ToI's Analysis of Medical College Admissions in 2017

I am not thrilled to see the word "merit" being used so casually, but Rema Nagarajan's ToI news story captures the essence in its title: Money, not quota, dilutes merit in medical admissions:

It is not caste-based reservation but money that compromises merit in medical admissions.

This is obvious from the difference of about 140 marks, or close to 20 percentage points, between the average NEET scores of admissions to over 39,000 government-controlled seats and those to the over 17,000 management and NRI quota seats in private colleges where fees determine admission.

TOI analysed details of nearly 57,000 students admitted to 409 colleges last year. The average NEET score of students in government-controlled seats was 448 out of 720, while the quotas under private control averaged just 306.

Incidentally, the average score of students admitted under the SC quota in government colleges was 398 and the overall average for SC students in all colleges was 367, both much higher than the overall average for privately controlled seats.

Gender and Affirmative Action Policies in India

S. Rukmini in has an interesting report in The Hindu -- ‘Quotas for education helped SCs, but boys alone reaped the benefit’ -- summarizing the results from a recent study.

The broad conclusion is in line with another study from 2008: Affirmative Action in Education: Evidence from Engineering College Admissions in India (pdf) summarized in a Mint op-ed by the authors. Key quote: "... we find that the affirmative action policy appears to hurt female applicants, given that a higher percentage of those let in through affirmative action are males."

Links

1. Curt Rice in The Guardian: Don't be fooled by the closing gender gap in science PhDs.

   [The researchers] found that historically men have had higher persistence rates than women, with a greater proportion of men having continued for a PhD. Since the 1990s, we see something else. The persistence rates have coverged: men and women continue in equal rates. That’s great news. Or so it would seem.

   Unfortunately, the new study doesn’t actually show a pipeline being tightened up to leak less – it shows the opposite. The convergence in persistence rates for men and women is not a result of an increase in the rate of women taking a PhD: it’s the result of a decline in the rate of men doing so, which now stands at 3%.

   Is this something to celebrate? I can’t imagine why. ...

   See also: Bob Grant's article in The Scientist -- New Look at the Leaky Pipeline -- on the same research.

2. An idea whose time has come! Noah Smith on Affirmative Action for Conservatives.

3. Finally, do yourself a favor and go to the 58th minute of the video below (or at YouTube) and wait for Bobby McFerrin's magic over the next several minutes. [I know I have linked to this stuff a long time ago, but it's worth watching any number of times]:

Links

1. A BBC news story on affirmative action policies in Brazil's federal universities. Almost all the arguments -- for affirmative action and against -- sound so familiar!

2. Mail Online on the extent of faculty crunch in India's "top institutes".

3. Greg Mankiw: A Carbon Tax That America Could Live With.

Links

1. Next Chief Justice of India favours reservation in higher judiciary.

2. Unreserved on the rolls: "The merit argument is debunked as many OBCs make it to IITs in general category."

3. Indian government's plans for 'heritage' universities:

   ... It will ... give a special 'heritage status' that are at least a 100 years old as per a Rs 100 crore proposal being finalised by the Union human resource development ministry.

   Making the cut are state universities like Universities of Calcutta , Bombay and Madras set up in 1857, Presidency University (1817), Osmania University (1908) Central Universities Aligarh Muslim University (1875) and Allahabad University (1887) and deemed universities like Indian Veterinary Research Institute (1889), Ramakrishna Mission Vivekanand University (1897) , Jamia Hamdard (1906), Indian Institute of Science (1909) and Bengal Engineering & Science University (1856). Jadavpur University, Forest Research Institute (1906), Indian Agricultural Research Institute (1902) are also being considered for the heritage tag.

