Observations On The Educational System In India

The other day my brother was telling me that the IIT's (Indian Institute of Technology) aren't ranked in the top 25 engineering institutes in the world. I replied that I had read that they aren't ranked in the top 100 even. (http://epaper.timesofindia.com/Default/Scripting/ArticleWin.asp?From=Archive&Source=Page&Skin=TOINEW&BaseHref=TOICH/2010/09/17&PageLabel=10&EntityId=Ar01000&ViewMode=HTML&GZ=T) He wondered why... In India the IIT's are very highly reputed.

That set off a train of thought. This is something I had been thinking about for a long time. The students who come out of IIT's are very bright. They go on to do their masters and doctorates in reputed colleges in USA like MIT, Stanford etc. And many of them go on to work in organizations like Bell Labs or NASA. There is no doubt that the students who graduate from IIT are "high class".

Now that raises a question. What are the factors that result in "high quality" of students who graduate from an institution?

I think there are primarily two factors.

1. What kind of students are taken in? What is the entry criterion like? Very high bar or a fairly low bar? (The better the quality of raw materials, the better the quality of finished product)
2. What kind of value add is done in the institution? Does it convert a fairly "inferior" raw material and process into a high grade?

It's my belief that in the IIT's, as well as in other institutions of learning, the quality of students graduating is largely determined by point 1 above and to a very small extent by point 2 above. Meaning the students who are gifted go on to do well. The school itself doesn't convert the not so gifted ones into high grade. And IIT's have a fairly high entry barrier and hence the students passing out also are relatively better. No thanks to what happens inside IIT (to the education inside).

Now is this really so? Meaning isn't there much quality teaching in IIT's? What about quality research?

To answer that question (just a yes or a no) let's see the factor that is relevant in an institution for higher learning. "How much are the institutions trusted (and hence) involved in solving problems?".

Let's take some examples:

1. Water scarcity in many towns in India in the coastal regions.
2. Electricity shortage and the abundance of sunshine in most parts of India.
3. The frequently recurring train accidents in India.
4. Heavy engineering: Construction of airports. Incidentally  yesterday, (a couple of days after I wrote the first version of this blog) I was talking to a friend of mine who is privy to the expansion plans of one of the major airports in India. The current status of that airport, as he explained to me, corroborates this point.
5. Cooum (a stinking river in Chennai) desilting.

I have listed some of the very common problems (or areas) faced in india which need an engineering solution. How many of these problems are solved in India (as against technology obtained from elsewhere and "adapted" locally? To the best of my knowledge most of these have been solved largely through technology acquisition than by indigenous development. This despite having 5 (now more) such IIT's - the so-called "high end engineering institutes". Now if we wont depend on our own institutions to solve our basic problems, if we won't eat our own dog food, it reflects a rather poor trust that we have on these colleges.

Why aren't these institutes ranked among the best in the world? Why don't we go to them for solving our problems?

This is not just because the teachers aren't good or that the facilities are not adequate. I think the issue goes deeper than that. Culturally Indians have always been good at solving (read: memorizing) solutions for known problems. 

Right from the time we were kids we were encouraged (I should say we were forced, not encouraged) to repeat what our teachers/elders said. We were actively discouraged from independent thinking. "Why don't you just write what is there in the text book? Why do you have to change the wording?". Remember these words from our parents/teachers?
In India it's perfectly acceptable (as a matter of fact it is insisted on) that we take large chunks of material from other textbooks or references and submit the same as a paper/essay. In USA, the same action would be strongly discouraged and considered as plagiarism (unless the material is given appropriate reference and inserted inside quotes). In India, we believe anything can be trusted only if we didn't create it ourselves. 

See the difference in culture? This resulted in Indians dreading any open ended problems. It's like how a child hates darkness. We have been forbidden to go THERE (THERE meaning the unknown); to look inside THERE, to explore THERE. Born and raised in such a culture, how do we ever send a man to the moon or  or design a new rocket or a submarine?

