GRE math subject test strongly recommended. Should I submit low score?

[ozborne]

Hi all,

I would very grateful if I could get some advice and hear some thoughts on my current situation. I am currently applying for PhD programs. I'm mostly applying to applied and computational mathematics departments, but also a few mathematics departments where they are also doing applied research.

Some of the programs I'm applying to (especially those in mathematics departments) either require (e.g. UCLA) or strongly recommend (e.g. Duke) submitting a GRE math subject test score. I did take the test in September, but unfortunately did really poorly: I got a scaled score of 650, which corresponds to the 49% percentile.

My question is this: For programs that strongly recommend, but don't require, the math subject test, do you think I'm better off just not submitting the test score at all?

One of the reasons why I'm leaning towards not submitting at all is that, first of all, my profile and research interests clearly are in applied rather than pure math, which ought to make the subject test less relevant. Second of all, by next fall I will have completed two master's degrees with strong grades: One 1-year MS in a heavily quantitative field from the math department of a top European university, and one 2-year MS in applied math from a strong US applied math department. I'm hoping that 3 years of graduate course work, mostly in applied mathematics, with strong grades will be a stronger indicator of my preparedness than the subject test. I'm not sure how the various admissions committees would look at this though.

Thanks in advance for any input.

[Neuro PolarBear]

I wouldn't submit it. If you did well in your masters program at a top school and have strong recommendations, I think you should be all set. How'd you do on the general GRE?

[svent]

650 isn't awful, but you should submit it (don't expect to get in though). If they say strongly recommended and you don't submit it, they'll probably assume you got like 500 on it or something (or less). It's sort of like graduating with a 3.2 GPA and applying to entry-level jobs without including your GPA on your resume. Recruiters will assume it's < 3.0 and trash it.

[ozborne]

1 hour ago, Micecroscopy said:

I wouldn't submit it. If you did well in your masters program at a top school and have strong recommendations, I think you should be all set. How'd you do on the general GRE?

The general GRE I did pretty well on: 163 (92%) on verbal reasoning, 166 (92%) on quant reasoning, 4.5 (80%) on analytical writing, which I'm quite happy with as a nonnative speaker.

13 minutes ago, svent said:

650 isn't awful, but you should submit it (don't expect to get in though). If they say strongly recommended and you don't submit it, they'll probably assume you got like 500 on it or something (or less). It's sort of like graduating with a 3.2 GPA and applying to entry-level jobs without including your GPA on your resume. Recruiters will assume it's < 3.0 and trash it.

That's a good point. I should also mention that my undergrad was in engineering rather than math, so hopefully they'll see that as another reason why I might not do the subject test. Or maybe that leads them to be even more interested in how I did my subject test...

[svent]

If your background is in engineering, and not math, then a 650 is pretty solid. Still probably doesn't get you into UCLA or Duke though, especially as an international student.

[ozborne]

1 minute ago, svent said:

If your background is in engineering, and not math, then a 650 is pretty solid. Still probably doesn't get you into UCLA or Duke though, especially as an international student.

Would you say it's better to just not include the score then? For UCLA I don't really have a choice though since it's required.

[svent]

No, you should still submit it. Read my first post. But don't expect to get admitted. Apply to some less selective schools.

[cwr]

2 hours ago, ozborne said:

Would you say it's better to just not include the score then? For UCLA I don't really have a choice though since it's required.

Just to be clear, it's my understanding that @svent is saying this:

1. Your chances at top schools are not great.
2. Your chances are better if you submit your subject test score, because otherwise they'll assume you did worse than you actually did. 

I know nothing about the programs in question, nor do I know anything about the math GRE. With that being said, I know how challenging the physics GRE is, even three years as a physics major. Given that your background is in engineering, it's impressive that you were able to pull off a score of 49th percentile on the math GRE, especially since you were likely competing against a lot of people who studied math as undergrads.

[Neuro PolarBear]

I'm still against it. Lots of programs that "strongly recommend" it really don't care that much and they won't know whether you took it and did poorly or didn't take it at all. Your demonstrated ability to do well in graduate coursework far outweighs the subject test, and you don't want to undermine that. Furthermore, I'm sure the types of math / applied mathematics that one does during a PhD doesn't even come close to what was covered on the subject test, so graduate programs know that it's not a good indicator of anything.