Women in Science in Japan

NYTimes ran a story on the massive underrepresentation of women in the sciences -- Japan's "Science Women" Seek an Identity by Miki Tanikawa. The story covers this phenomenon from many different angles, but this one stood out for me: the fate of a proposal to guarantee at least 5 seats (out of 54) for women in the Math department at Kyushu University:

Attempts at raising the number of women come against legal barriers, underpinned by social mores and cultural forces. In 2010, faculty members of the Kyushu University mathematics department concluded that a more proactive admissions policy might be needed to recruit more women. The number of female students was only in the single digits, out of a student body of more than 50.

So the faculty decided to set up a quota. The first group of 45 students selected would be done regardless of gender. But, in the second group, the department would admit at least five women out of nine slots. Ultimately, that meant a quota that guaranteed a minimum of only 5 women in 54 total places.

But months after the announcement was made, calls and e-mails poured in criticizing what was seen as “reverse discrimination” and the breaking of the “equality before the law” principle, said Masanobu Kaneko, dean of the department.

“They claimed it would be unconstitutional, violating Article 14 that guaranteed equality of gender before the law,” he said. “We realized that this could lead to a lawsuit,” possibly by male applicants who failed to get in.

“If we lost the case, it could result in irrecoverable damage” to the school’s reputation and cause problems for those who were admitted, Dr. Kaneko said.

On the advice of lawyers and constitutional scholars, the faculty decided that they could lose such a lawsuit. Baffled, they gave up the idea.

Links

1. Jennifer Carpenter at Science Insider summarizes yet another study whose results would appear to blame women for their low status in science: Study Finds Women Biologists More Likely to Avoid Spotlight at Conferences. Here's the study itself.

2. Rod Smolla has a great essay in Inside Higher Ed on The Legal Future of Affirmative Action in the US.

3. UGC announces a set of norms for what our universities can and cannot do -- mostly the latter. Among the cannots: offer distance education program for MPhil and PhD degrees, courses through franchisee institutions, courses outside the state in which they are established.

4. Trouble in CAT-land: results of some students have been tampered with -- not at the IIMs, but on the website that dishes out CAT scores to students and institutions. Private institutions that admit students using CAT scores are crying foul.

5. Finally, xkcd has put together a very informative document on how modernity has destroyed humanity affected, among other things, people' ability to connect with each other.

Links

1. WiseGeek: What does 200 Calories look like? Pictures of 200 Calories of Various Foods.

2. Robert Cottrell (the editor of The Browser: Writing Worth Reading) in FT: Net Wisdom.

   My first contention: this is a great time to be a reader. The amount of good writing freely available online far exceeds what even the most dedicated consumer might have hoped to encounter a generation ago within the limits of printed media.

   I don’t pretend that everything online is great writing. Let me go further: only 1 per cent is of value to the intelligent general reader ... Another 4 per cent of the internet counts as entertaining rubbish. The remaining 95 per cent has no redeeming features. But even the 1 per cent of writing by and for the elite is an embarrassment of riches, a horn of plenty, a garden of delights.

   The essay covers a wide variety of topics -- including what is so great about blogs by academics writing about their fields of expertise.

3. Makarand Sahasrabuddhe's answer in Quora to: India: Is reservation the best method of affirmative action in India?

Affirmative Action Goes to Wachington

Even before the US Supreme Court sits down to hear arguments in the latest affirmative action case to land at its doorsteps, it has been getting tons of reading material from various groups. One of them [link, link] makes an interesting case: replace race-based AA programs with one that's based on factors such as parental income and education, wealth of the student's neighbourhood, etc. A class-based AA program, in other words.

The broad idea, I think, is to create something that would achieve the diversity-related goals of institutions (without using race explicitly as a factor), and still be acceptable to the public, the lawmakers, and the courts.

The case itself is scheduled for a hearing in the coming days and weeks, and the SCOTUS verdict will probably arrive early next year. It will be interesting to see how this case is decided, and the arguments used for supporting that decision.

