How was the Oct 2012 Test? + My experience.

[Legendre]

Some thoughts after the test. I am not sure if we all have the same paper, so this might not be applicable to you.

1. The earlier questions appears to be easier than the ones at the back.

2. Time is a BIG issue. Questions in the later half (number 30-66) had so much text and took some time to set up + solve. I am not sure how to do them within 2.5-3 minutes. The initial questions were easy though: I did the first 20-22 questions in just the first hour. Got really hammered and demoralized during the 2nd hour though, and it went all the way downhill from there.

3. Difficulty level: feels like when I was doing the 2005 test (official GRE practice book) for the first time, which is far more difficult than Princeton Review or the older practice tests.

Overall, I did badly. Only managed answered 40 questions.

Wasted too much time (7-10 mins!!!) trying to solve questions I am stuck on, really need to force myself to move on after 3-4 minutes! Got really discouraged and distracted halfway through the test.

Taking it in November again. Hopefully will hit my target of answering 50 (or just 45!) questions then.

[mindreader]

A little background about myself...I am currently in my senior year, studying applied math in south east asia. I am interested in scientific computation as well as probability. For the past three years as an undergrad, my course work has focused on applied math areas like ODE, PDE, numerical analysis, optimization, game theory, and some statistics. I have also taken the first part of a real analysis course (chapters 1-5, 7 of Rudin) and I am currently taking complex analysis and an introduction to abstract algebra (just groups and their geometric applications).

I took the exam because it is recommended by many applied math programs I'm interested in. Many say that I should take "recommended" as "required." I only learned about it during the summer of this year and had I known earlier, I would've taken courses on abstract algebra and topology in my university. Nevertheless, indeed time was a huge factor during the exam. The time constraint pretty much affected my ability to analyze a problem like I normally would. I was surprised that a number of questions which could be solved in a straightforward manner were located near the end of the exam. So I answered them first and skipped the questions which were lengthy. Oh yeah the lengthy questions were very conceptual and time consuming! Another thing I was irritated about during the test is the presence of a physics question, too bad I really don't have any common sense in physics, primarily because of my high school background!

My suggestion: If you want to earn easy points, ensure that you are familiar with some theorems in high school geometry as well as fundamental stuff in graph theory! They're obviously a lot easier than topology.

Anyway I managed to answer 49 and left the others blank. I intended to answer around 55 but some were too challenging for me, specifically the ones on topology. I just hope that I only have a few mistakes among the ones I answered. Aiming for a score which is at least in the 70th %. As of now, I don't have intentions of retaking it because I'm running out of funds! GRE general, TOEFL, GRE subject, application fees, sending scores, and whatnot! Probably if I have to do a masters degree first then I'll take this exam again for PhD admissions but right now I'm just hoping the adcom will see that my math background isn't ideal for this exam (not making any excuses!!!).

[tikalora]

Here are my thoughts:

1. The earlier questions appears to be easier than the ones at the back.

1. That was a trend I had seen throughout the practice tests, so that wasn't a surprise to me per say.

2. Time is a BIG issue. Questions in the later half (number 30-66) had so much text and took some time to set up + solve. I am not sure how to do them within 2.5-3 minutes. The initial questions were easy though: I did the first 20-22 questions in just the first hour. Got really hammered and demoralized during the 2nd hour though, and it went all the way downhill from there.

2. SO TRUE. Sometime it takes 2.5-3 minutes just to understand the question or set up the problem. And then there are the theoretical ones that require much more than 2.5-3 minutes of thinking...

3. Difficulty level: feels like when I was doing the 2005 test for the first time, which is far more difficult than Princeton Review or the older practice tests.

3. Definitely on par with the 2005 test.

Overall, I did badly. Only managed answered 40 questions.

I also managed to only answer somewhere in the neighborhood of 40-45 questions, so we're definitely in the same boat. Seeing as we will not know our scores until well after the Nov. test date, I won't be retaking. Also, consider the fact that if time is our main issue we probably aren't going to get that much better. Also, $150 is too much money.

I'd like to meet whatever genius thought it was a GREAT idea to cram 66 hard math questions into 2 hrs and 50 mins of testing. Hopefully my other qualifications (like my Goldwater scholarship) will help to alleviate the pain of this subject GRE score.

[26186514]

Just got back.
Yes it was hard. Time is a big issue...
And I was late, can't belive myself...