[svent]

Except it is a good indicator of your ability to understand calculus. Note that your ability to teach calculus is to some extent what they're hiring you to do in a Ph.D. program. Also if you need to pass quals in algebra or analysis or topology, it's a good idea to recruit students with a decent understanding of those subjects.

[TakeruK]

Here's a similar story, maybe you can draw some useful conclusions from it.

I took the Physics GRE twice, scored 640 and 690 (44th and 53rd percentiles) in 2009 and 2011. The school I am currently at (top 10 for planetary science) "strongly recommends" a PGRE score submission. I read "strongly recommends" as "you better submit one unless you have a good reason not to" so I submitted both of my scores. I got in! 

Afterwards, I found out that "strongly recommends" doesn't mean what I thought it meant at all. In fact, only one other student in my cohort wrote the PGRE (the others did not even write the PGRE). One person explained to me that since Planetary Science is a very applied and interdisciplinary field (others in our cohort come from geology, chemistry, biology undergrads as well as physics/astro), it doesn't make sense to write and submit the PGRE unless your background was in Physics. But then I later found out that the other Physics majors did not write the PGRE either. 

I think there are many parallels to my story and yours ("strongly recommended", similar scores, and fields that are "applied" versions of the subject GRE test in question). My own takeaway from my story was that if a subject GRE score is only "strongly recommended", it seems like they will give the score a low weight, since my test scores are well below the typical PGRE scores for this school.

[ozborne]

Thanks everyone for your feedback. I ended up not including my GRE subject test.

On 11/9/2015, 5:25:23, svent said:

Except it is a good indicator of your ability to understand calculus. Note that your ability to teach calculus is to some extent what they're hiring you to do in a Ph.D. program. Also if you need to pass quals in algebra or analysis or topology, it's a good idea to recruit students with a decent understanding of those subjects.

I have been working as a TA for the duration of my current master's studies, so hopefully that will compensate for me not submitting my GRE subject test from, at least from the perspective of my ability to contribute as a TA.

On 11/9/2015, 8:47:24, TakeruK said:

Here's a similar story, maybe you can draw some useful conclusions from it.

I took the Physics GRE twice, scored 640 and 690 (44th and 53rd percentiles) in 2009 and 2011. The school I am currently at (top 10 for planetary science) "strongly recommends" a PGRE score submission. I read "strongly recommends" as "you better submit one unless you have a good reason not to" so I submitted both of my scores. I got in! 

Afterwards, I found out that "strongly recommends" doesn't mean what I thought it meant at all. In fact, only one other student in my cohort wrote the PGRE (the others did not even write the PGRE). One person explained to me that since Planetary Science is a very applied and interdisciplinary field (others in our cohort come from geology, chemistry, biology undergrads as well as physics/astro), it doesn't make sense to write and submit the PGRE unless your background was in Physics. But then I later found out that the other Physics majors did not write the PGRE either. 

I think there are many parallels to my story and yours ("strongly recommended", similar scores, and fields that are "applied" versions of the subject GRE test in question). My own takeaway from my story was that if a subject GRE score is only "strongly recommended", it seems like they will give the score a low weight, since my test scores are well below the typical PGRE scores for this school.

Thank you for this very useful feedback. I'm hoping that's the approach they take in the math departments strongly recommending the math GRE too.

[doctor-to-be]

I would not submit it. I personally believe that schools should either request something or not. Recommending this and that seems useless to me. 

[TakeruK]

26 minutes ago, doctor-to-be said:

I would not submit it. I personally believe that schools should either request something or not. Recommending this and that seems useless to me. 

I used to think the same way but recently, I have been revising these thoughts. I now personally believe that (and advocate for, whenever possible) schools should have a minimal amount of required items and then have a large list of possible "optional" items. The reasoning is that excellence can be defined and demonstrated in a lot of different ways and I think schools should strive to recruit a large diversity of excellence. When you have a strict list of things you are considering then you are strictly defining excellence to be a narrow thing. Students that are excellent but don't fit into the narrow mold would not do well in this system.

The Physics subject GRE in particular is a polarizing test in my field. Many people think it's not very effective because doing poorly on it doesn't really mean you're a bad physicist. But doing well on it could mean you really know your stuff. And others do argue that the Physics GRE is something that is more of an "equalizer" for those with non-traditional backgrounds---for example, students in countries without a Physics undergrad degree program can use a high Physics GRE score in order to demonstrate excellence in that way.

