curve

[GoesStation]

Surely you've heard of "grading on a curve" before? Say you get an 75 on a test. On a straight scale (90-80-70-60), 75 would be a C.


If graded on a curve, it depends on class performance. A 75 is bad if the class average is 85. It's good if the class averaged 60.

The phrase is always grading on the curve in my experience.

[Maybo]

The phrase is always grading on the curve in my experience.

What is the difference between "a" and "the"?

[GoesStation]

The expression we're talking about is always "grading on the curve", as far as I know. It's a fixed phrase.

[probus]

Is that the way things naturally occur, or is it something else?

It is very much the way things naturally occur. When the great mathematician Gauss first derived the equation for the standard normal distribution (the now-famous bell curve) he was studying the errors that occur in repeated meaurements of the same physical quantity.

[Tarheel]

Yes. In a random sample you can expect to get so many A's, so many B's, so many C's, so many D's, and so many F's. The biggest cluster will be in the C group. I don't think that's the same as grading on the curve. In that case, those who get the highest grades get A' s regardless of their test scores. (Correct me if I am wrong.) Any single class may have an unusual number of below average students or above average students (either by accident or design).