* * *

During the Mandal II debate in 2006-07, Yogendra Yadav and Satish Deshpande argued for a reservation system [see this post for links] that would, in essence, provide additional points to students from disadvantaged backgrounds due to their caste, sex, education and income levels of their parents, etc. In the event, the UPA-I government decided to run with a fixed quota for OBC students; the Supreme Court accepted this scheme with just one modification -- the so-called creamy layer clause that would exclude students from wealthy families from the quota benefit.

Unlike the US, there is an expectation in India that admission policies should be based on completely objective criteria; this explains the emphasis on cut-off marks, entrance exam scores, and the like. It's probably this that led Yadav and Deshpande to propose their "deprivation index" so that admissions officials would still have some objective numbers to base their decisions on.

OBCs to get the full benefit of reservation at the IITs

The Supreme Court has spoken:

Other Backward Classes (OBC) students will qualify for admissions if they have 10 % less marks than the eligibility level fixed for the general category students, the Supreme Court has ruled. The apex court Thursday clarified that 10 % relaxation being given to OBC students did not mean it will be 10 % less than the last student admitted under general category.

Since the IITs were following the second policy (which now stands quashed), the fraction of OBC students among this year's JEE-qualified students was just about 20% -- as opposed to the 27 % they were eligible for.

Thanks to today's SC verdict, OBC students will start getting the full benefit of the reservation policy from next year (this year's admissions will not be affected).

All in all, a welcome clarification from the Supreme Court.

* * *

Update (19 August 2011): J. Venkatesan's report in The Hindu has all the crucial details. In particular, the SC order prohibits re-allocation of OBC seats to the General Category as long as there are OBC students who meet the (relaxed) eligibility criteria.

13,602

That's the number of 16-18 year-olds who learned today that they are actually world-class, a privilege that allows all of them (and actually goads some of them) to live a life filled with a sense of entitlement and utter disdain for mere mortals -- especially their teachers. The press release is here.

* * *

Some highlights:

1. "A total number of 4,68,240 candidates appeared in JEE-2011 ... 13,602 candidates have been declared qualified to seek admission for 9618 seats in the IITs."

   Thus, about 3 percent of the candidates qualified. This figure is as high as ~5 percent for the Madras zone. I guess there's a lot of celebration at Hyderabad, which has probably overtaken Kota, Rajasthan, as the city of JEE dreams.

2. "This year 108647 girls appeared for JEE-2011 out of which 1491 qualified."

   There are several ways to see this result. First, the fraction of girls who qualified is 1.4 %, which is less than half the overall figure of about 3 %. Second, and more tellingly, boys are about two-and-a-half times more likely to be among the rank-holders than girls (3.36 % for boys against 1.34 % for girls).

   Third, girls form about 11 percent of rank-holders, while their share among the exam-takers was 23%.

3. In the OBC category, 138,104 took the exam, and 2700 qualified -- a ratio of about 2 %. A more relevant figure is the fraction of OBC students among the JEE-qualified: about 20%. While this figure is better than those of the previous two years (19% and 18%), it is still far short of the 27 percent supposedly set aside for them.

4. In the SC category, 45172 students took the exam and 1966 qualified -- yielding a pretty fabulous (by JEE standards) ratio of 4.3%.

   This also means that the SC quota has been filled to the brim. This is the second time that this key mark of progress has been achieved.

5. In the ST category, 766 students qualified out of 18125 -- for a ratio of 4.2 % which, while impressive, still doesn't allow them to fill all the available seats (mainly because their share in the group of exam-takers is low: 3.9 % which is about half as much as their quota.

6. The PD category is special: over 19 % of the PD candidates qualified (440 out of 2286); since their number among exam-takers is small, this huge pass-through ratio is still not enough to enable them to fill all the available seats.

More later, when the IITs release detailed information.

Affirmative Action at Oxbridge

Three things that I came across on my Google Reader stream in the last week or so:

1. Let me start with this NYTimes story:

   David Lammy, a former higher education minister and Labour member of Parliament, has used figures obtained under Britain’s Freedom of Information Law to reveal a dearth of black and other minority students at either of the country’s two oldest universities.

