Any ideas on what GRE/GPA cut-off scores are?

[lxs]

I read this article:

http://chronicle.com/forums/index.php/t ... 917.0.html

posted in another thread about the application process which says that applications which don't meet the cut-off for GRE or GPA don't even make it to the department. I know this varies by school, but does anyone have any ideas or knowledge of what scores are used to weed people out? I asked at all the schools I applied to but was given responses along the lines of "there is no cut off score" or "we consider everyone based on their entire application and not just the GRE." My score wasn't abysmal but also wasn't amazing and so the Chronicle forum post has me pretty nervous.

I'm applying to PhD programs in the Humanities, FWIW.

[Canadianpolsci]

It really is true that most schools no official cut offs. However, GPA and GRE are the easiest comparable data sets. Every competitive program needs to weed 30-50%, or more, of the applications; this is natural, as, aside from a few "for sure admit" starts of each round, a lot of applicants are going to fall into the "strong middle" category, and it requires effort, ratings, discussion, the works, to figure out which strong applicants to reject.

This means that, whether it's one poor junior prof., or each member individually, there has to be a weed out phase. In a way, it does not matter if there is a rule: "GPAs under 3.5, GREs under 1250 combined" or if it is ad hoc: the result is about the same.

Put yourself in the committee member's shoes, and you can see how it works. You have a pile of fresh applications no one has seen. They have either been sorted by sub-category (eg., in political science: theory applicants, IR applicants, etc.), or they have not. Either way, what does Junior Professor Smiley do with each app? The same thing: first he checks: a)what school the applicant went to b)GPA and GRE c)does he know any of the letter of reference writer's or recognize any names at least. Immediately all of the above info results in the frame of mind decision:

the applicant looks a)very strong b)strong (here, which school you went to is crucial...'He went you Yale'=he is certainly not DUMB) c)the scores are a bit low, and the school ain't so great d)the scores are too low, and the school ain't so great.

Then he has to weed. He will weed all the D)s. He will then likely weed half of the C)s.

If everything looks in order EXCEPT one main date-point, he checks: school pedigree.

In all cases where he can't easily arrive at a c or d, he reads the statement of purpose. In the cases where he is almost sure it is a c or d (weed or weed with great prejudice), he reads the statement of purpose QUICKLY.

If you are in the A) or "frame", he reads your application in its entirety CAREFULLY. If you are a C) or D), he reads everything as well, but he does so QUICKLY.

At the end of reading EVERY App in its entirety (or for some schools, only the A)s and B)s and some C)s ), he assigns a numerical value. The ranking system and breakdown varies. According to a paper published in a political science journal on the "Science of Political Science Admissions" in 1995, Harvard Government would rank out of 14. Why 14? Because 1/14 is about 7%, and that is also the average acceptance rate. In other words, you better get 11-14 to make it to serious final round debates in full committee.

How about the final rounds? For the low GPA/GRE folks: you better have an incredible reason to have gotten this far!!

How does the final round work? That's not clear. Certainly there are a small handful of students who just knocked the socks off everyone who read their app. They are in. That is maybe 1/4 to 1/5 of the spots. The rest are haggled over. Either you appeal to a few profs, or one prof in your area maybe just loves you, and fights for you.

And the rest, as they say, is history.

[ahug030]

Thanks polsci - very informative

[lxs]

wow - thanks a lot polisci! Very informative and thorough.

[crossedfingers]

wow!!! it almost sounds plausible!

I'm pretty sure thats how it functions, but my prof told me that it also depends on the person. he is in the admission committee for grad school and he said he begins reading the SOP. WHen he gets a feel for who the candidate is, he looks at the numbers (GRE/GPA). If for some reason they dont match up, (interesting SOP, not so great scores), he tells the committe that they should interview the person and find out more. However he did warn me, not everyone does it the same way.

i personally think its a big BAD gamble!!

[Minnesotan]

I have seen some programs, normally in the top 20 humanities ballpark, that explicitly state a GRE cutoff on their application FAQ. They are usually ridiculously low, though, like 600-650v for English or history programs.

[Canadianpolsci]

I am sure that certain schools, certain programs, have minimums. But yes, they are significantly lower than the average score of ADMITTED applicants.

Mr. 'I hate the GRE' is a regular on these forums. We know his problem: he screwed up the GRE. Still, he is going to plunK down 60 bucks, 5 times over, to apply to schools that admit less than 15% of applicants. Why? He is taking a chance.

