Strong GPA, Research, References--Bad GRE Score

[ope211]

Here's my situation. I have a high GPA at a very well known university, three research internships (including an REU and SULI,) a senior thesis, a potential publication, and strong recommendation letters. My physics GRE score for the September test? 510. BEYOND terrible. I'm taking it again but to be honest I'm not expecting to do much better, I studied pretty religiously for the September test and obviously it didn't help much so I'm not sure what more I can do.

I'm very concerned about the cost of applying to graduate school. Right now I'm looking at University of Chicago, Michigan State, University of Hawaii, Wisconsin Madison, George Washington University, and UC San Diego, all for nuclear/nuclear-particle astrophysics. If my GRE scores are going to prohibit me from getting into any of these schools, I want to know so I can consider other options because application fees are a huge expense. Does anyone have any advice? I'm also applying for the NSF graduate fellowship which I'm told I'm a good candidate for, and that doesn't look at GREs. What if I get the fellowship and don't get into grad school anywhere?

 

[Dishsoap]

I'm genuinely curious, since I have looked at quite a bit of research to determine how physics GRE scores affect one's chances in grad school admissions (I made this thread this past winter while applying to grad school myself). What was the problem with the pGRE, and why do you not think you are capable of doing better? If you have a high GPA at a "well-known university", it seems to me that you should have seen a lot of the material before. Did you not take the time to memorize equations? Did you not look at old tests (which, by the way, I would consider to be the greatest piece of advice one can give to someone who is taking the exam).

 

[ope211]

I'm genuinely curious, since I have looked at quite a bit of research to determine how physics GRE scores affect one's chances in grad school admissions (I made this thread this past winter while applying to grad school myself). What was the problem with the pGRE, and why do you not think you are capable of doing better? If you have a high GPA at a "well-known university", it seems to me that you should have seen a lot of the material before. Did you not take the time to memorize equations? Did you not look at old tests (which, by the way, I would consider to be the greatest piece of advice one can give to someone who is taking the exam).

You're not asking me anything I haven't asked myself already. I did look at old tests, I tried to take them in a test environment and all that stuff. I read "Conquering the Physics GRE" cover to cover, did every practice problem in the book. I'm trying to understand what happened, obviously I'd seen most of the material before. I obviously think I can do better but it really isn't like I didn't study, so I'm not sure what I can accomplish in the coming weeks that will push my score that much further up. I mean, maybe I'm not actually good at physics and it was all a ruse?

Thanks for your thread! It was very informative (if terrifying and discouraging...haha.)

 

[Crush1986]

I'd definitely study very hard for the next two weeks. How did you study the last time? I'd take a practice test and find some weaknesses, use a book like Halliday and Resnick for practice problems on them. That is pretty much how I did it and it helped a lot. My first practice test was a 620, my score 790 on the real test. I did a decent amount of studying the few weeks before the September test. I told myself I'd study for months before and learn upper-division quantum mechanics... Didn't go so well with an intense REU over the summer. I'm pretty sure a 510 will get you some auto-rejects. A score in the 600's may help you out with staying in the running at some universities.

*Just saw you replied above* I don't know what you can do then... I guess you have test anxiety or something.

 

[ope211]

I'd definitely study very hard for the next two weeks. How did you study the last time? I'd take a practice test and find some weaknesses, use a book like Halliday and Resnick for practice problems on them. That is pretty much how I did it and it helped a lot. My first practice test was a 620, my score 790 on the real test. I did a decent amount of studying the few weeks before the September test. I told myself I'd study for months before and learn upper-division quantum mechanics... Didn't go so well with an intense REU over the summer. I'm pretty sure a 510 will get you some auto-rejects. A score in the 600's may help you out with staying in the running at some universities.

*Just saw you replied above* I don't know what you can do then... I guess you have test anxiety or something.

Thanks for your reply. I've never been spectacular at tests and I guess I have test anxiety but I can usually get average or better. I'm going to mostly take practice tests and do Halliday-Resnick problems until the next test. Don't know what else to do. Do you think I'd get auto-rejects from all these schools? Don't want to waste hundreds of dollars if that's the case.

 

[Dishsoap]

When you took practice tests in a test environment, how did you do? I found that I actually did better on the real tests than on the practice ones. (upper 600s on practice tests, 710 and 750 on real ones).

 

[ope211]

When you took practice tests in a test environment, how did you do? I found that I actually did better on the real tests than on the practice ones. (upper 600s on practice tests, 710 and 750 on real ones).

I was doing in the 600-700 range.

 

[Crush1986]

Thanks for your reply. I've never been spectacular at tests and I guess I have test anxiety but I can usually get average or better. I'm going to mostly take practice tests and do Halliday-Resnick problems until the next test. Don't know what else to do. Do you think I'd get auto-rejects from all these schools? Don't want to waste hundreds of dollars if that's the case.

I'm not sure how many Universities auto-reject scores in the 500's. I've heard that some universities will do it though. You could probably just focus on a few subjects and do much better. Like mechanics and atoms (Rydberg equations always shows up), or EM it's a huge chunk of the test if you know a few of the equations like Coulumb's law and Ampere that is definitely a few questions. You don't need much to push it to the 600's.

 

[Crush1986]

When you took practice tests in a test environment, how did you do? I found that I actually did better on the real tests than on the practice ones. (upper 600s on practice tests, 710 and 750 on real ones).

I did too. The practice tests compared to the test I took were wayyyy harder. They made me mentally tired much more quickly.

 

[Vanadium 50]

I have some unpleasant things to say, I am afraid.

