overachieversmom

What is the equivalent of 700 on the old GRE in the new format. DD should be above 700 quanti for engineering programs she's applying to if not higher. I suppose some slack given the great gpa and magnitude of her research experience and recommendations.

DD took a computerized version of the GRE today and is convinced her scores suck.
I am finding too many online score conversion charts - none from ETS-- so not sure what to believe.


DD has a 166 verbal and 158 quant and seems really upset. That's the unofficial score at the test center-- no percentiles. Clearly 158 is borderline-- but does that translate into 700 ish?

I'm guessing this translates into 94% verbal about a 700 and a 77% quant about 720-- does that seem accurate? 77% tile is not so hot.

She asks me if she is toast. I dont' think so-- ??

awvvu

Here is the official concordance table: http://www.ets.org/s/gre/pdf/concord...nformation.pdf

aidami

Which schools are she applying to? For engineering, some of the schools look for a quantitative score of 700, others look for 750 (on the old scale). That's unofficial though- most schools say they don't have a hard cutoff point, and that other good things (like research) can compensate for a lower score.

TrueAce

158 seems to be a 740 according to the table.

If your daughter has succeeded in undergrad engineering courses, I don't think the ad com is going to have doubts about her ability to handle grad school, provided everything else in her application is in order.

I wouldn't give up.

eg1

She should be OK -- 740 is a decent score. A little below average for engineering (especially for top schools), but I think the rest of her record will count for more. I think she has definitely met the cut-off, and that's the most important thing.

Sichelgaita

The concordance table is interesting to me. FWIW, I'm a history applicant, and my scores from last July are:

Verbal: 770 -> 170
Quant: 770 -> 161

Even after taking into account quant being considered easier, I thought I was overperforming in quant. But the renormalized scores put me more in line with where I was on the SAT/ACT. Of course, I'm still happy with the scores.

I noticed the footnote advising people to take caution looking at 166s on quant from the old test, as they were actually perfect scores. Clearly, there was a need to differentiate more in the math section, which comes at the expense of eradicating verbal differentiation. Although 770 vs 800 really might have been noise and luck, it was still interesting to take a test that wasn't an easy verbal perfect score for me, and certainly heightened my nerves.

I'll be interested to see how things change in July of this year, as they state the percentiles are going to be updated then.

awvvu

What's interesting is this hard-to-find section of ETS's site: https://grediagnostic.ets.org/, which actually tells you how many questions you got wrong in each section and their difficulty.

My Q score was 166 which is apparently the best score, yet I still got 2 questions wrong out of 40 (both were hardest difficulty: 5)

Sichelgaita

That's interesting indeed.

On Verbal I missed 3 of 30, all level 5 and all in different categories. I took less than 30 seconds on 21 questions, only one of which I muffed. I was doing a good job identifying the harder questions, so possibly I should have taken even longer on them than I did. As I said in the score report thread, I left 15 minutes on the clock.

On Math I missed 5 of 28, one level 4 and the rest level 5, including both stats problems (to my surprise; I took AP Stats in high school and thought I had a grasp on that stuff.) There wasn't a strong correlation between time taken and performance in this case, nor even between time taken and difficulty (I spent only 0:06 on a level 4 quantitative comparison and got it right -- the last problem, as I was running out of time. I don't remember if it was a guess or not.)

I received no level 1 questions on either test. Is there no such thing, or is it only reserved for people performing poorly after the first few?

Acere

Thanks for giving that link, I didnt know about it.

In quant I missed only 1 question of difficulty 5 and got a 168. I assume a perfect score will obviously get you a 170.

How do you get a 169?

awvvu

Ah I see, the concordance table only has scores that matched with the old GRE score, so it is possible to get higher than 166. In the old system everyone would have been given an 800 though.

kiurys

Yes that was the reason for the new scale: to differentiate. Before people with 800 could of been a perfect score or they could of missed a couple and still gotten a 800. There was no way to set them apart.

The new scale accomplishes this.

TrueAce

If the number shown at the end of the test is accurate in 2012 then I got a 168 on the Quantitative (though maybe not, I suppose they need to compile the results of the day).

I'm guessing I probably only got one question wrong. Honestly I doubled checked everything from every angle and every question was easy. I'm shocked I even got one wrong.

That being said, 169 only probably comes into play depending on variance of test scores. If getting 1 wrong already drops you to the 96th or 97th percentile, then a 169 would be an impossible score for that period.

If for some reason a test a test is more difficult and getting only 1 wrong puts you in the 98th percentile, then 1 wrong would be a 169. So I can see how numbers at the top can be skipped.