than intended. The downside is fairness optics: the student who scored 36 gains 24 points while the student who scored 81 gains only 9.

Scaling to the highest score treats the top student's result as the new 100%. If the best score was 88, every score is multiplied by 100/88, so a 66 becomes 75. This method assumes the top performance defines what was realistically achievable, but note that a single outlier ruins it: one student scoring 98 on a test where everyone else scored below 70 produces almost no curve at all.

Whichever method you pick, compare the before/after mean and median here to sanity-check the effect. A curve that moves the class mean from 68 to 76 is a very different intervention from one that moves it from 68 to 92. All calculations run locally in your browser, so pasting real student scores is safe. For per-student grading, the Test Grade Calculator converts wrong answers to percentages, and the Final Grade Calculator projects what students need on the next exam.

How the Grade Curve Calculator Works

  1. 01Paste the class scores, one per line or comma separated.
  2. 02Choose a curving method: flat add, square root curve, or scale to the highest score.
  3. 03Compare the before and after statistics: mean, median, minimum, and maximum.
  4. 04Review the per-score mapping table to see exactly what each original score becomes, optionally capped at 100.

Choosing the Right Curve

A flat add shifts everyone equally: +5 turns a 58 into 63 and a 91 into 96, preserving all gaps, and is the standard remedy for a flawed question. The square root curve (10 x the square root of the score) compresses from below: 36 becomes 60, 64 becomes 80, 100 stays 100, so it rescues weak scores after an unexpectedly hard exam. Scaling to the highest treats the top score as the new 100%: if the best result was 88, every score is multiplied by 100/88. Watch out for outliers with that method, since a single 98 in a weak class produces almost no curve.

When to Use This Calculator

Use it when a test's raw distribution does not reflect what the class actually learned: a mean far below your other sections, a question that turned out ambiguous, or material that was not covered as planned. Run all three methods on the real scores and compare the resulting means before committing; moving a class mean from 68 to 76 is a very different intervention from moving it to 92. It is a decision-support tool for the instructor, not a policy.

Common Use Cases

  • Rescue a class after an exam that was harder than intended by testing a square root curve before applying it. → Test Grade Calculator
  • Give back the value of a flawed question with a flat add and verify the new mean is reasonable.
  • Compare all three curving methods on real scores to pick the one with the intended effect on the class mean.

Tips

  • Run all three methods on the same scores and compare the after-mean before deciding.
  • Check the maximum before scaling to highest; one outlier can neutralize the whole curve.
  • Announce the curve method to the class in the same terms used here; transparency prevents most grade disputes.

Frequently Asked Questions

What is a square root curve?→
Each score is replaced by 10 times its square root: 49 becomes 70, 64 becomes 80, 81 becomes 90, and 100 stays 100. It boosts low scores far more than high ones, which compresses the distribution upward. It is popular after exams that turned out much harder than intended.
Which curving method is fairest?→
Each has a different fairness logic. A flat add treats every student identically in points. The square root curve helps struggling students most, which some see as fair rescue and others as rewarding lower performance. Scaling to the highest anchors on peer performance. The transparent choice is to state the method openly and apply it to everyone.
How does scaling to the highest score work?→
Every score is multiplied by 100 divided by the class maximum. If the top score was 88, the factor is 1.136: a 66 becomes 75 and the 88 becomes 100. Beware of a single outlier: one 98 in an otherwise weak class shrinks the factor to 1.02 and the curve barely helps anyone.
Should curved scores be capped at 100?→
Usually yes, and the cap is on by default. A flat +8 would push a 95 to 103, which most gradebooks reject. Capping slightly compresses the top of the distribution: with +8 capped, a 95 and a 92 both become 100. Uncheck the cap if your grading system accepts scores above 100 as extra credit.
Does curving change the median and mean the same way?→
Not always. A flat add moves mean and median by exactly the same amount. The square root curve moves the mean more than the median when the class has many low scores, because those scores gain the most. Comparing both statistics before and after, as this tool shows, reveals how the curve reshapes the distribution rather than just shifting it.

Related tools

12 suggested

From the blog

Further reading