First, you should be having this conversation with your academic advisor. He or she knows much more than we do, and can provide more targeted advice.

Second, your application is probably not as strong as you think it is.

• You're right that a 510 is not a good score. Neither is a 600. Neither, really, is a 700. A 700 is at the 51st percentile. Roughly twice as many people take the Physics GRE as end up enrolling. Anywhere.
• A 3.7 GPA is at a university where the most common grade given is an A. You are now seeing the evil dark side of grade inflation. As they said in The Incredibles: "When everyone is super...no one will be!"
• Participating in research is not the magic that some people here think it is. What matters is what you got out of it, and that shows up in the letters. You say the letters are strong. How do you know? "Best student all year" is an average letter.

Given that, sure, apply to some top tier schools. But you should probably also apply to some other ones - in particular, look for smaller schools with strong programs in what you are interested.

 

[ope211]

I have some unpleasant things to say, I am afraid.

First, you should be having this conversation with your academic advisor. He or she knows much more than we do, and can provide more targeted advice.

Second, your application is probably not as strong as you think it is.

• You're right that a 510 is not a good score. Neither is a 600. Neither, really, is a 700. A 700 is at the 51st percentile. Roughly twice as many people take the Physics GRE as end up enrolling. Anywhere.
• A 3.7 GPA is at a university where the most common grade given is an A. You are now seeing the evil dark side of grade inflation. As they said in The Incredibles: "When everyone is super...no one will be!"
• Participating in research is not the magic that some people here think it is. What matters is what you got out of it, and that shows up in the letters. You say the letters are strong. How do you know? "Best student all year" is an average letter.

Given that, sure, apply to some top tier schools. But you should probably also apply to some other ones - in particular, look for smaller schools with strong programs in what you are interested.

Thanks for your reply. I don't really think my application is "strong" because I don't really know what a strong application is. Yeah, I'm aware that a "good" score is pretty much above an 800. I didn't know the stat about how many people take the GRE vs how many enroll, that's good to know.

I assume you looked at my post history for my GPA/school. Ahhh, the magic of grade inflation. Yeah. I guess you're right. I am pretty sure that an A isn't the most common grade given in a physics class at my university, and they do put the stats of how many people got the grade you got in the class on your transcript.

When you say it matters what you got out of research--what do you mean? I am working on a senior thesis, presented my work at the APS meeting, working on a publication. I'm not trying to toot my horn, just genuinely curious and a little confused about what you mean I should be getting out of a research experience. I say that about my letters because my professors and research advisors writing me letters have written them for me before for REUs and the like, to general success, and I'm pretty sure rec letters are important factors in those decisions. I've also only worked on projects where I was the sole student for the summer/year, so I've gotten to know my research mentors and they've gotten to know me and my work quite well.

I'm trying to look for small schools dedicated to what I'm interested in. I'm very interested in nuclear physics, hence, I found George Washington University.

Thanks for the advice. It's appreciated.

 

[Vanadium 50]

Saying "I did an REU over the summer" will barely move the needle when it comes to a decision. Having the REU advisor write a letter that says "She did X, Y and Z on this project and also learned how to do A, B, C. She was the best REU student in the last 10 years" will be a strong positive. In between is, well, in between.

 

[ope211]

Saying "I did an REU over the summer" will barely move the needle when it comes to a decision. Having the REU advisor write a letter that says "She did X, Y and Z on this project and also learned how to do A, B, C. She was the best REU student in the last 10 years" will be a strong positive. In between is, well, in between.

Makes sense. I am having all my research advisors write rec letters, so I hope they will be able to convey that.

 

[Student100]

I've never been spectacular at tests and I guess I have test anxiety but I can usually get average or better.

How did you end up with "a high GPA" in that case?

]I'm going to mostly take practice tests and do Halliday-Resnick problems until the next test. Don't know what else to do.

Where were you weak on the first attempt? Are you guessing on too many problems? You need to weigh the penalty of an incorrect guess versus how confident you are in said guess. Are you spending too much time on each problem? How were you doing at making order of magnitude guesses? You've taken the exam once, you should have a clearer picture of where you weaknesses are. Not "Don't know what else to do."

If you still can't score better than a 510, you might need to question if this field is the best for you.

]Do you think I'd get auto-rejects from all these schools? Don't want to waste hundreds of dollars if that's the case.

You'll get plenty of rejections. I don't think they're going to explicitly auto-reject you, but your score is certainty a negative. The rest of your application package would have to outweigh those who didn't bomb the PGRE and who aren't such large question-marks when it comes to passing qualifying exams.

You need to retake the exam, and score much better.

 

[radium]

Of the schools you mentioned, I have heard directly from people at UChicago (this was after I was admitted so I think it is pretty reliable) that they are pretty lenient about the PGRE. They actually compared professor ratings of applicants to PGRE scores and found little to no correlation (most likely with people above a certain score). I think is true at several other schools, even at the caliber of UChicago.

However, while these schools may not care about the difference between an 800 and 950 per se, they will most definitely be concerned by a 510. It is likely that they will automatically reject you because of that score . In order to avoid problems with the PGRE you should try to score a 700-750 at the bare minimum. If your score is at this level I think you can make up for it with different parts of your application.

I think it's definitely a good idea to take it again. I only took it once (which in the end made no difference since I got my top choices), but I think if I had taken the test again being more comfortable with the format I would have done much better.

 

[Vanadium 50]

I am looking at the https://www.ets.org/s/gre/pdf/practice_book_physics.pdf now, and the difference in raw scores between a 510 and a 750 is quite large. It's about 33 or 34 questions - a factor of 2.8x as many as you need for a 510.