grading on the curve

     Grading "on the curve" is familiar to most college and high school students. Instead of a standard that defines the limit of each grade--90 for an "A", 80 for a "B", for instance--raw scores are "adjusted" to fit a particular model. The model may be shaped like a bell--as in the well-known bell curve. The philosophy behind the bell curve is that most scores are average. The next largest "groups" of scores would be above and below average. The edges of the bell shape would be the exceptional scores--whether exceptionally good or exceptionally bad. In a class of 30 students, the bell curve may be used to adjust scores so that 3 people always get an "A", and 3 people always get an "F". The above and below average groups may contain 6 people who get a "B", and 6 people who get a "D". The remaining 12 people will get  a "C".
     Using "grading on the curve" the distribution of scores remains the same, even when all or most of the students being tested get a "raw score" of  below 50%--or above 90%. This is an extreme model--most teachers who use a "curved" grading system only do so when many students are failing. Failing students may simply have points added to their test scores, instead of having the scores adjusted to fit a model. They may be failing because the test was poorly designed, or because the material on the test has not been taught in class.
     Students who have been too often "graded on the curve" may be surprised when they encounter a standardized test. Their teachers may be surprised as well.