While South Africa hosted the World Cup Soccer game to perfection about a month back, one just has to look at the struggle India is going through, (to host the Commonwealth Games due to start in 2 months time), to realize how pathetic our performance has been. South Africa and India are ranked 129, 134 respectively, in the Human Development Index.  Here is an article explaining the basis of India's success as an economy
(http://www.globalresearch.ca/index.php?context=viewArticle&code=CUN20100925&articleId=21182). While the article is extremely critical, there is some truth in it, which we need to reflect on.

Indians knew about eclipses and could predict them thousands of years before physicists like Galileo Galilei, Newton or Kepler came along. Indians had a flourishing civilization, their literature dates back thousands of years. See this link which tracks the economic strength of India and China in the last few hundred years:
http://epaper.timesofindia.com/Default/Scripting/ArticleWin.asp?From=Archive&Source=Page&Skin=TOINEW&BaseHref=TOICH/2010/08/22&PageLabel=16&EntityId=Ar01601&ViewMode=HTML&GZ=T. I wonder how Indians solved problems then. They must have been were very good at problem solving many centuries back. Somewhere as time passed, they have gotten into a rut.

How does one get out of this rut? When do we learn to solve our own problems? When do we understand that it's ok to think independently?

Additional reading: (Updated on 16th July 2015)

1. http://www.thehindu.com/news/national/has-iisc-contributed-to-society-narayana-murthy/article7426651.ece
2. http://www.thehindu.com/opinion/lead/revamping-indias-scientific-ecosystem/article7492142.ece?utm_source=RSS_Feed&utm_medium=RSS&utm_campaign=RSS_Syndication

8/20/10, 5:52 AM

Pacific Daylight Time

Relation Between NT-SF, Natural Selection And Longevity

Of late I have been thinking about the MBTI types that are meant for survival and the ones that are not.

The statistics of percentage of people by MBTI type is available here. And you can see that the NT type, with all 4 subtypes ENTP, INTP, ENTJ, INTJ, forms about 10% of population while FP, with all 4 subtypes, is about 43% of the population.

If you split the types as NT, NF, SF, ST, each of the four types in theory should be 25%. As you can see, NT numbers are far lower at 10% and SF is higher at 43%. And majority of the SF's are women. Obviously there must be a reason why SF percentage is much higher than NT.

From another view point, we also know that the longevity of women is about 4-5 years than the average male as mentioned here and here. Women are in a way "fitter" than men and survive better. And perhaps for the same reason SF percentages are higher than NT's. Women form a much higher percentage of SF's than of NT's. SF is primarily a feminine trait while NT is a more masculine and cerebral trait. When SF percentages are so much higher than NT percentages in the population what does it say about the "fitness" of NT in the society?
By fitter I do not mean physically stronger. I refer to the fit as it is used in the expression "survival of the fittest".

Now we have the following data points. 

• SF percentage is much higher than NT percentage in the population. 
• Women live longer than men.
• Women form a higher percentage of SF than of NT and women outnumber men in SF.

What do these data mean?

The theory of natural selection says that we pass on those genes which help our species to survive and breed better. NT percentage is perhaps lower because we do not pass on the NT genes as often as we pass on the SF genes and that, in turn,  may be because NT's aren't well adapted to the environment. And as we have seen in the bullet points above female longevity is higher than male longevity.

Incidentally, research seems to indicate that our brain size has been getting smaller over the last 10 to 20 thousand years as per this, this, this and that we are becoming less intelligent than we were earlier. This seems to tie in with what I mentioned in the previous para - may be NT numbers (note that N's are more intelligent than S's) were higher earlier and these have become much less now. Even NF numbers are much lower than SF numbers. 

A quote from the 1st link: "Another likely reason for this decline is that brains are energetically expensive and will not be maintained at larger sizes unless it is necessary. The fact that we increasingly store and process information externally—in books, computers and online—means that many of us can probably get by with smaller brains. Some anthropologists have also proposed that larger brains may be less efficient at certain tasks, such as rapid computation, because of longer connection pathways. The way we live may have affected brain size. For instance, domesticated animals have smaller brains than their wild counterparts probably because they do not require the extra brainpower that could help them evade predators or hunt for food. Similarly, humans have become more domesticated. "