[Stochastic1]

One month ago,I thought and hoped to get 95 percentile on gre exam!
I took sample test GR9367. In the first attack I solved 51 problem within about 100 min so I had 70 min for 15 remain test!I finished that test by raw score 59
So I didn't feel any time pressure on that exam.And I thought that I don't need a deep review on topology , complex functions , multivariabale calculus and algebra.I have planned to solve such problems in that 70 min!
And also I found that there are many questions which seems long and difficult but they have short and easy answers too!
And last night, I took samlple test GR0568. My first attack take about two hours, and I solved 43 problem!I finished this sample test by raw score 51/5.
So the night before the exam I understood I need a deeper review in multivariable calculus , abstract algebra , topology , and complex variable!!!
I think GR0568 was much more difficult than GR9367 and unfortunately today exam was more difficult than both of them!I think my raw score will be about 48.which is far from what I've planned for!
I take the exam less than 10 hours after the sample test GR0568 , I feel the arrangment of the questions and the subject of them were really simmilar in these two test!!!
I like try once more , but these days 150$ is very very much in my country:(

[nico]

Just took it. I'm glad I wasn't the only one who totally ran out of time....I feel everything you just wrote.....I answered 54 questions....and who knows how many of those are wrong. If it all goes well hopefully less than 10 which brings my score down to like...40 :/.....I guess november will be just as miserable as this month....On a side note, does anyone know if it really takes 6 weeks to get the results in?!! I'm dying to know how I did here....:/

[CoxZucker_Machine]

Took the test today too. Answered 53 with a couple guesses in there. It felt essentially like the 05 test or the Princeton review practice test. Don't get me wrong, most questions were completely different, but they were of similar difficulty I think. The first half of the test was super easy but it got brutally hard in the second half. For the last 30 minutes I was just frantically skipping around from question to question without being able to make any progress anywhere. Once you start panicking, it's hard to get back in control of the situation. I'm guessing that I'll be in the mid 70th percentile. I spent months studying so damn hard for this thing! I'm glad it's finally over.

[26186514]

Agree on the difficulty.
I think ETS is trying to keep the scores in the middle. That's why there are super easy ones and damn difficult ones.

Took the test today too. Answered 53 with a couple guesses in there. It felt essentially like the 05 test or the Princeton review practice test. Don't get me wrong, most questions were completely different, but they were of similar difficulty I think. The first half of the test was super easy but it got brutally hard in the second half. For the last 30 minutes I was just frantically skipping around from question to question without being able to make any progress anywhere. Once you start panicking, it's hard to get back in control of the situation. I'm guessing that I'll be in the mid 70th percentile. I spent months studying so damn hard for this thing! I'm glad it's finally over.

[nico]

Took the test today too. Answered 53 with a couple guesses in there. It felt essentially like the 05 test or the Princeton review practice test. Don't get me wrong, most questions were completely different, but they were of similar difficulty I think. The first half of the test was super easy but it got brutally hard in the second half. For the last 30 minutes I was just frantically skipping around from question to question without being able to make any progress anywhere. Once you start panicking, it's hard to get back in control of the situation. I'm guessing that I'll be in the mid 70th percentile. I spent months studying so damn hard for this thing! I'm glad it's finally over.

Ditto....my summer was devoted to this thing....I wish there was an extra box on the thing specifying the number of hours, days, months you spent on the thing...you know, to consider genuine effort...:/

[mindreader]

I just realized from dumplinghao123's post in another thread that the exam is different across different countries. Are you guys aware that there's a version A, B, and C of the exam of which version C is the toughest? Damn I'm hoping that what I took is version C so that my lower raw score would translate to a higher scaled score!

[enomomo]

Took the test today too. Answered 53 with a couple guesses in there. It felt essentially like the 05 test or the Princeton review practice test. Don't get me wrong, most questions were completely different, but they were of similar difficulty I think. The first half of the test was super easy but it got brutally hard in the second half. For the last 30 minutes I was just frantically skipping around from question to question without being able to make any progress anywhere. Once you start panicking, it's hard to get back in control of the situation. I'm guessing that I'll be in the mid 70th percentile. I spent months studying so damn hard for this thing! I'm glad it's finally over.

Yah that was exactly my experience. On a lot of the later questions, I spent 4-5 minutes on them only to eliminate 1 or 2 options, which was very frustrating.

