Grading on a curve means adjusting students’ scores relative to how the whole class performed, rather than using a fixed percentage scale alone.
Teachers usually curve when a test turned out harder than planned. The word “curve” comes from the bell curve, but most classroom curves are simpler than that.
Three common methods, one test
Suppose the highest score on a test is 88 and your raw score is 70.
1. Flat curve. Add the gap between the top score and 100 to everyone. 100 − 88 = 12, so 70 becomes 82.
2. Top-score scaling. Multiply every score by 100 ÷ 88 (about 1.136). 70 × 1.136 = 79.5. Lower scores gain fewer points than with a flat curve.
3. Bell curve (norm-referenced). Grades are assigned by rank: for example, the top slice of the class gets A’s, the next slice B’s, and so on. Your 70 could be an A or a C depending on how classmates scored. This is where percentile rank comes in.
Other variations include the square-root curve (10 × √score, so 70 becomes about 83.7) and dropping the hardest question.
Try it with real scores
The grade curve calculator applies each method to a full class list. To convert a raw count like 21 out of 30 into a percent first, use the test score calculator, then check where the curved percent lands on your grading scale.
The fairness question
A fixed-scale grade measures what you know against a standard. A bell-curve grade measures you against classmates. That difference matters when talking about grade inflation and when a single letter grade decides eligibility for honors or scholarships.