   According to Mr. Lammy, last year the whole of Oxford University admitted just one black student, and Cambridge University does not have a single black faculty member.

2. The Guardian carried a debate on its pages: Should Oxbridge discriminate to boost the numbers of black students? You'll find an echo of the debate that raged in India in 2006 when the Union government chose to implement a 27 % quota for OBC rstudents in federally funded institutions (including the IITs, IIMs and AIIMS).

3. Richard Kahlenberg at Innovations: Oxford’s Research-Based Affirmative Action

   A large-scale British study, released last week, gives new empirical support for the drive to provide affirmative action to “strivers,” less advantaged students who, despite obstacles, perform fairly well academically. The research finds that students who attended regular “comprehensive” (public) secondary schools did better in college than those who scored at the same level on standardized admissions exams and attended “independent” (private) or “grammar” (selective public) schools.

   Pointing to the study last week, Oxford University’s dean of undergraduate admissions, Mike Nicholson, created waves when he declared that students who do well at poor performing secondary schools “may have more potential” than those from more-advantaged schools, and that universities should consider the context in which students compile an academic record. [...]

* * *

Addendum: See also:

• David Lammy: The Oxbridge whitewash.

• Sally Mapstone: These figures do not show that Oxford and Cambridge discriminate.

Links ...

1. Female Science Professor answers a reader's question about on how to deal with "professors emeriti who are well-meaning but who have not found productive ways to spend their days without distracting the more-busy and without wreaking minor havoc on various parts of the department infrastructure."

2. Celia Dugger in NYTimes: Campus That Apartheid Ruled Faces a Policy Rift -- the rift refers to the raging debate on the legitimacy of the use of affirmative action at the University of Cape Town.

   Affirmative action’s champion on campus is Max Price, the vice chancellor, who was himself detained as an anti-apartheid student activist in the mid-1970s. Dr. Price, who grew up as a child of white privilege, contends that preferences based on apartheid’s racial classifications provide a means to help those harmed by that system to gain critical educational opportunities.

   The university has an openly stated policy of admitting blacks who have substantially lower test scores than whites, but whites still outnumber blacks almost two to one — 45 percent versus 25 percent — among the 20,500 South African students at the university. In South Africa, 79 percent of the population is black and only 9 percent is white.

Science in India: "Quantity need not mean a loss in quality"

The Nature India portal features an interesting column -- The Quality - Quantity Conundrum -- by two colleagues of mine, Prof. Gautam Desiraju and Prof. M. Giridhar. [Note: the site requires (free) registration.] Here's a quick excerpt:

An earlier analysis on the contributions of the best 10 research institutions in these countries to their overall publication records showed that the premier Indian institutes contributed nearly 30% over the span of 25 years. This is in keeping with the sluggish Indian curve in figure 1. However, in China, the contribution of the top institutes has decreased from 53% to 39% during the period of 2002 to 2008, which coincides with the exponential overall growth in that country. Clearly this means that the second and third rung research institutes have begun to participate actively in research. This can only be possible if trained manpower who studied in the top institutes went on to teach and do research in these institutes.

* * *

While much of their analysis is admirably data-driven, I wish they didn't take a loose, evidence-free swipe at the reservation policy:

There is also a real need now to assess the consequences of the caste based reservation system. Where has 50 years of reservation taken us, in a world where no quotas are applied in competitive activity?

The issue may appear in the form of a hard-nosed question about the consequences of reservation, but this sort of rhetoric is intellectually lazy -- as lazy, in fact, as that flowing in the opposite direction: where has 60+ years of upper caste domination of higher education and scientific research taken us?

Fifty percent limit on reservations

Looks like this arbitrary, nationwide limit is now history:

The Supreme Court today virtually lifted its 50 per cent cap on job and education quotas, allowing Tamil Nadu to exceed the volume if “quantifiable data” justified it.

