Sunday, August 02, 2009

Media reactions to JEE-09 disclosures

After yesterday's disclosure of data by the IITs, I found two newspaper reports on the results of JEE-09. Predictably, both played up the reservation angle (just as I did), but in radically different ways.

First up, here's Charu Sudan Kasturi's report in The Telegraph:

OBCs bust quality myth

Two of every three OBC students selected to the IITs this year would have made it without any quotas, 2009 entrance test details reveal, debunking fears that this year’s quota hike would lead to a drop in student quality.

At least 1,300 of the 1,930 Other Backward Class students admitted to the IITs through the Joint Entrance Examination secured marks that would have guaranteed them seats even without the quotas, details released today show.

In chemistry, the OBC topper notched 126 — four more than the general category (and overall) top ranker.

Next up, we have Hemali Chhapia's report in The Economic Times:

IIT cut-off falls to 18% for SC/ST students

A minimum of 35% is essential to be promoted to a higher class under almost every Indian school board or university. But you don't need that much to make it to some of the finest technological institutes in the country.

On Saturday, when the Indian Institutes of Technology released report cards of students who joined these engineering schools this year, it transpired that the entry bar for the reserved category students had dropped to a mere 18% (89/480).

The IITs were forced to make various concessions to fill SC/ST seats this year. Entry levels were lowered to half of what the last general category student who got through to the IITs scored. So, as the last general category student admitted to the IITs bagged an overall score of 178 (out of 480), the cut-off for an SC/ST student was brought down to 89 (half of 178). Till last year, the cut-off for SC/ST students used to be 60% of the score of the last general category student.

Even if she wants to highlight the fact that 18 percent is what the SC/ST students needed to get into the IITs, she chooses the wrong number -- 35 percent that one needs in Board Exams -- for comparison. The relevant number should have been the subject-wise averages in this year's JEE itself: 7%, 4% and 7% (in math, physics and chemistry, respectively).

In fact, one of the striking features of JEE is the very low averages -- percentages in single digits. Remember, this is an exam that students self-select into. Remember also that this is an exam for which they prepare hard and pay good money for coaching.

Given these facts, the single-digit averages are atrociously, horribly, low.

I'm sure Hemali Chhapia knows of no board or university exam with such low averages. And yet, she compares 35% in them against 18% in JEE -- a fiercely competitive exam.

Is Hemali Chhapia really this clueless? Is she being intentionally misleading? Or has she simply internalized her newspaper group's attitudes?


  1. Raj said...

    "At least 1,300 of the 1,930 Other Backward Class students admitted to the IITs through the Joint Entrance Examination secured marks that would have guaranteed them seats even without the quotas"


    1) Quality has not suffered much because of the quotas.
    2) We can now dispense with the quotas as OBCs are able to make it on merit anyway.

  2. Abi said...

    @Raj: Nice try on the second "conclusion". I knew I could count on you to tease out all the conclusions from a given observation. ...

  3. Anonymous said...

    But seriously Abi, why do you think the second conclusion is not a valid conclusion?

  4. Abi said...

    @Anon: If you read my previous post, you would know the answer to your question. The JEE-09 results make it very, very clear that we need affirmative action -- reservation, in Indian context -- to enable OBCs to gain access to educational opportunities. Without reservation, the data show that only 14 percent of the seats go to the OBCs.

    I am yet to see anyone make the claim that OBCs form 14 percent (or less) of the population in India.

  5. Anonymous said...

    Same Anon: 14% is not an insignificant fraction, like say, 0.5%. Your numbers show they are doing well.

    Are you suggesting that if in any exam/job, we do not have densities in outcomes nearly mirroring the demographics of the state, we should have reservations to "correct" the variations?

    By the way, I am not against reservation per se, esp. SC/ST. (Mandatory line to avoid getting clubbed with some clueless ET reporter!) In the present context, the OBC slope does not seem too steep to merit reservation. Why do you think it is otherwise?

  6. Niket said...


    I don't think anything I say would convince you to take a re-look at your views on reservations... so I won't go in that direction.

    Still, single digit averages is actually NOT "very low" due to negative marks. The mean marks for each paper if you were to answer questions randomly in JEE is zero.

  7. Abi said...

    @Anon: The logic for affirmative action to expand opportunities for SC/ST, OBCs and PDs (and, if I may add one more category --women) is the same: If the participation rate for a group is far lower than its share in the common pool, that group may be a good candidate for for affirmative action.

    You may argue that 14% is a good enough representation without the benefit of reservation. I don't know how you would define "good enough." The way I see it, it's not good enough.

    I have seen all kinds of figures for the OBCs' share of the population; none of them places it below 27 percent for which I drew Figure 3 in my previous post. The estimates at the higher end are larger than 50% (Mandal Commission has it at 52%).

    @Niket: The average for random answers may be zero -- which would assume that JEE-taking students are a bunch of dummies. But aren't they a bunch of self-selected, highly motivated students? Don't they come in with expert coaching? Given these characteristics of the pool of candidates, I would argue that an average of 4 or 7 percent is very low -- and this shows how brutal the examination is. I have ranted about this before, so I won't say any more here.

  8. Anonymous said...

    Hi Abi,

    This is very standard Hemali Chhapia fare. I have often been irked by her casual, almost banter/gossip-like reporting on matters she does not understand, particularly the IITs, and have even slammed her and her editor on this. Being an alumnus of IITB, I have often found her opinions to be poorly researched, highly prejudiced and fakely sensationalized.


