Graduate admissions process makes no sense

[drb]

The graduate student admission process seems to be as follows: Shortly after the application deadline, a cohort of highly competitive students are admitted with funding. A second-level cohort are either admitted without funding, or are identified but not contacted. As April 15 draws near, decisions are made, and only a subset of the top-tier cohort accept (since they have multiple offers). Meanwhile, the second-tier group has visited elsewhere, gotten funding elsewhere, etc. Then what happens? Is there some frantic attempt to engage and recruit these second-tier students on April 14? Or does the school just pass on them and make offers to students who have not been accepted anywhere; i.e. they end up accepting a third-tier cohort? Or they just live with the subset of first-tier students who accepted, even if its not sufficient for filling TA-ships, etc?

Seems like an inefficient way to train the next generation of scholars.

[surprisecake]

In India we rank every single application based on merit and then release "merit lists".

For e.g. if there are 50 spots open, they first release the top 50 names. If 20 accept, they release the next 30 names for the remaining spots. They release a new list on a weekly basis until all their spots are filled! Its very time-bound and things get done a lot quicker.

[waddle]

In India we rank every single application based on merit and then release "merit lists".

For e.g. if there are 50 spots open, they first release the top 50 names. If 20 accept, they release the next 30 names for the remaining spots. They release a new list on a weekly basis until all their spots are filled! Its very time-bound and things get done a lot quicker.

This does sound a lot more efficient; what is the criteria for "merit"? Also, I'd imagine such an admissions system wouldn't be very personal--something that seems to be important to professional science in the U.S.

[surprisecake]

Yes, I did quite conveniently leave that part out

These days, each field has its own common entrance test for that subject and your program ranks you according to your score. Of course I much prefer the holistic application system they use in the States, but in India, this system is a great leveler. There are opportunities that lower income and rural students just don't have compared to city kids. They simply never have the chance at extra curriculars and meeting prestigious recommenders. Thats why I love the common entrance test system- you get in based strictly on your aptitude for THAT subject. Our academic system is so cut-throat because of our huge population; if admissions included intangibles, our schools would be filled with privileged city kids.

What would work better is a hybrid of both our systems. I believe the program I applied for at Berkeley uses a kind of numbers system. 60% of your application is your GRE and GPA and 40% is your LORs, SOP and CV.

Edited March 18, 2011 by surprisecake

[phoenix245]

I'd imagine a lot of schools get enough of an intake from their initial '1st tier' wave. For example, if the school is looking for 25 grad students to matriculate and from their past experience 30-40% accept, all they practically have to do is invite around 80 students for the Open House. If they choose this out of 500 apps, their work is almost done, except for a few odd ones out...of course this gets more difficult with smaller programs where there's higher variability in the percentage of people who eventually accept the offer.

[drb]

[

For example, if the school is looking for 25 grad students to matriculate and from their past experience 30-40% accept, all they practically have to do is invite around 80 students for the Open House

While this works for undergrad admission, I would think grad admission would be more finely tuned, given the costs. Consider your example - the difference between 30 and 40% is 8 students; at around $25K this is a difference of $200K to the department (not including tuition). It would be less for smaller departments, but as you note, the variability would be greater, and they might end up having to take a larger fraction of their class from the third-tier.

It just seems that, given the multi-year commitment of both students and school, there should be a little more communication with the second tier, so that there is clarity as to the degree of mutual interest. In particular, it amazes me that at this late date there are programs that have not even informed applicants whether they have been admitted.

Edited March 18, 2011 by drb

[starmaker]

These days, each field has its own common entrance test for that subject and your program ranks you according to your score. Of course I much prefer the holistic application system they use in the States, but in India, this system is a great leveler. There are opportunities that lower income and rural students just don't have compared to city kids. They simply never have the chance at extra curriculars and meeting prestigious recommenders. Thats why I love the common entrance test system- you get in based strictly on your aptitude for THAT subject. Our academic system is so cut-throat because of our huge population; if admissions included intangibles, our schools would be filled with privileged city kids.

But how do you determine who actually has research aptitude or fit with a department's faculty based on a test? A test of undergrad-level subject material doesn't tell you who's going to be a good researcher. And it doesn't test aptitude, just current ability.