The order implies that any state can now do a headcount of all groups it deems backward, place the numbers before the state backward commission, secure its recommendation for higher or additional quotas, and pass a law implementing them. Such higher quotas can no longer be legally challenged for exceeding the 50 per cent ceiling but only on other grounds, such as a faulty headcount.

Here's what this ruling implies for the states of Tamil Nadu and Karnataka:

The quota volume in [Tamil Nadu] can, therefore, rise even above the existing 69 per cent if the data justify it. Tamil Nadu would ideally like 88 per cent reservations.

Neighbouring Karnataka, which had enacted a 50 per cent quota for the Other Backward Classes in addition to 23 per cent for the Scheduled Castes and Scheduled Tribes — raising the total to 73 per cent — too has been allowed the same leeway.

Till a year from today, the quota volume in Karnataka will stay at 50 per cent, but the state can within that period do a survey of backward communities. If its backward commission then accepts that the data justify a hike to 73 per cent or higher, the state can pass a law to that effect.

Karnataka wants 57 per cent reservation for the Other Backward Classes alone, which means a total quota volume of 80 per cent.

JEE-2010

Some quick comments. First, some groups that have done extremely well to reach major milestones:

1. SC Students have managed to fill all the seats (and then some) reserved for them. For the first time.

2. Andhra Pradesh: First and second rankers are from AP, which is also the home of seven among the top ten. The state has scored in other ways too:

   Students from the state also scored well in the reserved category. In the OBC category , AP bagged seven out of the top 15 ranks. Among all reserved categories, including SC, ST, OBC and Physically Handicapped (PH), the state secured over 50 ranks in the top 200.

   Officials from IIT-Madras said that students from AP already account for 21 per cent of the total strength in IITs.

   [Be careful with that ToI report, though: it has howlers like, "An estimated 30 per cent of the 50,000 students who wrote the exam from the state cleared the test this year." If these numbers are correct, AP alone would account for 15,000 of the 13,104 ranks!]

3. Hindi speakers: The number of JEE rankers who took the exam in Hindi tripled to 554 from 184 last year -- despite the problems that plagued the Hindi paper this year. [See items 2 and 3 in the corrective steps taken by the IITs].

Some other interesting bits:

1. JEE-2010 was taken by 4,55,571 students, up 18% from JEE-2009's 3,84,977.

2. The number of JEE-qualified candidates has gone up 30% to 13,104 from 10,048.

   [Not all of them can demand a seat at an IIT (or JEE affiliates), however. The number of IIT seats is only 9,509 this year.]

3. This year, 1476 girls have qualified, an increase of over 40 percent over last year's 1048. Girls' share in the exam takers also increased 15% from 98,028 to 1,13,127, and Their success rate went up from 1.06% to 1.3 %. So, this is a clear gain for girls.

4. Akansha Sarda is the topper among girls -- with a rank of 18. She's not joining an IIT, however; she's going to MIT.

5. Sahal Kaushik of Delhi at No. 33 is just 14 years old, and home-schooled. And, what does he want to study? "I want to study pure science, physics or mathematics, not engineering. ... I took the JEE because I could also get science courses through it."

6. Anand Kumar's Super 30 achieves 100% success -- for the third time in a row. [An ally-turned-rival has raised some questions about this institution's record, however.]

Caste Census

Three links.

In the first, Pratap Bhanu Mehta goes ballistic in opposing the very idea of a caste census. Scratch the surface a little bit, and you too would wonder how much of his opposition is driven by an almost visceral hatred towards the so-called OBC politics:

Third, giving in to a caste census is giving in to a discourse of raw political power. The blunt truth is that designing remedial measures for Dalits, including addressing discrimination does not require a census. This demand has rather been fuelled by politically assertive groups like OBCs, who first hijacked the Dalit discourse on deprivation to their own ends.

Nandini Sundar, on the other hand, goes into the question of whether a caste census can do anything to reduce inequality, and concludes:

... [W]hile the census can provide base figures, it cannot substitute for the kind of information needed both for inclusion of castes in an OBC list or for ‘graduation' of castes out of the list, even assuming the latter were ever to be politically feasible.

