The IIT-JEE: Is It Really The "Toughest Exam In The World"?

I have, in a recent column here, described the over-zealous focus on Multiple Choice Questions at Kota coaching classes as damaging and destructive - to students and, eventually, to IIT.

The big question is why, in fact, the IITs still use a Multiple Choice Question (MCQ) format for the JEE or entrance exam. This format was adopted to enable machine grading, which, in turn, was necessitated by the vast and ever-growing number of candidates. Evaluating papers in which every "long" answer had to be read was becoming logistically impossible; there were simply no resources within the IIT system for a manual checking. A possible solution could have been to outsource the evaluation process but the IITs have fiercely guarded the fairness of the examination precisely by not involving any external agencies. It is also very likely that when the MCQ pattern was brought in nearly 12 years ago, no one really anticipated that it would lead to such severe problems.

The significant thing to do now is to change the examination pattern so that it becomes necessary to justify a correct answer with detailed working with machine grade-able MCQs. All those who mark the correct answer(s) will have their working checked manually as well - which means resources will be need only for the pool of students who have got the answers correct while ensuring that all the guesswork is rendered useless. There is the option of awarding part marks for the right approach but a wrong answer. Perhaps some questions can also be designed to elicit a series of numerical answers to be written into boxes (very easy to implement on-line; one variant of this is already being done), which may be accepted within a reasonable range of accuracy, provided their working is found to be correct; the working should be checked manually only if the answers are acceptable. 


The IITs still use a Multiple Choice Question (MCQ) format for the JEE or entrance exam

This sort of exam, which tests how students arrived at their answer, is much harder to coach for and is likely to curb the propensity of the coaching industry to market the IIT dream to all and sundry. As for students, studying for the changed examination pattern (involving long problems) will provide them with a reality check of their own abilities and aptitudes. "Success" in solving MCQ type problems can generate a false self-assessment by students about their aptitude for conceptual learning and creates false hope; such self-deception is not easy to indulge in when stuck in a long problem that needs clear fundamentals. 

A more radical set of reforms that have been suggested but never really debated seriously center on the idea of reducing the importance of the JEE rank by not using it to allot branches to students at the time of admission. Instead, the scores obtained by students after their first year should be used to determine this. Not only will this reduce the desperation to get the highest rank in the IIT-JEE - and therefore some of the coaching pressure to be a topper - it will also "force" students to focus on academics in their first year inside an IIT. An added benefit is that some students may discover and choose branches they really like rather than following the hackneyed trends of previous years. Of course, this idea needs to be developed further as there can be many possible variants such as taking the top 10,000 students and randomly allocating them to different IITs, or letting students select only an institute of their choice (based on their JEE rank) but not the branch, and so on. These ideas are likely to face resistance as they will upset the current equilibrium, most noticeably the covetedness of specific institute-branch-JEE rank combinations (e.g. IIT Bombay - Computer Science - top 100 ranks).

Another concern that comes up frequently is why does the IIT-JEE have to be so tough. Firstly, the IIT-JEE conforms by and large to the syllabus followed by CBSE up to Class 12, so it is factually incorrect to say that it covers an exceptionally broad set of topics. Second, the charge of being "tough" is used in comparison to board examinations. The comparison is flawed as the two examinations are different in numerous ways - the purpose of a competitive test is to sort aspirants into a merit list so, on purpose, it must be more challenging than a board examination where the purpose is to broadly test to how much one knows; also, board exams in general are not sufficiently "analytical", based as these are on significantly "rigid" and "rote-ish" teaching styles. Thus, in contrast, any question paper that requires thinking, depth of understanding, or originality can be problematic. Ideally, a reasonably competent student - who has good scores in a board examination - should be able to do reasonably well in any competitive examination. But because of the limitations discussed above, a board topper is not necessarily a person with clear concepts or analytical abilities. So the problems in the IIT-JEE are tough with respect to a reference that itself is very different. Asking institutions to "adjust" their entrance examinations to board examination standards and style is therefore not a very valid proposition. 


The IIT-JEE conforms by and large to the syllabus followed by CBSE up to Class 12

Thirdly, a fundamental attribute of an examination whose purpose is to create merit lists is that it has to be "hard". If we make it too simple, we will get clustering of students in the higher marks ranges; if it is made too hard then the clustering will be near the low marks ranges. Both outcomes are not very helpful for creating a ranked list of students, so a well-designed competitive examination will be somewhere in between, and will always be "hard". Even in the pre-MCQ era, the IIT-JEE was tougher than it is now for the simple reason that the long problems of that era needed in-depth working; in the current format, only a few minutes are given per problem which means the questions asked now are of a lower level of complexity. 

Coaching will not be eliminated by these measures. As long as inadequate and bad schooling exists and the difference between the number of aspirants and available seats remains huge, coaching of all kinds, including the Kota sort will thrive. Ultimately, it is only good schooling that can make them irrelevant.

(Anurag Mehra is a Professor of Chemical Engineering and Associate Faculty at the Center for Policy Studies, at IIT Bombay.)

Disclaimer: The opinions expressed within this article are the personal opinions of the author. The facts and opinions appearing in the article do not reflect the views of NDTV and NDTV does not assume any responsibility or liability for the same.