I've generally thought of intangibles as a benefit to less privileged students, because it means that a program can take, for example, the fact that you didn't do well in your early college years because your poor high school didn't prepare you for college, into account, and not count it so strongly against you. Or, maybe you went to a low-quality undergrad college because you couldn't afford to go elsewhere, so you don't do as well on subject-matter tests as the people from top colleges, but you have a glowing recommendation from a faculty member at the more prestigious place where you did summer research, assuring the grad programs that you have great research aptitude.

[Eigen]

It just seems that, given the multi-year commitment of both students and school, there should be a little more communication with the second tier, so that there is clarity as to the degree of mutual interest. In particular, it amazes me that at this late date there are programs that have not even informed applicants whether they have been admitted.

I think a lot of this has to do with how large the "gaps" are between the tiers. I think a lot of what I've been seeing people say this year is that probably the first 3 tiers of students (ie, 3x the number of spots) are all excellent students they'd be fine with having.

As to the unfunded 2nd tier... A lot of schools will offer the funding from a first tier to a second tier student if possible.

[drb]

As to the unfunded 2nd tier... A lot of schools will offer the funding from a first tier to a second tier student if possible.

Indeed. But you would think that having gone to the trouble of identifying the desired students, they would be more proactive regarding communicating this likelihood. Otherwise the student may make a commitment elsewhere. Your point regarding the third tier being satisfactory is well-taken, but, again, given the many year and $ commitment, a modest amount of active engagement during March would be in order (vs. "check the website").

Further, those in the third tier who are still uncommitted close to April 15 may be largely those who have no funded offer anywhere, so you might be going pretty far down the list.

Edited March 18, 2011 by drb

[ringo-ring]

I've generally thought of intangibles as a benefit to less privileged students, because it means that a program can take, for example, the fact that you didn't do well in your early college years because your poor high school didn't prepare you for college, into account, and not count it so strongly against you. Or, maybe you went to a low-quality undergrad college because you couldn't afford to go elsewhere, so you don't do as well on subject-matter tests as the people from top colleges, but you have a glowing recommendation from a faculty member at the more prestigious place where you did summer research, assuring the grad programs that you have great research aptitude.

I agree that "intangibles" do not require preparation in particular subject, but what do they require then? Let's see! To get a good letter of recommendation, you need to have a good relationship with recommender and somehow make him think good about yourself. To write a good statement of purpose, you have to know how to convince the reader, hot to form the opinion you need through writing. In short, these are all social skills, not research skills.

And disadvantaged background affects these social skills much more than it does to IQ. Even worse, there is no easy or obvious way to compensate for this! If you had bad education, for example, there are LOTS of free ebooks and courses available online for free, so anyone who was enough passion and determination can educate himself up to the any level required by school. However, for soft skills there are no obvious ways to train them up.

So my opinion would be that intangibles are actually shortcuts through which mindless kids of privileged parents can make their way into grad school, even if they do not have any dedication and/or smartness to study to get a good GPA/GRE scores.

[drb]

Um - can we keep this thread on topic? If you wish to discuss the merits of stat vs. intangibles-driven admissions, and its implications for less-privileged applicants, please feel free to start another thread. Thank you.

[DariaIRL]

Admissions is definitely a complicated dance, because there are so many players and things can change slowly or quickly. The first hump applicants need to wait for is the processing of their materials. Most everyone puts off actually submitting their materials until the DAY of the deadline, which can sometimes double or triple the workload of the office for processing these.

Then, as you've described, the first tier people are identified and go through interviews and perhaps receive offers. These accepted individuals will most likely wait until close to April 15 to decide, which leaves those in the second tier doing their own thing while HOPEFULLY the school is letting them know where they stand. An offer can be made after April 15, and usually it's for someone who didn't get in anywhere [that they liked?] or maybe only applied to the one school.

Some programs send out more offers in the first round, with the thought that a portion will accept (such as make 40 offers when you're hoping for 30). In this case, there may not be a second tier if they've already over-offered.

It's not a smooth process, that's for sure!

