Grade Curve Calculator

Adds the same fixed number of points to every student's score. The boost equals the difference between the desired top score (usually 100%) and the highest raw score in the class. If the top score was 88% and the target is 100%, every student gets +12 points. Simple, transparent, and easy to explain.

Treats the highest score in the class as the new 100% and scales everyone else proportionally. The top performer always ends up at exactly 100%; a student who scored half of the top score ends up at 50%. Best when you want the relative ranking between students preserved exactly.

Takes the square root of each score and multiplies by 10. This gives a bigger boost to low scores and a smaller boost to high ones — a 49% jumps to 70%, while a 90% only rises to 94.9%. Often used when a test was harder than intended and many students scored in the 50–70% range.

Shifts the entire distribution so the class average lands on a target value (commonly 75%). Every student gets the same additive adjustment — the shape of the distribution is preserved. Unlike pure statistical bell-curve normalisation, this simple version doesn't change the spread, only the centre.

What does "grading on a curve" mean?

Grading on a curve means adjusting students' raw scores so they reflect relative performance within the class rather than an absolute percentage standard. It is typically used when a test turns out to be harder than intended, resulting in a lower-than-expected class average. Curving can take many forms — from simply adding a flat number of points to more complex statistical adjustments.

Which curving method should I use?

It depends on your goal. Use Linear Rescaling when you want to reward the top performer with a perfect score while boosting everyone equally. Use Ratio Scaling when you want to preserve exact relative rankings. Use the Square Root method when lower-scoring students need a bigger boost than high scorers. Use the Bell Curve / Mean Shift when you want the class average to hit a specific target without changing who ranks above or below whom.

Does curving grades change who ranks first in the class?

Linear rescaling, square root, and bell curve (mean shift) all preserve the original ranking order — a student who scored higher before the curve still scores higher after. The ratio method also preserves rankings. Only methods that normalise using individual z-scores (not implemented here) can reorder students.

Is the square root curve fair?

The square root curve is considered fair in situations where a test was unusually hard because it gives more relief to students who struggled most. However, it can feel arbitrary to high-achieving students whose scores change less. Whether it is "fair" depends on the context and the instructor's philosophy about grading.

What is the difference between the bell curve method here and statistical bell-curve grading?

Statistical bell-curve grading maps scores to a normal distribution using z-scores, forcing a predetermined percentage of students into each letter grade. The bell curve method in this calculator is a simpler "mean shift" — it moves the whole distribution up or down so the average hits a target, without forcing a fixed grade distribution or reordering students.

Can a curved score exceed 100%?

In this calculator, all curved scores are capped at 100%. With the linear method, if a student's raw score plus the boost exceeds 100%, their score is capped. This is the standard approach in most classroom settings — extra credit is a separate decision.

How many students can I enter?

There is no hard limit. Use the "+ Add student" button to add as many rows as needed. The calculator re-runs all curves instantly as you type. For very large classes, you may find it more efficient to use a spreadsheet — but for up to 50 or so students, this tool works well.