Mr. 'I hate the GRE': do you best. Some places will screw you early, most will omit you from the final cut list. You will never even know who, what or why.

My advice to this type of dude: if you don't suceed, re-apply with new GRE scores. Can't get GRE score of over 650 in each category and a 5 in writing (and you are not a foreigner): don't go to grad school. You think these applications are rough, wait till you see the job application competition.

What the difference between a PhD application process, and a Academic Job process --competition wise: THAT EASY! EVERY LAST DARN FELLOW AND LADY APPLYING FOR JOBS ALREADY GOT INTO GRAD SCHOOL -- EITHER WITH GOOD GRE SCORES (DUH) OR BAD ONES (WOW!)

The PhD application process is not quite the real world. It gets HARDER from here.

[Canadianpolsci]

I hope you find these posts of mine informative and/or amusing.

I am aiming very very high this fateful application season. If my horse does not come in, I'll need to find something to do for a year, while I re-apply. And my idea is: develop a personal tutoring service to help people get into Law, MBA and PhD programs. I am completely serious. Some might as well pay you to learn from your own mistakes.

[Intaglio5]

I have seen some programs, normally in the top 20 humanities ballpark, that explicitly state a GRE cutoff on their application FAQ. They are usually ridiculously low, though, like 600-650v for English or history programs.

650v is low at 93-95%?

[Minnesotan]

If you want to go into a field like English or History, yes. If you can't break into the 90th percentile as a potential humanities PhD, you really have no business applying to grad school. You definitely shouldn't have received a bachelor's degree in English!

Less than 1% of Americans earn PhDs. Shouldn't these people at least be in the top ten percent in all of their academic pursuits?

[dubitocogito]

Canadianpolsci mentioned "The Science of Political Science Graduate Admissions" article. For those who have not read it, here is a link to a pdf of the document: http://gking.harvard.edu/files/PS93.pdf.

[flyingwalrus]

Less than 1% of Americans earn PhDs. Shouldn't these people at least be in the top ten percent in all of their academic pursuits?

You're assuming that the GRE accurately measure what they puport to. My understanding is that the quantitative score is generally a much better guage of an applicant's abilities than is the verabl score. You may be interested in this post on a blog maintained by several political scientists: http://www.themonkeycage.org/2008/01/post_28.html

[crustaceangirl]

90th %ile on the GRE doesn't put you in the top 10% of the general population, it puts you in the top 10% of people who took the GRE, i.e. college graduates who are applying to graduate programs.

[farty14]

Does Quant score matter at all for History PhD admissions. I would assume it does not.

[flyingwalrus]

Does Quant score matter at all for History PhD admissions. I would assume it does not.

I bet it doesn't matter too much, but a good score will naturally look better than a poor one. Some quant skills could actually be valuable for some historians though, so I wouldn't write it off.

[dubitocogito]

Does Quant score matter at all for History PhD admissions. I would assume it does not.

The admissions officer at Boston University emphatically told me that quantitative does not matter. However, it seems to definitely matter at more competitive programs.

Some schools post this information. Here are some links I have gathered:

-UC Berkeley (http://history.berkeley.edu/graduate/faq.html): "For 2007, the average GRE scores were 91% verbal; 78% quantitative; and 5.5 analytical writing on a scale of 1 to 6."

-Duke University (http://www.gradschool.duke.edu/about_us ... ithist.htm): 646V, 623Q.

-University of Virginia (http://www.virginia.edu/history/graduat ... cation.php): "The average scores of applicants who were offered admission for the Fall of 2007 were 666 Verbal, 644 Quantitative, and 5.3 Analytic Writing."

-William and Mary (http://www.wm.edu/history/grad/faqs.php): "The students admitted for the fall of 2007 had an average verbal score of 653 and an average quantitative score of 630."

-University of Maryland (http://www.history.umd.edu/graduate/adm ... lines.html): "The average GRE scores, expressed as percentiles, were: Verbal: 87%; Quantitative: 59%; Analytical: 74%."

-University of Minnesota (http://www.grad.umn.edu/data/stats/ad/1038000.html#gre): 645.6V, 664.7Q for 2006-2007.

-Princeton also posts their applicant and admit GRE information (http://www.princeton.edu/pr/facts/profile/07/graduate/).

Best of luck. I'm applying for History as well.