[basst]

I think most people get into some schools and waiting for some schools. The problem is if everyone decide really close to April 15, schools don't have enough time to fill the empty spots. How can applicants decide on Apr 15 if they are still waiting for the results?

[rising_star]

The graduate student admission process seems to be as follows: Shortly after the application deadline, a cohort of highly competitive students are admitted with funding. A second-level cohort are either admitted without funding, or are identified but not contacted. As April 15 draws near, decisions are made, and only a subset of the top-tier cohort accept (since they have multiple offers). Meanwhile, the second-tier group has visited elsewhere, gotten funding elsewhere, etc. Then what happens? Is there some frantic attempt to engage and recruit these second-tier students on April 14? Or does the school just pass on them and make offers to students who have not been accepted anywhere; i.e. they end up accepting a third-tier cohort? Or they just live with the subset of first-tier students who accepted, even if its not sufficient for filling TA-ships, etc?

Seems like an inefficient way to train the next generation of scholars.

It starts happening well before April 14. That's why everyone encourages people to turn down offers as soon as they know they won't go somewhere.

Also, schools can issue acceptances with funding AFTER April 15 if they want to and many do if they still have spots available. But, in this economy, many schools are probably hoping that not everyone they accept enrolls because it will cause pressure on their budgets now and in the future.

I think most people get into some schools and waiting for some schools. The problem is if everyone decide really close to April 15, schools don't have enough time to fill the empty spots. How can applicants decide on Apr 15 if they are still waiting for the results?

You make a decision based on the information available. If you change your mind after April 15, you have to get a written release (if the school is a CGS member and your offer included funding) before you can commit to another school. This is not particularly difficult to do, but must be done tactfully so you don't burn any bridges. If you check past years of the "April 15th" forum, you'll see questions and answers about this sort of thing.

[waddle]

I'm thinking a "match" system--like that of U.S. medical residency admissions--would work wonders for graduate programs. Basically, this involves a centralized application system where the applicant inputs a list of programs ranked by preference, and the individual programs also rank the applications they get by preference. And on April 15th, the system runs the algorithm and "matches" students to programs. So, a student would get one--and only one--acceptance for the entire application cycle (i.e., take it or leave it), and the grad schools won't need to scramble to cover gaps left by applicants who decline an offer.

[Ujjawal]

I absolutely agree with the point that admissions should be less based on the intangibles. There are some schools where it is next to impossible to find a superstar professor who could write a recommendation for you. Not all schools have superstar professors. Also when we try to get in contact with some star professor some where else, no matter how relevant our e-mails are they never reply to our mails. So again one more intangible comes into picture and that is "luck". If you are lucky enough to get in touch with a professor then you have the chance, otherwise how passionate you are about doing a phd makes no sense to them. But the matter of the fact is we are the applicants and we have no authority to do anything but to just accept the decision whatever it might be. So in my opinion this thread can be of a good use to just let out our frustrations because after all we will not be able to change anything.

[Ujjawal]

I'm thinking a "match" system--like that of U.S. medical residency admissions--would work wonders for graduate programs. Basically, this involves a centralized application system where the applicant inputs a list of programs ranked by preference, and the individual programs also rank the applications they get by preference. And on April 15th, the system runs the algorithm and "matches" students to programs. So, a student would get one--and only one--acceptance for the entire application cycle (i.e., take it or leave it), and the grad schools won't need to scramble to cover gaps left by applicants who decline an offer.

I completely support this system. By this the applicant will be happy to get the school according to his/her preference and even the University will be happy to get a student whom they wanted. The applicant will also be well aware of his/her in terms of preference and will be able to decide where he/she wants to go well in advance before applying. No redundant admissions so everyone is happy.

[drb]

I'm thinking a "match" system--like that of U.S. medical residency admissions--would work wonders for graduate programs.

While theoretically attractive, this is probably not feasible. Firstly, there is the "herding cats" aspect of getting graduate departments to agree - medical residency programs are much more regulated by and accountable to their accrediting boards. Second is the highly varied constituent departments. For example, a student interested in neural engineering may end up applying to a biomedical engineering program, an electrical engineering program, and a neuroscience program. Getting all of these into the same match system would be difficult, as their applicants would otherwise not likely overlap. (Residencies, in contrast, are much more homogeneous in nature - you pick surgery, or pediatrics, but rarely both). Finally, this would ultimately be a disservice to the student. In contrast to residencies, alot of a student's evaluation of a program is made after admission, when in-depth consideration of and interviews with individual PIs are done.