Quote from the 2nd link: "Over the past 20,000 years, the average volume of the human male brain has decreased from 1,500 cubic centimeters to 1,350 cc."... "The impact of these mutations remains uncertain, but many scientists say it is plausible that our temperament or reasoning abilities shifted as a result. Numerous phone calls later, it dawns on me that the world’s foremost experts do not really know why our organ of intellect has been vanishing. "

As per the 3rd link, "Geary and his colleagues used population density as a measure of social complexity, with the hypothesis that the more humans are living closer together, the greater the exchanges between group, the division of labor and the rich and varied interactions between people. They found that brain size decreased as population density increased. "As complex societies emerged, the brain became smaller because people did not have to be as smart to stay alive," Geary told AFP."

If this trend continues we will see that the brain continues to shrink, that socialization (F) becomes more critical than abstract thinking and NT - SF difference in population will become even more marked than it is now.
And either the future population will have more women than men or the population will have more SF's in it. Men also will become SF's increasingly. The 16 MBTI types, in future, may change into a different set of and perhaps numbers of types also.

One data I really want to get my hands on is the longevity statistics of NT and SF. I want to see if longevity of NT is lesser than that of SF.  I guess it will likely be so. 

In a lighter vein, thinkers should consider switching to feeling (becoming feminine) when they still can. Warren Buffet seems to feel the same thing - see "peals of laughter" here http://www.financialexpress.com/market/warren-buffett-predicts-dow-at-over-1-million-and-thats-in-a-conservative-estimate/865762/

Additional reading:
https://www.washingtonpost.com/news/business/wp/2017/07/27/grocery-stores-are-adapting-to-more-male-shoppers-who-they-treat-like-knuckleheads/?utm_term=.1026029792f9

Splitting A Blog Into Two Bloglets

Couple of days back I felt that I had to split my blog into 2 bloglets. I planned this for about 10 minutes and then got going. I decided that blog posts of a certain kind (or topic) would remain in the original blog (blog A) and the rest would go to blog B.

It wasn't as simple as that. There were some posts which weren't easy to decide as to whether they fit in blog A or B but then since the only person who would make the call was me it was simple.

A little bit into what caused me to split my blog - I felt that some of my posts were largely providing information - for example about laptops or mobiles or best places to eat or the railways booking policy or about an interesting book or blog I read recently etc. In such posts the value add I did was minimal to zero. I just provided the links. I moved all such posts to blog B. They were and are useful for me to refer to. 

Blog A (current blog) retained all the posts where I wrote my content and where external references were not much. There are some posts in this blog (A) where there are plenty of quotes and each of those quotes would make me think. Quotes that made and still make me think while writing or after writing ended up in A. And that's the way I want things to be. About 70% of all posts were moved to blog B.

Maybe in future i may go for one more round of disinvestment or pruning. 

Now as to how I moved the 70% posts - I did not now that I would end up moving 70% of the posts. At the end of the exercise it turned out to be so many: 

1. I took a backup of the blog and restored the same into blog B. So now blogs A and B were identical. While restoring I did not choose the "automatically publish on import" option. So for the general public the content was only available in blog A.
2. I chose blog B as my reference blog. Those posts that I knew would definitely be in A (by filtering on certain label values), I deleted from B. Similarly those posts that I knew would definitely be in B, I selected using labels in blog A and deleted them from A.
3. For the rest of the posts, working in blog B I went through each post and decided whether it would remain in blog B (in which case I deleted in blog A) and if not I deleted in blog B. Occasionally I deleted the post from both blogs when I thought I did not want that post at all. To make my job easier, I had opened two browser windows with blog A in the 1st and blog B in the 2nd window. So I could switch tabs and work fast.
4. Finally there was the issues of URL references - meaning some of my posts would refer to other posts of mine. Both posts were originally in A. After this splitting exercise when the referred post, which could be in blog A or B, was in blog B then the 404 error would come. So I had to change the references in referring posts to blog B from blog A. 
1. Initially I used a free online broken link checker which listed a whole bunch of broken links, some of my own posts and some external. But these were too many. 
2. I then searched within each blog separately for "blogA" (assuming that my blog address was blogA.blogspot.com). That gave me a bunch of posts in each blog. Some of those references would be Ok where the post was not migrated to blog B. Others would not be OK and the references would have to be changed. 
3. I would open each of the posts that came up from the search and then search inside the post for blog A and click on the hyperlink. If it opened successfully then fine, else I would search for that post in blog B and find the new URL and replace. Often it would be just easier to replace the blogA with blogB in the URL and then click the link to see if it worked. If not, of course go to blog B and find the correct address. Occasionally the referred post wasn't there in either blog A or B. God knows where it went. In such cases, there were 4 or 5 such posts, I just removed the reference.
4. This would still leave external references which may not work and also some of my own posts where the actual text in the hyperlink was not the URL but something more sensible. In such cases, seach for blogA would not show up. I would have to run a broken link checker to identify the remaining issues.
5. Having gone through this exercise I would have preferred it if Blogger gave an option to identify posts from blog A and also the option to move the selected posts to blog B 
1. while also deleting the exported posts from blog A and
2. changing the URL references in both blogs to the exported posts. Maybe such an option exists and I was not aware of it.

