Long: Grading system needs overhaul
January 19, 2012
What is your cumulative GPA? For the average student, if the true idea of “average” holds, it would be about a 2.0. See, there’s a problem in that statement. Average at school is no longer average. The “average” grade, if we had one now, would be a B, at minimum. In fact, in 2008, nearly 43 percent of grades given in American universities were A’s. About 35 percent were B’s. That means that around 78 percent of grades given were rated as some order of “above average.”
As one of those students who probably has been rated as “above average” at some point in your academic career, does it make any logical sense that 78 percent of anything can be above average? It doesn’t make sense to me either.
Why does this matter, you may ask? Well, your GPA is vitally important when it comes to competing for jobs in a difficult market, or applying for post-graduate school of any kind. If you compete for any type of scholarships during college, a minute differential in GPA can be the difference in thousands of dollars.
To solve this problem, we need to look at some basic statistical principles. Your cumulative GPA is basically a statistical representation of your performance over time. It measures your performance against an assumed “average” performance standard. It just is a bit flawed, currently and needs adjusting.
To do this, I need to make a few things clear. I am intending that the GPA would be used to compare students against each other. In a competitive world, where thousands of people compete for a few seats in law and medical school or jobs on the job market, the comparison is necessary.
And second, for this to work, everyone would have to readjust what they perceive as normal. Whereas now, students cannot graduate Iowa State with less than a 2.00, it would have to be changed. With the GPA, 2.00 would be very close to the mean performance of all students at Iowa State. Thus, Iowa State, along with all job markets and post graduate institutions, would also need to readjust their standards.
Now that that is out of the way, I suppose I should explain how the grade would be figured. It is relatively simple, using a well-known statistical measure. I propose that the standard deviation of the class be used.
Assuming a normal distribution from the mean, the standard deviation shows accurately how close to the true median a value is. According to the three-sigma rule, under a normal distribution, 68.7 percent of all values in the distribution falls within one standard deviation (positive or negative) of the mean. An additional 13.6 percent falls on each side within the next standard deviation, and with the rest being beyond that.
So, what if we called everything within the first deviation some form of a C (the exact grade point would vary with the distance from the mean, so a value very near the first deviation on the high side would get closer to a 2.66, and one close to the lower deviation would have a value closer to 1.66). The next deviations on each side would be the bounds of the next letter grade.
To be clear, this would not assign a set number of A, B, and C grades to be handed out. It simply would compare the total points you earned against the average points earned in the class, and assign a value to it.
Now, of course that could be adjusted (a C could go from +.5 to -.5, a B could go from +.5 to +1.5, and an A could be everything above +1.5, for example). That would make it easier to earn an A, but also easier to fail.
Of course, one problem with this would be that you can only measure against other students who took your exact course. This could have a great effect in one class, but over the course of a collegiate career, it would even out.
It also would place an even greater emphasis on the reputation of the academic institution you go to, the perceived difficulty of the courses there and the quality of students admitted.
But it would give a greater representation of how you performed during your career against students who were placed in the exact same courses in the same session. It needs some tweaking, but it’s worth consideration. My GPA would be worse, but it would be a better representation of my performance.
No matter the solution, system changes are required. When 78 percent are counted as above average, average becomes meaningless.