[runonsentence]

While theoretically attractive, this is probably not feasible. Firstly, there is the "herding cats" aspect of getting graduate departments to agree - medical residency programs are much more regulated by and accountable to their accrediting boards. Second is the highly varied constituent departments. For example, a student interested in neural engineering may end up applying to a biomedical engineering program, an electrical engineering program, and a neuroscience program. Getting all of these into the same match system would be difficult, as their applicants would otherwise not likely overlap. (Residencies, in contrast, are much more homogeneous in nature - you pick surgery, or pediatrics, but rarely both). Finally, this would ultimately be a disservice to the student. In contrast to residencies, alot of a student's evaluation of a program is made after admission, when in-depth consideration of and interviews with individual PIs are done.

Exactly my opinion.

The match system is actually currently under a bit of debate, as it means that residents can't negotiate for their pay: they are contractually obligated to attend the university/hospital/program they are matched with. (Now, of course, most doctors will be making high six-figure salaries just a few years down the road, but the point is still there.) We, meanwhile, do have the flexibility to negotiate, to a certain extent.

Further, most universities don't know what their budgets for the next academic year will look like until March or April, making funding something that's impossible to know about during the application process. Much of any applicant's decision rests on whether or not she received funding, and how livable/competitive that funding is. A match system would take away an applicant's ability to take funding into account in her decision-making process.

Edited March 19, 2011 by runonsentence

[Eigen]

I absolutely agree with the point that admissions should be less based on the intangibles. There are some schools where it is next to impossible to find a superstar professor who could write a recommendation for you. Not all schools have superstar professors. Also when we try to get in contact with some star professor some where else, no matter how relevant our e-mails are they never reply to our mails. So again one more intangible comes into picture and that is "luck". If you are lucky enough to get in touch with a professor then you have the chance, otherwise how passionate you are about doing a phd makes no sense to them. But the matter of the fact is we are the applicants and we have no authority to do anything but to just accept the decision whatever it might be. So in my opinion this thread can be of a good use to just let out our frustrations because after all we will not be able to change anything.

I think a lot of people focus on getting a recommendation from a "star" professor, and I think that is not at all necessary. A professor that knows you well and thinks you can do well in graduate school is all that's needed.

Graduate admissions really isn't all that complex: Do well in classes, have a good understanding of the base material, have several years of in-depth research experience (that you can explain and discuss well) have some good extra-curriculars/awards/society memberships, and have 3 professors that know you well enough to write good, personal letters for you by the time you graduate.

[Eigen]

Exactly my opinion.

The match system is actually currently under a bit of debate, as it means that residents can't negotiate for their pay: they are contractually obligated to attend the university/hospital/program they are matched with. (Now, of course, most doctors will be making high six-figure salaries just a few years down the road, but the point is still there.) We, meanwhile, do have the flexibility to negotiate, to a certain extent.

Further, most universities don't know what their budgets for the next academic year will look like until March or April, making funding something that's impossible to know about during the application process. Much of any applicant's decision rests on whether or not she received funding, and how livable/competitive that funding is. A match system would take away an applicant's ability to take funding into account in her decision-making process.

Not to mention that the match system also doesn't allow you to definitely specify an *area* for your residency. I can't imagine many grad students being happy with getting shunted from say, chemistry to physics or biology just because there were more openings in those fields/they weren't competitive enough for chemistry (choosing semi-random related fields here, it applies across the board).

[The Dudester]

I absolutely agree with the point that admissions should be less based on the intangibles. There are some schools where it is next to impossible to find a superstar professor who could write a recommendation for you. Not all schools have superstar professors. Also when we try to get in contact with some star professor some where else, no matter how relevant our e-mails are they never reply to our mails. So again one more intangible comes into picture and that is "luck". If you are lucky enough to get in touch with a professor then you have the chance, otherwise how passionate you are about doing a phd makes no sense to them. But the matter of the fact is we are the applicants and we have no authority to do anything but to just accept the decision whatever it might be. So in my opinion this thread can be of a good use to just let out our frustrations because after all we will not be able to change anything.