Anyway, version 1 of the bloglets A and B are now live. It took me about 30 man-hours (over 3 days) to do the complete exercise. The size of the exported blog was about 10MB. It is rather surprising that the export files of blogs A and B after the exercise are both about 5MB despite the fact that blog B has 3 times more posts than A (remember posts in A:B = 25%:75%).

Blog B is named Calmisc.

Recipe - Pulikachal

A cousin of mine makes excellent pulikachal. It's supposed to be spicy and mixed with rice and eaten.

Here is her recipe.

Mix tamarind with water, heat it a little, extract the tamarind juice and boil. Add groundnuts roasted for a couple of minutes. Sautee mustard, haldi powder. Add all this and salt to the tamarind - water.

Roast urad dal (a little), chana dal and red chilies, solid hing and very little methi and grind the mixture. Add this also to tamarind water. Add a little curry leaves also. My cousin said too much methi gives the product a sambar like taste. A lot of gingelly oil needs to be added. Groundnut skin has to be peeled.

Heat and stir the mixture until its like a paste. After I make it myself I will insert pictures.

Fried with Solid Hing, Haldi, Methi, Urad Dal, Chana Dal, Then added red chilies also

All fried stuff to be ground.Added salt also

Sauteed Mustard, Hing powder. Imli water is added, then ground stuff, curry leaves, lot of gingelly oil.

Final Pulikachal

Additional reading:

• http://vbala99.blogspot.com/2015/08/a-list-of-indian-mostly-tamil-lunch-and.html
• http://vbala99.blogspot.com/2016/08/making-sambar-and-rasam-powder-and.html

Estimating AIR (All India Rank) And Percentile

This post has to do with number crunching and perhaps would be only interesting to people who are interested in that area and who perhaps want to estimate the AIR and percentile from the NEET marks.

Now that NEET marks (out of a total of 700) are available, I have had to estimate the rank of a candidate who has got, say 600, marks. I looked up this link under the section "NEET 2017 Cut-off For AIQ Seats". As per this link, 14315 students will have marks from 685-468 marks (out of 700).

Now comes the interesting part.

How do I estimate the AIR (All India Rank) of a student who has got 600 marks?

The easiest way is to assume that all 14315 student marks are spread out equally between 685 and 468. In such a case, the rank of student-600 is about 14315*(685-600)/(685-468) =5607 approx.

But then this is very likely to be grossly incorrect because it assumes that the number of students between 685 to 585 is the same as between 585 to 485. In reality, perhaps student curve may be more like a bell curve with very few students near the maximum marks and again very few students near zero or minimum marks while the majority of students will be near the middle. How do we model this?

Here I assumed that students having marks between 685 and 10 marks less than 685 (= 675) will be 20. And students having marks between 675 and 665 will be 20 times a factor that is more than 1 (say 1.3). 26 students (=1.3*20) would have got between 675 and 665. Number of students who got between 655 and 645 would be even more than 26 (I assumed it will  be 26*1.3) =34. And so on. Until we get the number of students having marks between 475 and 465. If you remember, number of students having marks between 685 and 468 will be 14315.