[chocobear]

How did you know that it was different version? I kind of forget everything in the test after it ended...

I just realized from dumplinghao123's post in another thread that the exam is different across different countries. Are you guys aware that there's a version A, B, and C of the exam of which version C is the toughest? Damn I'm hoping that what I took is version C so that my lower raw score would translate to a higher scaled score!

[26186514]

I just realized from dumplinghao123's post in another thread that the exam is different across different countries. Are you guys aware that there's a version A, B, and C of the exam of which version C is the toughest? Damn I'm hoping that what I took is version C so that my lower raw score would translate to a higher scaled score!

Sure hope so too! That was a really hard test...

[mindreader]

How did you know that it was different version? I kind of forget everything in the test after it ended...

I just realized from dumplinghao123's post in another thread that the exam is different across different countries. Are you guys aware that there's a version A, B, and C of the exam of which version C is the toughest? Damn I'm hoping that what I took is version C so that my lower raw score would translate to a higher scaled score!

dumplinghao123 posted a couple of questions in another thread...didn't have any of those in my exam! well i thought my exam was much more difficult than the sample test ETS provided. that's just my opinion though!

[chocobear]

Yeah. His is different with mine as well. I heard from other math phd that the difficulty level is increasing as years go by, so it should be harder than the practise one. But I noticed that the score conversions is stricter, ie, raw score of 50 questions was 84% but last year it was about 81%. Ah, the school has to admit us coz we are paying so much money to toture ourselves on a test...

How did you know that it was different version? I kind of forget everything in the test after it ended...

I just realized from dumplinghao123's post in another thread that the exam is different across different countries. Are you guys aware that there's a version A, B, and C of the exam of which version C is the toughest? Damn I'm hoping that what I took is version C so that my lower raw score would translate to a higher scaled score!

dumplinghao123 posted a couple of questions in another thread...didn't have any of those in my exam! well i thought my exam was much more difficult than the sample test ETS provided. that's just my opinion though!

[aards]

I managed to answer 65 of the questions , gotta admit that was a brutal suffering for the ones after question number 40 despite of a few straightforward ones. I did guessing on around 10 or 12 of the questions after ruling out 2 or three of the answers.

[mindreader]

where are you based chocobear?

[Legendre]

I managed to answer 65 of the questions , gotta admit that was a brutal suffering for the ones after question number 40 despite of a few straightforward ones. I did guessing on around 10 or 12 of the questions after ruling out 2 or three of the answers.

Wow!! 65!!! Are you some kind of quick calculation genius?

More thoughts:

1. I am in south east asia, and my test was different from dumplinghao123's USA test. None of the questions he posted showed up in mine.

2. One thing I find unfair is that we don't really know the exact syllabus. There was a question that asked something specific to a topic I didn't know was in the syllabus. Are we even allowed to discuss the syllabus we saw in the test? Mathematics is so vast. It is so unfair to not know how and what to prepare for.

3. Based on everyone's experience, I think the test has degenerated into a speed test that involve a lot of luck: whether you happened to be familiar with the paper's syllabus, whether you guessed some choices correctly, whether the questions you chose to spend time on are easily solvable.

Also, it tests the ability to run at the first sign of difficulty, use shortcuts and fish out problems with quick easy solutions, which is the opposite of grad studies, where one has to have the tenacity and perseverance to hack away at difficulties.

[rmg512]

Overall, the test seemed about on par with last year's October exam and the released 2005 exam. I attempted 59 of the questions and felt good about almost all of them. Hopefully I won't get slaughtered by the guessing penalty! Oddly, I thought mid-test was the most brutal portion. To me, the algebra and topology questions were easier than they've been in the past once you thought about them, and the last few questions weren't terrible.

I was not expecting that physics-y question, and that combinatorics question was hard though!


For those of you complaining about the amount of time to take the test, I recommend taking tons and tons of practice tests. I believe there are 16 available. Eventually, you'll get down the proper pacing for optimum performance. It's a fine line between being efficiently fast and careless, though.

[Legendre]

For those of you complaining about the amount of time to take the test, I recommend taking tons and tons of practice tests. I believe there are 16 available.

There are only 4 official practice tests available. Unless you're counting the Princeton Review test and the REA book tests.

Btw, 59 questions is amazing! On the 2nd attempt for the 2005 official practice test, I was able to only do 45.

