“How did you do on your test?” my friend inquired. It is a simple question and it deserves a simple answer. Unfortunately, I am unable to give him one. Most of the grading I have encountered in engineering has been completely incomprehensible. It has been based on curves, multipliers, averages, and who knows what other hocus-pocus. Standardization may not be easy, but it is sorely needed.

Let us look at a materials course I had last semester as an example. On one test, I received a score of 38.4 out of 70 points, with some points being free. I will not attempt to explain the basis because I do not understand it myself. Seeking more information, I played Sherlock Holmes and found that my 55 percent was slightly above average. My professor could have legitimately complimented me, “You got an F! Good job!” On another test—take-home this time—I received an A-, and I was thrilled. I played sleuth again and found that most people had received A’s. I wondered, did I do well on that test or not? After this and a number of homework assignments, graded out of 28 total points, the course was over. I had absolutely no idea of how I had performed. If one laid out all of the possible final grades—A, B, C, D, and F—I would have assigned them all equal probabilities. This is totally unacceptable. Students have the right to know how they are doing without ambiguity and guesswork.

Nowhere else would you find this phenomenon and the effects are grossly negative. There is a good reason why SAT scores mean the same thing, year after year. When I entered this school as a freshman, I was very serious about my grades and objective performance. Now a senior, RPI has so desensitized me to curves and subjectivity in grading that it has become irrational for me to dwell on it. These days, when I get 47.6 out of 84 points on a test, I just laugh at the absurdity of the evaluation and place it aside.

There is no valid reason why a curve should consistently be used in any course. Besides the confusion it causes, the use of a curve is blatantly unfair. Why should a student’s performance be based arbitrarily upon how the other students perform in a given semester? Courses should be based on a total of 100 points with 10-point downward increments, regardless of performance distribution. Any deviation indicates a questionable and ultimately lazy grading scheme. Even if it takes many years of practice, all professors should be expected to develop a reasonable standard with which to judge students, based upon individual student performance and not upon peer performance, 20-sided dice, or the phases of the moon. This can and should be done in higher-level engineering courses because I have seen very reasonable standardized scales employed in physics, humanities, and core engineering courses at RPI with success. Kudos to them.