Now that we estimated number of students in each slab of 10 marks, we can sum up and estimate the total students who got marks from 685 to 465. Now this may not equal 14315. If it is not equal then we tweak the factor which we assumed to be 1.3 until the sum total of students with marks between 685 and 465 matches 14315 more or less. And thus we get a good estimate of the number of students in each slab of marks 685-675, 675-665 and so on till 475-465. 

With this, we are in a position to find out the AIR of the student who got 600 marks. That rank will be approximately the number of students who got more than 600 marks. Let's say his All India Rank came out to be 737. And let's say 200,000 students wrote the NEET exam. The percentile for the student would be (200000-737)/200000=99.63. 

200000-737 being the number of students whose marks were lower than that of the student we are considering. Actually this itself is another approximation. There would have been many students who got the same 600 marks as our student. So the number of students who got lower than 600 would be less than 200000-737. But this is ok for an approximate estimate.

This is the spreadsheet which shows the calculations for NEET. Of course the estimates are heavily dependent on the assumption that 14315 students got marks between 685 and 468.

If anyone is interested in getting this estimate, you can leave a comment on this post.

There is another link which predicts NEET rank. The link predicts the ranks against marks as follows (data as on Jun 20 2017, 1PM)

Marks   Estimated Rank
680        1 -     10
670      11 -     20
660      21 -     50
650      21 -     50
640      51 -   100
630      51 -   100
620    101 -   150
610    151 -   200
600    251 -   500
590    251 -   500
580    501 - 1000

Additional reading:

• http://vbala99.blogspot.com/2017/06/neeting-medical-requirements.html
• https://www.quora.com/In-this-years-NEET-is-it-only-600-and-above-that’s-considered-a-good-score

22 Shey Shraabon

https://en.wikipedia.org/wiki/Baishe_Srabon: I watched this movie today on Amazon Prime. This is about tracking a serial killer in Calcutta..

I was reminded of the movie Bone Collector. Both movies seemed very similar in the way the killer leaves some clue behind about the next murder.  Of course one difference was that Denzel Washington's speech was very sophisticated (read upper class) while in the Bengali movie the guy used vulgar or coarse language despite probably being quite educated.

A very interesting scene in the movie, quite unrelated to the theme, was one where a guy and his live-in partner Raima Sen drive in the white car and they have a small tiff. Sen subsequently gets off the car and walks away. But the way she uses her eyes and tears him to pieces in a minute, without raising her voice nor using crude language, was so well done. So utterly feminine. And the guy was left speechless.

It was nice listening to Raima speaking in Bengali after having watched her only in Tamil movies. I do not know whether she needed dubbing for her Tamil or her Bengali or both or neither.

The English movie is much more watchable. The Bengali one just didn't move. I lost my patience.


Additional reading:
http://vbala99.blogspot.com/2012/11/adaptations-in-from-movies.html

NEETING The Medical Entrance Requirements - Identifying The Best Colleges From The Student's Perspective

I wrote a few days back about NEET and various quotas: AIQ (All India Quota), SQ (State quota) and caste quota.

This post tries to address the following question.

As a MBBS candidate (or her parent) how do we go about identifying the colleges? There are thousands of colleges in India, some government and some private. 

As a parent we want the best medical college for our child to get admission into. Most of us can't afford the pay for the management quota seats and we want our child to get into a college where he learns well and becomes a good doctor and in future...

So maybe we ask our friends about good colleges. We write down the names of great medical colleges such as AIIMS Delhi, CMC Vellore (I have left out mentioning Gypmer in this post because I am not sure how to spell it). The total number of seats in these colleges is about 500. Not all of those seats are UR (= Unreserved). And about 175,000 students are looking for medical seats in India. So where does that leave each of us?

Now if you / your child are a genius and get NEET All India Rank 10, you do not need this post. You read this link instead that a friend of mine sent me: http://www.theweek.in/theweek/cover/indias-best-colleges.html (go through the list of top medical colleges in the link). 

But if you are a bright student or are the parent of a bright child whose NEET AIR is perhaps around 5000 then this post will help you. If you / your child is likely to get AIR 12000, this post may still be of help to you. But you may need to shortlist more colleges than I have done. English Grammar aficionados can find out whether the verb that follows "if you/your child" should be is OR are.