[aards]

I managed to answer 65 of the questions , gotta admit that was a brutal suffering for the ones after question number 40 despite of a few straightforward ones. I did guessing on around 10 or 12 of the questions after ruling out 2 or three of the answers.

Wow!! 65!!! Are you some kind of quick calculation genius?

More thoughts:

1. I am in south east asia, and my test was different from dumplinghao123's USA test. None of the questions he posted showed up in mine.

2. One thing I find unfair is that we don't really know the exact syllabus. There was a question that asked something specific to a topic I didn't know was in the syllabus. Are we even allowed to discuss the syllabus we saw in the test? Mathematics is so vast. It is so unfair to not know how and what to prepare for.

3. Based on everyone's experience, I think the test has degenerated into a speed test that involve a lot of luck: whether you happened to be familiar with the paper's syllabus, whether you guessed some choices correctly, whether the questions you chose to spend time on are easily solvable.

Also, it tests the ability to run at the first sign of difficulty, use shortcuts and fish out problems with quick easy solutions, which is the opposite of grad studies, where one has to have the tenacity and perseverance to hack away at difficulties.

Legendre, I wont say I am a "calculation genius" , but I did around 1000 test questions including 11 practice tests(real and simulated) and i think that was what helped a lot in the quickness in my reaction to the test questions today. However, as you said , that may not be a true indicator of my grasp of all the undergrad courses that were on the Subject test syllabus.( I had never taken courses such as probability, numerical analysis or abstract algebra back in my country, and I managed to familiarize myself with those courses during the last 6 months as I prepare for the test.

I agree with you when you say that subject GRE is inclined to emphasize too much on some "irrelevant" skills to what is expected in the grad schl , specially in phd program. However , it might also be fair to say that the performance on this test is just one component of the vector that is at least in 10-dimensional space.

[Legendre]

just one component of the vector that is at least in 10-dimensional space.

Another terrible thing about the subject test is that if you do well or exceptionally well, its an almost irrelevant part of your application. But if you do poorly, it will almost surely kill you, especially for top departments like Berkeley.

Work your ass off and it counts for nothing. But you got to work your ass off because if you don't, it counts for a lot.

[aards]

just one component of the vector that is at least in 10-dimensional space.

Another terrible thing about the subject test is that if you do well or exceptionally well, its an almost irrelevant part of your application. But if you do poorly, it will almost surely kill you, especially for top departments like Berkeley.

Work your ass off and it counts for nothing. But you got to work your ass off because if you don't, it counts for a lot.

are you based in us? you also took the test today, am i right?

[Legendre]

are you based in us? you also took the test today, am i right?

Yes I took the test recently in south east asia.

[chocobear]

Canada

where are you based chocobear?

[meredith]

Blearg. That was a fiasco. I definitely agree with Legendre about the whole "only can hurt you thing" and I definitely hurt myself with this (and I'll admit, I'm not the strongest applicant to begin with). How seriously do you guys figure they take this score? I'm sure some schools probably have a minimum cutoff, but in general, it's going to be way less important than your actual academic record, and letters of rec, I assume.

Regardless, I'm not sure how I'm going to make it through the couple of weeks of waiting to get the results back.

[26186514]

Blearg. That was a fiasco. I definitely agree with Legendre about the whole "only can hurt you thing" and I definitely hurt myself with this (and I'll admit, I'm not the strongest applicant to begin with). How seriously do you guys figure they take this score? I'm sure some schools probably have a minimum cutoff, but in general, it's going to be way less important than your actual academic record, and letters of rec, I assume.

Regardless, I'm not sure how I'm going to make it through the couple of weeks of waiting to get the results back.

Same here. Very anxious... I kinda wait for the result to decide whether I should apply this year or not.

[Legendre]

How seriously do you guys figure they take this score? I'm sure some schools probably have a minimum cutoff, but in general, it's going to be way less important than your actual academic record, and letters of rec, I assume.

Bad score - They will take it VERY seriously.

Good score - Way less important than the other things.

Yep. Its damned if we do (waste time on not very important aspect) and damned if we don't (bad scores taken very seriously).

[marco]

But what would be considered a bad score? Is percentile 60 a bad score? What about percentile 70? Also what do you guys think will be the raw score cutoff for percentile 60 and percentile 70?