However, there are also other thoughts that show the Physics GRE places others at a disadvantage. For example, if a student is working part-time to pay tuition or supporting a family, the extra time required to study for and do well in the Physics GRE can be a barrier. Also, if you are living in rural parts of the United States or outside of the United States, then it's even harder and more prohibitive to register for and write the test. Rejecting students on the basis of poor Physics GRE scores alone is generally a bad practice.

So, to me, I think the clear answer is to make the Physics GRE scores optional! If you are able to do well in the Physics GRE, then you should be able to submit it and demonstrate your excellence there. However, if you are not able to do well in it (or if you don't want to do it), then you don't have to submit it and you should not be penalized. Instead, you can demonstrate your excellence through other means. To me, I think this fits the purpose of graduate admissions a lot better---the goal is to find signs of excellence in the applications, not signs of failure, so we should structure the applications to give students the best chance to demonstrate why they are great for our programs.

(Note: All of this is assuming your post means you want schools to just say things are "required" or "not required" and remove categories like "recommended" or "optional". I agree with you that "recommended" is useless but "optional" is a good category. "Recommended" places a value judgement that shouldn't be there if you really intend it to be optional. So if this is also what you meant then I guess I'm not really saying anything new :))

[svent]

"For example, if a student is working part-time to pay tuition or supporting a family, the extra time required to study for and do well in the Physics GRE can be a barrier."

How is this any different from the general GRE, or doing well in (unrelated) classes, or finding time for research?

Edited November 15, 2015 by svent

[TakeruK]

3 hours ago, svent said:

"For example, if a student is working part-time to pay tuition or supporting a family, the extra time required to study for and do well in the Physics GRE can be a barrier."

How is this any different from the general GRE, or doing well in (unrelated) classes, or finding time for research?

compared to General GRE: yes, it's not any different and I would have also said that the general GRE should be optional, but I didn't want to get too off topic (since the OP is discussing subject GRE).

compared to doing well in unrelated classes or finding time for research: I am not sure why it has to be all or nothing. What I mean is that if a student only has X hours outside of classwork to do things that will help their graduate school chances, then having fewer required things will help students with limited resources. For example, if all programs made the Physics GRE optional, then if the student invests all X hours into research, they can demonstrate excellence there. Or, if they invest all X hours into doing really well in coursework (maybe their school doesn't have research opportunities), then they can demonstrate excellence there. But if you require the Physics GRE as well as the other things, then excellent researchers without a ton of spare time will have to split their efforts and look worse on paper.

This argument takes into account these assumptions though:
1. Doing well on the Physics GRE is not necessary to do well in graduate school.
2. A school usually asks for students to provide information on many things: X, Y, Z, etc. such as LORs, GPA, GREs, research experience etc. I'm working under the assumption (as has been my experience in graduate school) that grad schools don't necessarily only want students that have excellence in all categories. Instead, most graduate schools seem to want to use the application process as a way to identify excellence. Each item, X, Y, Z etc. is another chance for an applicant to show excellence.
3. Schools are trying to ensure they get a variety and diversity of excellence---that is, not just the "traditional" grad student route and that schools are careful to ensure their application system does not "weed out" students they actually want to attend but don't meet the "mold".

So, if that is the case, I think schools will be better off being less restrictive on what materials are submitted. Instead of treating applications as a "test" or a "trial" where they look for reasons to reject, they can allow candidates to submit their best materials and look for reasons to accept. The down side is that with more variety in materials submitted, it could mean more work for the admissions committees. But I think it's worth it.

[holeinyougraph]

I know it's a bit late but, 650 is not that low of a score. From talking with some of my professors, who got their degrees from Group I schools (PSU and Michigan), that was about the average subject test score (for american students with undergraduate degrees). I know it's too late to change your decision, but for others asking themselves a similar question with scores in the 620-680 range, I think the correct call would be to submit your scores. @svent Seems to be the one with the best sense here. They either assumed that you did not take the test which could be seen as a lack of effort or that you scored abysmally, both cases hurt your application. 

[athulyaram]

Hi, this seems really late, but I'm in a very similar situation right now. Did you end up submitting your scores? And did you get in?