What led to this exercise

A friend of mine and I collaborated together towards this exercise. While I may use the first person singular while writing this post, it was she who first made the master list of medical colleges and started evaluating colleges based on a set of parameters. I joined her in this exercise subsequently. I have good amount of prior experience in the medical field as described in the post here.

While looking at Google reviews for colleges or at collegedunia, college360.com sites, we found that the data or opinions were too narrow or too broad and there was no justification for the same. 

For example: A college was rated as follows: Faculty 5.6/10, Placements 4.7/10. How were these arrived at? Or more subjective reviews such as "Superb college, one day I want to become a student there", "I took my brother there and we were treated so efficiently by the doctors there". 

See the data was in either too refined a form or in such crude form - it made us uncomfortable.

What bothered us was also the fact that AIIMS (non Delhi) campuses consistently were rated "not high" by their own students. So the question we had was - what caused this "low" rating. We googled and got hold of a list of all candidates who got admission to all AIIMS colleges couple of years back. We started to ping some of them randomly and we connected to a couple of students through networking sites and chatted with them to get the insider information. There was also a report few months back in a leading daily about how an MBBS student from one of the newer AIIMS colleges felt bad about not getting adequate clinical experience. 

And we found in 2 separate cases that the students' exposure to patients was limited in the new AIIMS colleges (Rishikesh, Bhopal etc). And hence this, adequate exposure to patients, became a crucial requirement for shortlisting colleges. 

So we found ourselves facing the question: how could we identify the appropriate medical college for us / our children? If we had an all India ranking of all medical colleges based on an appropriate set of parameters, this post would have been unnecessary.

How the colleges were selected:

What we did was for a student from TN state. For students from other states the spreadsheet would be very similar - just change the candidate state from Tamil Nadu to the appropriate state in the spreadsheet. I often mention a spreadsheet in this post. Yes it exists and no, it is not public yet. 

Only Government colleges were considered because they are definitely cheaper and often better than private colleges. A total of about 90 colleges were selected. Which were those? 

• The best colleges such as AIIMS Delhi, CMC etc 
• The new AIIMS colleges such as Bhopal. Bhuvaneswar etc
• Colleges in domiciled state (in this example Tamil Nadu) - since 85% of the seats would be available for domiciled students (subject of course to various caste quotas), a good number of govt colleges in the state were listed down. 
• States were identified and rejected. AP, Telengana and JK were rejected because there is no All India Quota (AIQ) in these states. North eastern states, Haryana, Bihar, UP were rejected by me for other reasons - we didn't consider colleges from those states. Thank god, we are not rejecting women or handicapped people in such an arbitrary fashion in this post. Otherwise I would have to answer to feminists and handicappists and what not. Government colleges were thus identified from other states.

Thus the list of prospective colleges was formed from the point of view of an UR Tamil Nadu student.

And the focus was on male candidates. Hence Lady Hardinge College Delhi, being a college only for girls, was eliminated. And in AFMC only the male student quota was taken. For female candidates these would need to be tailored.

Getting the list of colleges was the easy job. On what basis do we shortlist the colleges - with special focus on the non-best and non-highly branded colleges?

Next we identified list of parameters to be collected for each college. 

As we started collecting data and analyzing we identified more parameters and collected data for those as well. These are the column headings in the spreadsheet. The data collected had to follow these rules:

• Data for parameters chosen should be available for each college in the public domain - hopefully recent data.
• Data for parameters should be meaningful.

What do we mean by meaningful? Meaningful data should help us with the decision to accept or reject a college. Data (or secondary data, typically ratios, derived from the data we collected) should be comparable across colleges.The data (or ratios) should help identify issues of inconsistency. For example consider a 1000 bed hospital having 100 deaths yearly (not including DOA cases). It is inconsistent because a 1000 bed hospital would have many more deaths than 100 in a year. 

The data we collected could also be inconsistent within a college and also across colleges for the same column (=parameter) because they were obtained on different dates (assessment pdf's made on different dates for the same college by the same or, as is more likely, by different assessors ). This would often be the case since all required data may not have been present in the latest pdf. Also, different assessors could also have brought in their own personal idiosyncrasies.