In the meantime, Amitabh Bachchan knows what he will say when the census enumerator asks for his caste: Indian.

Art of the Possible: The Curious Case of the Caste Census

The caste census is (most likely) on! Let's hope the formal decision will be announced soon.

To see the significance of this (impending) decision, we need to jiggle our neurons and go back to 2006 -- to the height of the OBC reservation debate.

The more wonkish of the opponents of the reservation policy kept questioning the Mandal Commission's estimate of 52 % for the OBC share in the population. They kept hammering away at the fact that the last caste census happened way back in 1931 -- almost four generations ago! They kept denying the validity of caste data from surveys such as the National Sample Survey.

They kept claiming that the economic and educational situation of OBCs is not so 'backward' as to require affirmative action. [Specifics in 2006-08 era posts, especially those under the category Caste]

Most importantly, they kept pounding into us that the OBC share in the population was quite likely far smaller than what the government claimed.

Since a caste census would prove them right, they kept insinuating that it would be killed by a pliant government too scared to take on the "OBC leaders" who would surely oppose it.

Guess what?

[In the Lok Sabha debate,] Sharad Yadav, Mulayam Singh Yadav and Lalu Prasad — all OBC leaders — warned the Centre against committing the blunder of omitting caste. The Bahujan Samaj Party, Biju Janata Dal, Telugu Desam Party, Trinamul Congress, DMK, CPM, CPI, ADMK, NCP, Akali Dal, Shiv Sena and the Janata Dal (Secular) unambiguously supported caste enumeration.

In the two major national parties, the initial response to the demand for caste census was not so much as to oppose it, but to side-step it and hope that it would go away. BJP chose to focus on Bangladeshi 'infiltrators', before outraged leaders like Gopinath Munde forced its hand.

The Congress too has its own supporters of caste census [see this report for a summary of the Lok Sabha debate]; but in a Cabinet meeting, they were sidelined:

At Tuesday's Cabinet meeting, Home Minister P. Chidambaram, whose Ministry oversees the census, argued against the demand to include caste. He said the enumerators lacked the sociological sensitivity to record and classify the population on the basis of caste and sub-caste. According to sources, he suggested that the backward classes commissions at the State and Central levels be better placed to conduct a detailed survey. Others not inclined towards a caste census were Finance Minister Pranab Mukherjee and Human Resources Minister Kapil Sibal.

The Natural Party of Governance appears to have come around to accepting caste census. And this is indeed good news.

It's not done, yet. So, let's wait for the official announcement before hailing it as an example of politics that works.

JNU and Reservations

It's sad to read Anuradha Raman's Outlook story on the ongoing 'debate' within JNU on faculty reservations:

Given [Vice Chancellor] Bhattacharya’s ambivalence on reservation -— amply demonstrated, faculty members say, in two meetings of the academic council and the last executive council meeting of April 6—many at the university think his continuance will thwart any effort to introduce reservations in the faculty. In fact, the meeting of the executive council—the highest body in the university—was marked by a vicious debate. It was even decided to refer the matter to the solicitor-general of India, prompting executive council member P. Sainath to ask why the highest legal officer of the country should opine on the administrative matters of a university, especially when the university had submitted to the Supreme Court in 2008 that it would implement faculty quotas in two years.

Reservations and JNU faculty

It was quite shocking that a bunch of JNU faculty members resorted to the same cliched arguments against reservations at the faculty level -- especially at the full professor level.

That was two weeks ago.

Now, in a welcome move, quite a few of them have backtracked from their earlier stand.

As for the official stand, this is what the Vice-Chancellor B.B. Bhattacharya had to say:

... [T]he university was in the process of furnishing its response to the query. “We have to first get a clarification on whether reservations in the academic faculty is a law; if so, we won’t go against it.”

While this seems to me like a strange thing to say, I hope he finds the time to figure out the right thing to do, and do it.