# About the GRE Mathematics Subject Exam and its scoring

I want to apply to a Ph.D. in Mathematics in the United States. Therefore I'm about to take the GRE Mathematics Subject exam. I have some confusion about the grading procedure and the interpretation of the grades.

In the official practice book from the GRE website they write:

GRE Mathematics Test total scores are reported on a 200 to 990 score scale in ten-point increments.

However, the conversion in page 63 shows a maximal grade of 910. I also found this unofficial table from 2019, which states that 63-66 in the practice test (GR1768) is indeed equivalent to 910 in the real test, but in the real test it is possible to score up to 970.

So what is the maximal score when I take the test? And does it mean that the practice test is easier than the real test? I mean, it is the only source offered by the GRE website for preparation, so that shouldn't be right.

• As you may expect, GRE and the likes are just obxnious crap to categorize people in a quick and dirty way. Every decent university (or business school, the most common users of such "indicators") will set a minimum GRE requirement, so it is not relevant how much you score as long as you score >600 (or other threshold) Commented Sep 12, 2022 at 9:38
• Maybe I should have mentioned that I want to apply to Ph.D. in mathematics Commented Sep 12, 2022 at 11:49
• Luckily some university are phasing out the GRE test. mathematics.uchicago.edu/graduate/mathematics-phd-program/… Good luck anyway! Commented Sep 12, 2022 at 11:55
• and another more comprehensive list mathalliance.org/wp-content/uploads/2020/09/… Commented Sep 12, 2022 at 11:56

Note that the GRE is becoming required in fewer places, especially since the onset of COVID, but also a realization of its limitations.

Don't worry too much about the score. It will be what it will be, but I'll offer a warning if you take it. At the end of the exam you will feel terrible, as if you are sure you "failed" it. But that is because it is extremely broad (or was when I took it more than a half century ago). There were some questions for which I'd never seen the terminology, though I was a top math student.

But then, the scores came in and I found I'd done quite (very?) well. Much better than my expectations.

Furthermore note that in US evaluation of doctoral admissions is very broad with more emphasis put on letters of recommendation than some other places. Committee members (and it is usually a committee) are looking for predictions of success in doctoral study and thereafter. A Statement of Purpose can be very important if it is asked for. So, the GRE, even if required is usually given less weight, though there are some places that might use it as a gate, so an especially low number might be an issue in some top institutions.

Note however, that even a low GRE can be overcome by good letters from trusted faculty. I once had a case in which one of my "top" undergraduates in CS had low GRE scores that might have prevented further consideration. But my support of him at the time was seen by the dean (who I knew) and he was admitted and succeeded. The dean later wrote me to tell me all the above. That person succeeded at this (top) place and he is now a full professor at another top place.

In my personal case, my acceptance to doctoral study was much more influenced by recommendations and support of trusted professors than by anything else. I actually have some evidence of this.

So, for the GRE, do what you can do and don't worry about it. Get good letters of recommendation and make a broad search for a position.

Yes, I realize that the above advice doesn't answer the question about "how it is graded". Assume fairly, but also "fairly strictly". I've never been a grader. Answer a lot of questions assuming that you are pretty sure you have an answer before you start. Ignore questions for which you have no background on the first pass.