All data were obtained from internet and more than 95% of data from site: mciindia.org

Another friend who is a doctor provided some suggestions (she mentioned that even formal assessments could be tainted or incorrect - some colleges tend to organize additional beds or even doctors on the assessment date to fudge data much like in Munnabhai MBBS). Nevertheless I based my analysis on the assessment reports - I decided that those assessment reports were the most trustworthy than any other reviews I could lay my hands on. I am neither able to hack into the hospital sites nor do I have access to cabinet ministers. I am just a common man.

Based on these factors:

• The top colleges across India were tagged as best. There was no point analyzing those colleges. Data collection, analysis were not done for these colleges.
• Next came the new AIIMS colleges (Bhopal, Jodhpur etc) which were branded but being too new were not that good as colleges. Admissions into OPD were low for example. Data collection was not done much for these colleges - actually not much data was available anyway for these colleges. 
• Some colleges were rejected straightaway. These were tagged as "No". Opinions were entered in Remarks column in the spreadsheet as to why a college was rejected (or pushed to Yes-2, see next line)
• Some colleges were borderline yet accepted. These were tagged as Yes-2.
• Rest were accepted and tagged as Yes (meaning the college was a good choice). All that remained was whether a student got adequate marks to get selected into that college.

While categorizing colleges as Yes, Yes-2 and No there was some amount of trial and error. The process was a little subjective. Eliminating old (established in 1950s and 60s) colleges took some time to digest especially while at the same time accepting colleges established after the year 2000. We were uncomfortable shortlisting (tagging with Yes) any college established after 2010. We did select a few, we have mentioned the reason why in the spreadsheet.

Factors we considered to be important were:

• Number of outpatients daily in the entire hospital
• No. of beds (Supported by no. of surgeries, deaths, blood units consumed, X ray and CT scans per day). Low number of deaths probably means serious cases don't come to this hospital. Higher number of deaths preferred. For example if only patients with a cough come to a hospital then it's death rate is likely to be low. But do we want to get trained in such a hospital? 
• Level of Faculty Shortage
• Whether the college had Computerized Record Keeping
• Issues highlighted in MCI INDIA Assessment reports
• PG Intake per year (higher the better). If there were more PG students there would be more experienced teachers, the college would have more resources and better information exchange through senior-junior student interaction.

What did we achieve at the end: (note these are approximate figures and can go up or down based on the trial and error)

Of the 86 colleges identified totally, leaving the Best ones and all the AIIMS colleges out, we have shortlisted 30 colleges across India which together have a total seat of 594 available for a student from TN (applying 85% to TN colleges, 15% to other state college seats and also applying UnReserved quota percentages on top). We have further shortlisted another set of 16 colleges across India with total 237 seats available (with the same AIQ/SQ/UR percentages applied).

From the total 46 (=30+16) short listed colleges:

• Of the 594 Yes seats all over India, 302 were in TN and the rest were in other states.
• Of the 237 seats in Yes-2 all over India, 78 were in Tamil Nadu and the rest were outside.

While the Best, Branded and Yes colleges may be the Dream college for a prospective student, it is important to have a contingency plan. The Yes-2 colleges provide that Plan B. In our shortlist there are 13 Yes-2 colleges outside India and 3 within TN. 

Of course if CBSE's NEET exam do not form the basis for MBBS seat allotment, then the analysis of AIQ, SQ etc may not be current. The 85%, 15% allocation of seats may change as well as the basis on which students get admitted to a college. I wonder whether even the list of 46 short listed colleges would undergo a change in that case.

The detailed spreadsheet for public use is planned to be uploaded after a month. The links here and here are for beta testers.

In the next post, I have estimated the AIR and percentile for a student who got 600 marks out of 700 in NEET

Additional reading:
http://www.thehindu.com/opinion/editorial/lessons-from-a-fiasco/article19547540.ece.