The other side of this is that if the intangibles mean nothing, then you better ace your GRE. Otherwise, just quit and never try. There is more to people than test scores and GPA.

[lewin]

Yes, I did quite conveniently leave that part out

These days, each field has its own common entrance test for that subject and your program ranks you according to your score. Of course I much prefer the holistic application system they use in the States, but in India, this system is a great leveler. There are opportunities that lower income and rural students just don't have compared to city kids. They simply never have the chance at extra curriculars and meeting prestigious recommenders. Thats why I love the common entrance test system- you get in based strictly on your aptitude for THAT subject. Our academic system is so cut-throat because of our huge population; if admissions included intangibles, our schools would be filled with privileged city kids.

What would work better is a hybrid of both our systems. I believe the program I applied for at Berkeley uses a kind of numbers system. 60% of your application is your GRE and GPA and 40% is your LORs, SOP and CV.

I can't imagine the GRE-related complaining (or complaining about its replacement) that would occur in these forums if admissions were based on standardized test scores. "A single test would even the playing field!" versus "I'm more than just a number!" would be an epic battle.

[Eigen]

And would fall directly upon the lines of people who did/didn't do well.

The GRE is far too simplistic to be used as an overall measure of an applicants ability.... Even the subject GRE. Really all they tell you is how well someone can study for a test- they don't show you how productive the applicant can be, how well they work on projects, how creative they are, etc.... That's what you get from the SoP and the letters of rec... The full story about the applicant you're accepting.

Honestly, a good GRE holds little to no correlation about how you'll do in graduate school- it just shows that you have some knowledge of very basic math and a decent vocabulary. Test scores and transcripts are nice numbers to have- they show some baseline academic performance. They say next to nothing about how you'll do in graduate school.

Success in getting a PhD is more about drive and determination then it is about grades and test scores... And you can't read drive and determination off a transcript or from test scores. Adcoms want to take students who seem genuinely interested in the field, because without some deep interest it's really hard to last out the years. Interest is (of course) another characteristic you can't see from test scores and transcripts.

[equinox]

"Graduate admissions really isn't all that complex: Do well in classes, have a good understanding of the base material, have several years of in-depth research experience (that you can explain and discuss well) have some good extra-curriculars/awards/society memberships, and have 3 professors that know you well enough to write good, personal letters for you by the time you graduate." -- You are absolutely right, I wish someone had told me that when I started my undergrad.

As for the matching system- there is NO way it would work. There are too many subfields, and too many programs where to have to name an advisor. The medical profession is far more straight forward- there are no research interests involved, and the credentials are way more standardized. There would have to be a separate matching process for math, applied math, psych, chemistry, physics... and so on. And then what happens to interdiscplinary people? I applied to pure math programs, applied math programs, computational biology programs, and Ecology and Evolution programs. They all had people I would have loved as my advisor, and I would have done almost the same research in each place. I got into one pure math, one Ecology and Evolution, and wait listed at an applied math. How on EARTH could a matching system deal with that?

And I only figured out which I wanted to go to after extensive vists and discussions with potential advisors.

As for the indian system- I call bulls***. Graduate school is about doing RESEARCH, not how well you learned a particular subject area in undergrad. It's about CREATIVITY, which cannot be tested. Knowledge is important too, but it is ONLY the foundation of the building, not the structure.

I did poorly on the math GRE because it had been 7 years since I took calc I and II and diff eq in high school. I was taking graduate math courses that semester, so I didn't want to devote half of my time to preparing. Does that mean I don't know calculus or differential equations? heck no, I just can't do the problems in 2.5 minutes, it takes me maybe 6 to figure it out, instead of spitting out a formula and then doing a small calculation.

The point is, Some graduate schools held it against me, others (ranked just as highly) didn't care. If every school uses the same ranking system, how can it NOT be discriminatory to some people? The strength of the US system is in its diversity. If one school is biased for a particular reason, the other won't be.

Edited March 21, 2011 by equinox