Grading curve

The grading curve is a system of grading students relative to the performance of their peers.

Method

A grading curve is created by graphing the frequency distribution of marks earned by students. Typically, there will be a cluster of popular marks, with exceptional students above and below - thus creating a curved graph. Grades are then set so that the most popular mark will result in a middling grade, whereas unusually high marks will attract higher grades and unusually low marks will attract lower grades.

(This needs an explanation of what "marks" are - it sounds like you get a grade based on your grade - not possible?)

Benefits

• Automatically allows for difficulty of tests. The grading curve tends to automatically compensate for the test that was used to obtain the marks being unusually easy or unusually hard. If the majority of students have high (or low) marks then the middling grade will be adjusted there and higher or lower grades awarded based on this performance.

• Reduces difficulties with grade margins. The grading curve eliminates the problems that may arise if a student's score falls near to a grade margin. Clustering of marks will have been explicitly examined to establish where the margin should be placed.

Problems

• Someone always does badly. There must always be at least one student who has a lower score than all others, even if that score is objectively quite high. Thus, grading on the curve guarantees that some students will always do badly, or fail, even if they have objectively performed will on the test.

• A student's performance is affected by their peers. For example, a student could - in one year - be in a class of high performers and recieve a low grade. Then, next year, the student could try the same course again and deliver exactly the same performance but recieve a high grade because the others in the new class did worse.