Curve applier
Straight from a spreadsheet column is fine. Non-numbers are ignored and counted.
Off by default so the raw arithmetic stays visible.
Per-student before and after
| Before | After | Change | Letter |
|---|
Letter distribution, before and after
Counted on the common plus-minus default scale (A 93 / A- 90 / B+ 87 … D- 60, F below).
| Letter | Before | After |
|---|
Questions
Which curves does this tool apply?
Six: add a flat number of points; multiply every score by a percent; shift so the top score becomes 100; rescale so the top score becomes 100; the classic square-root curve (10 times the square root of the score); and shift the class mean to a target you choose. Each shows exactly what it did in the result line.
Why does the top-to-100 curve sometimes refuse to run?
When the top score is already 100 or more, shifting or rescaling it to 100 would lower every score. The tool refuses and says so instead of quietly curving downward. The target-mean curve is the exception: it may shift down, because you named the target - and the result says the shift was downward.
Do square-root and rescale curves treat students equally?
No, and that is the point of showing the per-student table. The square-root curve lifts low scores most and high scores least; the rescale curve multiplies, so it hands the most points to the highest scores. Same roster, opposite distribution of the help.
What does the cap toggle do?
After the curve is applied, any score above 100 is trimmed to 100. It is off by default so the raw arithmetic stays visible. Capping slightly lowers the after mean whenever anything hits the ceiling.
Do the scores I paste go anywhere?
No. There is no server and no storage - the math runs in your browser and the numbers exist only on this page until you close it.
Which scale does the letter column use?
The common plus-minus default (A at 93 down to D- at 60, F below). It is fixed on this page to keep the table readable; the percent-to-letter converter has the fully editable version of the same scale.