Index solves grade inequality problem

At the meeting last Thursday of the Arts and Sciences Council there was an almost complete disconnection between the presentation of the proposal for the Achievement Index and the statements made in opposition. Professor Val Johnson gave a clear and complete explanation of the plan's rationale, its method and its implications. This was followed by a long series of prepared statements by people who seem not to have heard a word he said. What we got instead was a string of red herrings. I will comment here on only one of them, the matter of grade inflation.

Grade inflation is not the issue addressed by the AI. Carried to its logical conclusion, grade inflation would result in a system with only one passing grade (presumably "A") and the need for the AI would disappear. The issue is inequality of grading standards, the lack of any consistency among instructors as to what the various grades signify, which results in gross unfairness to students. The only purpose of the AI is to correct for the distortion this produces in class rank.

That this distortion is significant is shown by comparing class rank based on the grade point average with class rank based on the AI. At the very top and bottom of the ranks there is little difference, but for a very large number of students the GPA rank is quite different from the AI rank. If one grants that the AI does indeed correct for inequality of grading standards among instructors -and those who do not grant it have an obligation to show why not-then this disparity between the rankings reveals a significant problem of fairness to our students.

In the face of this evidence, a responsible faculty cannot simply turn away and ignore the problem. There are three alternatives: radically change the grading system; deal with the problem at the source by imposing uniformity of grading standards on all instructors; or deal with the problem retrospectively by taking account of the biases statistically with something like the AI. Given that the first alternative is unlikely and the second unworkable, adoption of the AI seems to me clearly the best way to obtain a fair ranking of our students.

Lawrence Evans

Professor

Department of Physics