Students generally assume that curving means an upward adjustment of low test scores, but the basis of the practice derives from assumptions about statistical distributions of scores (bell curve).
A simple method for curving grades is to add the same amount of points to each student's score. A common method: Find the difference between the highest grade in the class and the highest possible score and add that many points. If the highest percentage grade in the class was 88%, the difference is 12%.
Per definition, grading by a curve usually means that the students are assigned grades based on the statistical distribution of the test/exam results. No matter what, say 20% of students will always fail, and only say 10% will get a perfect mark.
Gradually, grading on a curve improves students' grades by raising their actual scores by a few notches, probably improving a letter grade. Several teachers use curves to adjust the tests' scores, while others choose to adjust the letter grades assigned to existing scores.
The Downsides of Grading on a Curve However, if they were in a class of 40, curving will only allow eight people to get A's. This means that it's not enough to get a grade of 90 and above to get an A; if you get a 94 and eight other people get higher, you end up getting a grade lower than you deserve.
When courses are graded on a curve there is less interaction between students, hence less learning. We have measured that there is also generally overall lower motivation.
Grade curving in university is a practice designed generally to protect a median of letter grades, meaning that the majority of the class would get B's or C's and smaller percentages would receive A's, D's, and F's.
Many professors curve their exam in some way to alter the grade distributions. Some professors think this isn't necessary until the end of the semester and "curve" by altering cutoffs rather than adjusting final averages. Sometimes you, as a student, deserve the grade you got.