[KUNRD07]

I bet it doesn't matter too much, but a good score will naturally look better than a poor one. Some quant skills could actually be valuable for some historians though, so I wouldn't write it off.

From the Harvard study posted above:

"Based on our analyses, the quantitative GRE score is a better predictor of grades in graduate school than the other two scores, regardless of the field of study within the Government Department. This is not because mathematics is essential to a graduate education at Harvard, but, because mathematics questions are by their nature less ambiguous, it appears to be a better test. Moreover, quantitative, verbal, and analytical skills correlate very highly, and so a high score on a math test, even for an aspiring political philosopher, often predicts success in his or her field."

[Minnesotan]

90th %ile on the GRE doesn't put you in the top 10% of the general population, it puts you in the top 10% of people who took the GRE, i.e. college graduates who are applying to graduate programs.

But the flaw in your reasoning is this: 100% of PhD applicants (at reputable, accredited programs) are college graduates. The criterion for inclusion in our little group here is a bachelor's or master's degree, not a pulse. PhD candidates should be the cream of the college crop, not just anyone who can't figure out what else to do with their lives. And even if you have a strong desire and a clear vision of your career goals, that doesn't mean you have the tools to succeed.

Think of it like this: you want to be a professional hockey player. Your skating speed and power is excellent. You're a strong, physical player. You can shoot from the hashmarks, inside the crease, or back at the blue line with a high level of relative success. Unfortunately, your stick-handling is at the 55th percentile.

Should a team still pick you if you skate (sorry - pun) through life being just above average? Not putting in the effort to study for a standardized test? Whining instead of proving people wrong? If you've lived the majority of your life in an English-speaking country, you should have had an easy enough time learning the few words you didn't know from the word lists. If English is not your native language, then you should be taking the TOEFL, or equivalent test. Honestly, we've all (Americans, anyway) been taking these same achievement tests since second grade, using the same word-relationship exercises/analogies. Even if you didn't grow up taking these tests, you would have had to completely forsake all preparation for the examination to have not understood what you were getting yourself into!

Is the GRE perfect? Probably not. Is it fair and easy to study for? Yes.

Why blame the test when you don't prepare enough (for you, using your own successful study habits and test-taking strategies)?

p.s. I don't mean to be argumentative. It's just that the constant GRE whining really annoys me. People blame the test when they fail in any pursuit, when most often it's their own fault.

[Minnesotan]

From the Harvard study posted above:

"Based on our analyses, the quantitative GRE score is a better predictor of grades in graduate school than the other two scores, regardless of the field of study within the Government Department. This is not because mathematics is essential to a graduate education at Harvard, but, because mathematics questions are by their nature less ambiguous, it appears to be a better test. Moreover, quantitative, verbal, and analytical skills correlate very highly, and so a high score on a math test, even for an aspiring political philosopher, often predicts success in his or her field."

Precisely! For jobs where non-linear thinking is a nuisance (i.e. the government) more than a desired commodity, the verbal holds little meaning. For fields where one might have to use abstract problem solving skills and linguistic creativity (say, History, English, or Philosophy) the verbal might actually be a useful benchmark.

The tide of GRE naysayers never ends. Sure, I did well, and I have a vested interest in defending the relevance of my score, but I also believed - even before I took the test - that the test was administered as fairly as possible to all test-takers, and that it had value. I also realized that it's a straightforward test that anyone can study for successfully. If you didn't study properly, or hard enough, then maybe that's another indication of your suitability to grad studies.

[ahug030]

But it must be said that anyone studying for a standardised test like the GRE has kinda missed the point - it's supposed to be designed to test skills accumulated over years, not one's ability to cram from wordlists. But then again, people do study for the GRE for >2 months in some cases. Why is it fair that someone should get good scores just because they can prove that practice makes perfect? You could get someone without any knowledge of english, get them to rote learn a huge wordlist over a period of months and they'd probably get 600v. The whole idea seems silly... I got 640v and 760q after studying for 5 hours total. I guess I should count myself lucky but I can't stand it when people say - 640V/760Q RATE MY CHANCES AT X UNIVERSITY etc. when any professor with half a brain would dismiss the GRE as a formality rather than a true indicator of research capability.

My point is that the GRE probably measures a candidate's self-perceived degree of inadequacy (=motivation to study) rather than some underlying goodness as a PhD candidate.