[nico]

Since apparently lots of people are more than unhappy with the whole thing, is anyone planning on retaking the exam in november? Do you think it would dramatically increase your performance?

[Legendre]

But what would be considered a bad score? Is percentile 60 a bad score? What about percentile 70? Also what do you guys think will be the raw score cutoff for percentile 60 and percentile 70?

x < 50% = Atrocious. Consider not sending.

60% > x > 50% => Bad. I would feel bad sending this. But I heard >50% might be ok.

70% > x >= 60% => Ok. Might work for applied maths or non-math PhDs.

80% > x >= 70% => Good. I am very happy sending this to non-top 10 programs.

x > 80% => Excellent. UC Berkeley actually demands this so I guess this is the standard for top programs.

IMHO these are the broad 4 categories. However, subject test scores counts for little in the sense that they only really come into play when comparing two close candidates or schools with cutoffs. If one is perfect GPA Harvard with publications, and the other is 3.0 GPA from University of Degree Mill, then even if the Harvard guy bombs his subject test, the other guy don't stand a chance even with a 120% score.

[colldood]

In relation to the above post, it may be of interest to know that I applied with 50-60th percentile last year, and was very close to getting into two top 25 schools (waitlisted). This was for pure math PhD.

[tikalora]

In relation to the above post, it may be of interest to know that I applied with 50-60th percentile last year, and was very close to getting into two top 25 schools (waitlisted). This was for pure math PhD.

Personally, just looking at the applicant information forums, it seems that there is no strict correlation between your general and subject GRE scores and the rest of your application. There are many people on that forum who did "poorly" (< 50%) on the subject test and still got into some of the schools they wanted to. So, if the rest of your application is stellar, you should still have a chance. I think what is important for us is in our personal statement that we emphasize our strengths at mathematics to make up for our poor subject scores. Many of us have other qualifications (high math GPA, research, publications, scholarships) that should show we are good at math without that silly subject test.

Another thought: ETS claims the test is 50% calculus and differential equations, but on this test and the other practice tests I only remember there being 1 or 2 questions on diff eqs. Seeing as diff eqs are my specialty, I feel a little ripped off by this. Has anyone else noticed this trend?

[Legendre]

Personally, just looking at the applicant information forums, it seems that there is no strict correlation between your general and subject GRE scores and the rest of your application. There are many people on that forum who did "poorly" (< 50%) on the subject test and still got into some of the schools they wanted to. So, if the rest of your application is stellar, you should still have a chance. I think what is important for us is in our personal statement that we emphasize our strengths at mathematics to make up for our poor subject scores. Many of us have other qualifications (high math GPA, research, publications, scholarships) that should show we are good at math without that silly subject test.

Even though I did not take the math subject test last cycle, I was interviewed by an ivy league non-math (but quantitative + science) department.

However, my impression from the applicant information thread here is that those with higher scores tend to get into better places. No? This is corroborated by admissions statistics showing that tougher places tend to admit students with higher scores.

Another thought: ETS claims the test is 50% calculus and differential equations, but on this test and the other practice tests I only remember there being 1 or 2 questions on diff eqs. Seeing as diff eqs are my specialty, I feel a little ripped off by this. Has anyone else noticed this trend?

Yep, differential equations are 2-3 questions tops.

[tikalora]

Personally, just looking at the applicant information forums, it seems that there is no strict correlation between your general and subject GRE scores and the rest of your application. There are many people on that forum who did "poorly" (< 50%) on the subject test and still got into some of the schools they wanted to. So, if the rest of your application is stellar, you should still have a chance. I think what is important for us is in our personal statement that we emphasize our strengths at mathematics to make up for our poor subject scores. Many of us have other qualifications (high math GPA, research, publications, scholarships) that should show we are good at math without that silly subject test.

Even though I did not take the math subject test last cycle, I was interviewed by an ivy league non-math (but quantitative + science) department.

However, my impression from the applicant information thread here is that those with higher scores tend to get into better places. No? This is corroborated by admissions statistics showing that tougher places tend to admit students with higher scores.

I guess I should qualify my statement... this is a trend I have seen with applied math applicants. I agree for pure math programs it is much more important.

[26186514]

can't wait the results to come out...

[mandelawillson]

Just got back.
Yes it was hard. Time is a big issue...
And I was late, can't belive myself...

Really appreciate with you time is big problem.Time management is most important thing in exam.And every time i am not time management proper way.