  9. Anonymous said...

    Im a jee aspirant, a dropper, i gave last years (2009) paper WITHOUT any coaching or any practice, and got 129 marks.

    on that basis, isnt 89 an abysmally low percentage for admission? agreed that 35% is a ridiculous reference frame and shows the naivete/bias of the reporter.

    But 7%,4% and 7% are AVERAGES of the FOUR LAKH students who gave the exam, of which ONLY 10000- 2.5% of the total got in. the last admitted (general category) student got 75-71-35 in P-C-M, not even close to those low subjectwise cutoffs.

    the exam is perhaps the only 'brutal' one left in india- in the maharashtra CET for example, anything less than 180/200 is considered a mortal sin. More than the board marks and other exams, this is a true test of understanding and is brutal for us, used to rote learning for the boards.

  10. Raj said...

    Abhi, the goals, boundary conditions and charter of ‘affirmative action’ need not be held as sacrosanct for all times to come. Some re-validation of old assumptions is called for. And a recalibration wherever required.

    Recent data (one you have provided here, and from Prof TTR’s blog that you had linked to earlier) seems to provide evidence that a) a high proportion of OBC students who have got their admission would have made it through the unreserved channel anyway and b) among the OBC students who have got admitted, a fair percentage perform almost as well as the students who got in through the unreserved channel.

    The ‘affirmative action’ cheer band has been quick to use this data to bust the quality-will-suffer-myth. As an academician steeped in analytical methods, you know far better than I do that the data can provide more than one inference. For instance, what were the characteristics and profile of those OBCs who performed well in the entrance as well as in the semester exams? Were they relatively more affluent than the rest of the OBC candidates? (In which case, it’s wiser to introduce an economic criterion). Were they the children of educated parents? (In which case, it is wiser to introduce quotas for first generation graduates, cutting across caste lines). And so on.

    If you blindly tell us that affirmative action must continue without a sunset clause, till OBCs constitute 27% of the total admissions, and every classroom is a microcosm of society, then I’m afraid you’ll be ignoring the lessons that the emerging data could provide, and lose out on the chance to make a mid-course correction..

  11. Abi said...

    @Raj: If you went back to the 2006 archives (May-August), you would find that I belong to the Yogendra Yadav - Satish Deshpande camp that takes an evidence-based and calibrated approach (that demands continual tinkering and, therefore, a sunset). I belong to the same camp that also said, eventually, that in the absence of such an approach, some type of affirmative action is better than none. This is because the data are clear: OBCs are grossly underrepresented in higher ed. The data from this year's JEE are a ringing confirmation. I don't know how the data can be viewed to support the view that OBCs don't need reservations.

    The anti-reservation folks' main objections were focused on how OBC reservation would "dilute" quality. Given the huge noise they made and all the media attention they got, I think Charu Sudan Kasturi has done the right thing by highlighting that nearly two-thirds of OBC students would have got into the IITs without reservations. Similarly, the Outlook article is right to highlight that the OBC students in the IITs do about as well as the rest of the class.

    To refer to them dismissively as "affirmative action cheer band" is bad enough. But to use their "cheering" to argue against the need for OBC reservations ("all right, this just shows that OBCs don't need reservations") is to betray a certain disrespect to the data that you claim you want to use as an aid to policy.

    It's interesting that you posed questions about the economic and educational background of beneficiaries of OBC reservation (in the interest, of course, of mid-course corrections). It occurred to me that we never raised such questions about anyone until OBC reservations kicked in. Maybe it's time we asked these questions -- not just about OBC students, but *all* JEE rank-holders, and seek to introduce an "economic criterion" and "quotas for first generation graduates".

  12. Raj said...

    Abi, sorry if I rushed in without dipping into the archives first. This is not a subject that usually catches my attention, nor can I claim to be well-informed; I just read this post of yours as a stand-alone one, and shot off the comments.

    I asked this question of Prof TTR and I am repeating it here, hoping that you would shed some light. In support of its claim that quality has not been compromised, the Outlook article says that the average grade in the non-reserved category for the 1st semester was 7.92, while it was around 7.22 for the reserved OBC category. As one well-versed with the grading system, do you believe that only a small difference separates the two grades?
    My impression, based on feedback from my daughter who is in an engineering course and grappling with this system is that this is the kind of gap that separates the mediocre from the really good.

    Also, it is incorrect to say that the main objection of the anti-reservation front was that quality would suffer. It was more on a point of principle, to wit, against the dismantling of a well-tested system based on meritocracy.

    But given that affirmative action has been implemented and is irreversible, there’s no point in arguing on the basics again. I would be glad if we would make course corrections, in the light of new data that would keep coming in.

    I entirely agree with the last para of your reply. If the data suggests that ‘economic status’ and ‘first-generation-graduates’ are what held back a section of society from these institutions, by all means, let us use these criteria for all categories and fix quotas

  13. Anonymous said...

    @abi and the rest:
    I understand reservations do have a role in uplifting certain sections of our society. However, at what level and form do our reservations need to be implemented. Is IIT-JEE or other entrance exams the right place? Do we have to wait for kids to reach up to this point of education in their lives. What about those who did not choose to not write JEE, CEE, etc.
    Aren't we looking at a superficial solution to the problem by doling out reservations in these entrance exams?


  14. Anonymous said...

    And now this....

    great... not only admission but special and give them one more chance.... if you have been taught the same way as the rest, why ask for more?