NEET UG (Undergraduate) - Some Useful Links

For the year 2017, 

• 15% of all seats in a state are reserved for students of other states and my assumption is that students of a state (say TN) are not eligible for the 15% AIQ (All India Quota) for medical seats in TN. AIQ does not apply to medical colleges in Jammu Kashmir, AP, Telengana. All their seats will be available in their respective SQ only. See this link from Quora by Anjum Khan: Quote from there: "Furthermore, if particular XYZ college has 120 MBBS seats then, 85% of 120 seats i.e. 102 seats are for state domicile candidate and rest seats i.e. 18 seats are for All India students (candidates not resident of Kerala state)."
• 85% of all seats in the state are reserved for students of the state - SQ (State quota). For AAP,  Telengana and JK, 100% of the seats will be available for the respective state's students since there is zero AIQ for medical colleges in those three states. 

And within each quota (AIQ and SQ - I use AIQ and SQ frequently so make a note of what they mean), a good percentage is reserved for various castes. In this post we are only going to discuss the general quota for students not belonging to OBC, SC/ST etc.

https://www.toppr.com/bytes/neet-home-state-quota/: as per this link AIQ list will be announced first. And then individual states will announce their respective 85% seat allocations.

I assume a student will apply for both the AIQ (for other states only) and SQ (for his domiciled state only).

As per this link "students must understand that state of origin gets precedence over state of domicile while filling the application form. What it means is that if a student is originally from Andhra Pradesh but is a domicile of Gujarat, then in the column of state, the student is supposed to fill Andhra Pradesh." - What about Sindhis and Konkanis and Tulus? What is their origin? Karnataka for Tulus and Goa for Konkanis I guess. And for Sindhis? Hmm. And it becomes trickier. In some states, domicile is considered based on whether the student wrote the class 12 exam from that state. Other states look at whether his parents have a house there. So a student could be a domicile of both states BUT apparently it's illegal to apply to the SQ of more than one state. NOW, who is going to find out if I applied in TN and Assam? We will let deftly bypass these tricky questions.

A friend of mine sent me this link which seems to indicate that 3850 is the number of seats all over India available under AIQ. The link also gives the schedule for counseling for June and July 2017. 
As per this link, the total AIQ seats across India is about 2582 (and for TN it is 313 - I summed the figures for all colleges in TN in the table in the link). My guess is that the actual AIQ figure for 2017 would be somewhere between 3000 and 4000. Now, TN has 313 seats towards the 15% allocation and hence the SQ of 85% for TN will be 313/15% * 85% = 1773 seats.

About 50% of seats are available for general students (non backward ) in AIQ and about 25% seats are open/available for general students in SQ. 

AIQ for general (not backward) caste = 50% * 2582 = 1291 - assuming half the seats are available under general category or unreserved (UR)
SQ for general (not backward) caste =  25% * 1773 =   443 (assuming 25% available for UR)

Now let's consider 2 options

1. A student from TN gets AIR 1000 (which is less than 1291 seats available under AIQ) and state rank 40 and hence he gets the pick of his choice outside of his state. But if his pick is CMC Vellore, which is in Tamil Nadu, he won't be eligible under the AIQ. He has to wait for the SQ for admission into CMC. I assume in this case it makes sense for the student to apply both for the 15% AIQ (for other states) and for the 85% SQ. IF he hates to do medicine in his domiciled state he would apply only for the AIQ and not apply to SQ at all. The approximate seats available under AIQ and SQ are mentioned in italics above.
2. The student gets Rank All India Rank 10,000 and state rank 300. In this case he will not be eligible for the 15% AIQ since his rank is more than 1291. He can apply only to the medical colleges in his state. State wise merit or qualification marks data are here and here. I am assuming the student will get an intimation regarding counseling where he can "book his ticket on any state college of his choice" that currently has seats available.

Comments welcome. 


Additional reading:

• This is the list of students who got admission into all AIIMS in 2015: https://www.aiimsexams.org/pdf/Final%20list%20of%20MBBS-2015.pdf
• Estimated marks to get into top 25 colleges: https://www.quora.com/What-was-the-cutoff-rank-in-NEET-for-the-top-25-medical-colleges
• http://www.neetfortn.in/2017/05/05/neet-2017-expected-india-tamil-nadu-cut-off-marks/