To retake or not to retake...

[scyrus]

Brief summary of my situation: I'm a CS student seeking to apply to PhD programs this coming Fall, and have a 4.0 and a couple of years of research experience from a top 10 CS institution as an undergrad(top 10 is in Graduate CS rankings, though). I don't have any publications. I scored a 770 in math, and a 680 verbal on the GRE.

Currently, I'm in a summer research program at a very prestigious research institution, and this research will likely, but not definitely result in a peer-reviewed publication. This research program is also offering me a subsidized Kaplan GRE test prep course. I pay $50, they pay the rest of the overwhelming majority. Therein lies my dilemma...

When I got my GRE results, I was content, but not happy. I know I can get an 800 on the Math, I even did so on the practice tests. However, the GRE being a minor metric used in admissions, I figured a 770 is good enough. If I buy the $50 course, I will probably want to take it again and waste hours of time on the prep. If I don't buy the prep course and don't get in to the universities I want, I will probably feel that I wasted a valuable opportunity. What do you guys think? Will a 770 in math keep me out of the top CS schools? Should I take the course and, by extension, take the test again?

[anthropologygeek]

Only you can answer that. Is taking it again for 30 max points worth the money and the risk of getting a lower score? Also every place you apply will see the 770 since the see every attempt for a five year period.

[DrFaustus666]

Hi Scyrus,

It's true that probably half of the people you're competing against (top CS grads going for PhD's, that is) did score 800. But that means that the other half didn't.

And your 4.0 average, coupled with your 680 in verbal, which I think has got to be in the top 1% of Computer Science majors, are compelling.

I wouldn't worry about it too much; i.e., I don't think I'd retake the GRE for $150 for 30 extra points ---- but if you share in your SOP some new ideas on neural networks or a really new and different sort algorithm you invented yourself---and that runs 50% faster than any other known algorithm --- anything like that, and the AdComm will blow off your 770 as an aberration---maybe you accidentally hit the key for "square root of 2" when it should have been "square root of 2 / 1" ... and of course, then couldn't go back.

The moral of this lengthy homily....I think you're in a very good position. If by some chance (which I doubt), you are rejected by all of your top choice schools, you can always try the GRE again.

My two cents,

John

[scyrus]

Thanks for the responses. After some soul searching, this is what I've come up with: If any school is going to reject me because of a 770 instead an 800, I probably don't want to go there anyway. I'd rather spend the time on focusing on my research and actually graduating now =).

By the way, Dr.Faustus666, how did your applications go? Or were you going to apply this fall, and not last fall?

[DrFaustus666]

Thanks for the responses. After some soul searching, this is what I've come up with: If any school is going to reject me because of a 770 instead an 800, I probably don't want to go there anyway. I'd rather spend the time on focusing on my research and actually graduating now =).

By the way, Dr.Faustus666, how did your applications go? Or were you going to apply this fall, and not last fall?

Unfortunately, life circumstances have caught up with me, and I'll need to wait until Fall 2011 at the earliest. I am divorced, my ex-wife's income is low, and so I must continue working full time to pay off my children's undergraduate college tuition for at least another year. This issue has dogged me since 2005 (when I began the as-yet uncompleted M.A. number 2) .... my PhD dream may fall by the wayside ... I HOPE NOT, but it's well possible.

The bright side is I'll be able to continue working on Quantitative, and maybe bring that score into the mid-700s: where I've been told it needs to be for my ultimate research ambitions to seem plausible.

Thanks for asking, scyrus.