The JEE Debate Rolls On

August 25, 2008

P V Indiresan, the former director of IIT Madras has a comment piece in the Times of India today on how the JEE has run its course. His prognosis is gloomy – there’s no way to improve the entrance test without decreasing the number of applicants:

World-class universities like Harvard and Stanford get 10-11 applications for each available seat. It would appear that the IITs are much better off with their ratio of 70:1. Yet, IITs are unable to select the best because the JEE is mechanistic; Harvard selections have greater depth. Harvard uses tests like SAT, GRE, GMAT for shortlisting only, not as the arbiter for final selection, the way the IITs use the JEE.

The enormous number of applicants is the problem. It requires hundreds of examiners working in tandem to correct papers. In order to prevent individual bias, questions have to be so set that answers are all in a standard format; answers that demonstrate originality are out. Therefore, the JEE can be mastered by drill; innovative thinking is not necessary. Only when the numbers are small, is there time to ask incisive questions, to separate the truly intelligent from the merely industrious. Two-tier tests have been attempted to prune the numbers. That mechanism is of little use; it merely separates the more industrious from the less industrious and not the intelligent from the industrious.

The IITs would do well to follow Stanford, Harvard and so on. That is, IITs should first identify good schools and let those schools do the shortlisting for them. If the number of candidates entertained from any school is linked to the performance of older students from the same school, the schools will be under pressure to make their recommendations as efficiently as they can. If enough scholarships are offered in those schools for poor but competent students, the selection will also be broad-based; bright children from poor families will have a chance they do not have at present.

(Times of India)

Alas, for Dr. Indiresan’s recommendations to work, all changes made would have to happen in schools, and non-IIT engineering colleges (and even non-engineering colleges and universities), and not in the IITs themselves. This is a systemic snafu, and India is not particularly good at dealing with those.


Re-engineering the IITs

August 19, 2008

Sorry, couldn’t resist that title.

Andy Mukherjee has a column about the IITs making the admission criteria public, how coaching factories are skewing the results, and how there’s no readily apparent way to get through this.

It doesn’t say anything that hasn’t already been said in Indian newspapers, but it puts everything together much better.