Relatively Speaking…

100marks.jpgVery few people know about this, and those who do are very surprised but the IITs are one of the few colleges in the country which use a system of relative marking. Since that itself is not very informative, this article is all about relative marking, or grading as some people like to call it and of course, its pros and cons.

First, about the method itself. Relative, of course is the antithesis of absolute and the same applies here. Absolute marking, something most of us have had to put up with throughout school and probably still are in college is straightforward, you answer questions, score marks, give many exams and patiently wait for the result. All your examiners have to do is total all the marks, divide it by a scaling factor and give you a nice (or not, depending on how hard you’ve worked), round total out of 100. This ‘new’ system is pretty much the same except this time the examiner is hard-pressed to actually work hard and do some hard work himself. Basically, this is a point-based scale in which at the end of a semester, a professor teaching a course compiles the ‘absolute’ marks obtained by all students and orders them. Now he gives ‘full’ points (this usually means a 10 point) to an arbitrarily selected top percentage of the students (based on the absolute marks obtained by them), 9 points (in this example) to the next, again arbitrarily, selected section and so on. The most conspicuous effect of all this being (it is also obvious from the name) that you are ultimately graded based on your performance relative to your peers. However, there are wheels within wheels and even though, in my opinion this method is far superior to absolute marking, it has its own little advantages and disadvantages.

Good things first. Like I already mentioned, everyone is graded on their relative performance. So if you completely messed up a paper, and so did everyone else, then good for you! Just kidding. But I think it actually gives a much better idea about the academic standing of a candidate. If ABC scores a 10 in Mathematics, then there is no doubt that he is a whiz at it, especially when the marking has been done against 300 –odd other students. You could go back to the old he-only-performed-well-in-an-exam but it is still a fairly reliable indicator. The second other substantial argument in the favor of this system is the fact that examiners can actually set difficult papers. And by difficult, I mean really difficult papers which can test each and every fundamental of the student taking the exam. And in an engineering institute this is of vital importance. When you have absolute marking, you also have a pass percentage, let’s say 40. Consequently, the paper has to be such that even a mediocre student should at least score a 40. This means that a quarter of the questions are a cakewalk. And you’ll have all the intelligent kids whining about this very fact after the exam. On the other hand, in relative, an examiner could fill the paper with one crunchingly hard question after the other and still have everyone pass. To make things clearer, in an extremely hard paper like mentioned, even if the highest score is 25 out of 100, the guy getting a 25 finally gets a 10. And even a ridiculously low score like 10 (absolute), if it is the average, would end up getting a 7. Another, smaller advantage is that it removes the importance of single marks that is such an ugly characteristic of the absolute system. How many times have you cribbed about that ‘one silly mistake’ that cost you 2 marks? Countless, I know. Not that silly mistakes are suddenly a good thing now but nobody rants about less than 5 marks. On a lighter note, everyone’s happier.

The flip side is minute. If a someone is indeed a prodigy and he scores 80 in a subject where the average is closer to 10, then he does lose out because he will just get a 10 while another getting 30 might get a 9 because he has the second highest marks. My next point is the fact that the professors do have complete control over the criteria for giving different points. On their own whim, they could choose to give only the topper a 10, or the entire top 10% of the class a 10. Some even choose cut-offs in advance, for example 80 above gets a 10, 70-80 gets a 9 and so on. This leaves no difference between the two methods. Now I am about to contradict myself here, but in rare (and I repeat, rare) instances a single mark could suddenly have a much greater impact than it ever would in the absolute system. If the professor is of the third kind, and he sets ranges such that a 79er gets a 7 and a guy getting 80 gets an 8, then there is no end to the distress of the 79er, even though it’s a source of great hilarity for friends. For the mathematically inclined, let me mention that losing one point is the equivalent of losing 10 marks in the absolute 100 scale. Luckily, we only find out our final points/grades and not the absolute marks.

So that’s about the gist of this system and I’ll leave it to you to evaluate it. Do leave your opinion in the comments section…

Siddharth Gupta