I'm ranting now, but as a final note has anyone looked at the correlations between e.g. GPA and GRE verbal score?? As I recall they're very low - like p=0.2 or something. Someone correct me if this is off?

[crustaceangirl]

Minnesotan, I didn't mean to imply that a good GRE score is not important in the admissions process or that it doesn't indicate anything. I was just pointing out the inconsistency between saying that less than 1% of the general population gets a PhD, thus one should get such-and-such percentile on the GRE to be admitted to a PhD program. Being "average" for the population of college grads seeking further education (which I would argue is a different population than college grads in general -- a lot of people have a "rubber stamp" view of a bachelor's degree and consider it obligatory, but would never think of going to school for longer than they feel is strictly necessary) is not an awful thing to be. Yeah, you won't get into a top-tier school, but there are schools in the middle for people whose application stats are in the middle.

Anyway, I'm not trying to be argumentative either. I tend to do well on standardized tests, so I didn't stress over the GRE and was happy with my scores. Just wanted to explain my comment in a little more detail (sorry I didn't use any hockey analogies... as a Floridian, I have no idea what the heck hockey is about).

[canuck]

In my humble opinion, motivation to study and ability as a PhD researcher are virtually equivalent. In all but the most esoteric fields there is such a colossal amount of literature to wade through to get from the undergraduate level to the 'state-of-the-art'. If the GRE is a measure of your attitude then perhaps it is a better indicator of research potential than some would hope to believe; ie, treating your PhD studies like 'I don't care if other people work hard, I'm gonna just put in a few hours and get by on my brilliance' is not going to get you anywhere just like it doesn't on the GRE.

[Minnesotan]

But it must be said that anyone studying for a standardised test like the GRE has kinda missed the point - it's supposed to be designed to test skills accumulated over years, not one's ability to cram from wordlists. But then again, people do study for the GRE for >2 months in some cases. Why is it fair that someone should get good scores just because they can prove that practice makes perfect?

While I understand your sentiment, I would counter that if you want to get into grad school, the best way to do it would seem clear: do the best job you can on all aspects of your application. This means studying for the GRE, even if it is not in the spirit of why we use standardized testing.

(sorry I didn't use any hockey analogies... as a Floridian, I have no idea what the heck hockey is about).

This, the unkindest cut of all!

Ermm... I think you're right Crusty - we're arguing uselessly. I agree with the distinction you made between professionalizing bachelors and bachelors who are motivated to continue studies; there is often a difference in terms of desire. But, in the end, however we frame the numbers, it's the intent of the argument that counts. I was merely trying to say that a potential humanities grad should score very high on the verbal, and that most of the top universities do take notice if someone has less than a 650v.

Yes, the "cutoff" limit gets lower as we work our way down from the ivies/public ivies to the less prestigious programs. However, a strong verbal score should be important, no matter what school you're applying to (in the humanities, I mean - scientists don't need to be able to string two sentences together). =)

[LaraAnn85]

I think that GRE scores and GPA are generally a good way to assess someone's possible ability and performance. BUT, there are times when those numbers simply cannot tell you enough about a person. I have done research for over a year now- running my own experiments, putting in way more hours than required, constantly being complemented on my work, co-authored a paper, etc.- which is what you look for in a graduate student. During the past two years I have had about a 3.7 GPA with an upward trend. All is good and well until you look at my first two years' GPA- probably about a 2.4, if that. However, what my transcript cannot tell you is that I was pre-med and suffering from an eating disorder, so how is that fair? Once I recovered and changed my major and figured out what I wanted to pursue, I was a star candidate. But because of those two years, my overall GPA is not competitive and I'm sure a lot of schools won't have a chance to see that I would make a great graduate student. My GRE scores are good, but not amazing because I didn't have time to re-take them while taking a full load of classes and working in a lab. So in my case, I am hoping that LOR and research experience can get me further than raw numbers. But that's just me.

[ahug030]

I found the section on the 'standardized tests' interesting in the article mentioned earlier (http://gking.harvard.edu/files/PS93.pdf) - especially the bit about how high the uncertainty is in the GRE results for individuals (+- 80 points at 95% confidence). A good point is "...all students who receive scores of above 688 are indistinguishable (statistically)". But then the article backtracks a bit and claims that on the whole the GRE does provide a good way of distinguishing between students... so I guess the bottom line is, the GRE is the worst way to judge aptitude, except for all the others (to borrow a turn of phrase).