640GMAT vs 1300GRE

Question

Which do you think is better to report?

A GMAT of 640 (38V/40Q) or a GRE of 1300 (600V/700Q)

lagomez · Answer

600Q and a 700Q = roughly 640

Report either one I guess

coaks · Answer

I say send whichever gives you the highest and most-well-rounded percentiles. If they are weighted more heavily in different areas, send both.

Posted from my mobile device

lagomez · Answer

Your GRE percentiles are going to be converted to GMAT percentiles so it's not going to matter much, unless your percentiles aren't balanced, even if you get a 50% on the GMAT and a 90% on the GRE. Percentiles are not equal. This is not my opinion, this is from various admissions at schools that take both. Use this tool by the GRE to confirm https://www.ets.org/Media/Tests/GRE/pdf/ ... n_Tool.xls

coaks · Answer

Interesting link. Thanks. But the confidence intervals on that comparison tool seem absurdly large. The 90% CI around any given score is over 200 GMAT points. Also it makes no attempt to translate quant to quant or verbal to verbal and the demographics of their sample differ significantly from the demographics of GMAT test takers. I hope adcoms aren't using that translation tool but, sadly, I assume they are.

coaks · Answer

One last note: After inputting a bunch of scores from different GRE percentile bands (even q/v break-outs only), it appears that the translation tool and dataset would argue that GRE/GMAT percentiles are equivalent.

lagomez · Answer

Unfortunately, I think they are using this tool. However, I am sure there is some room for error.

007MBA · Answer

Thanks for the info! Interesting approach on their part- I would assume that there is a certain demographic to individuals that take both, hence the redac confidence interval. And only 500 people !?!

jzd · Answer

I entered my GRE score from 10 years ago V550 Q800 - Total 1350 and it outputs 680 GMAT

My real gmat is 690 (Q51 V31). It is frighteningly accurate and even sadder that even after 10 years I am at the same "skill level" for standardized tests

coaks · Answer

It's surprising but 500 people should offer a fairly robust estimate of the intended population. The question becomes, however, what are they actually trying to measure? Are they trying to estimate the translated performance of all GRE/GMAT test takers or are they trying to estimate the translated performance of only the subset of GRE/GMAT test takers that would be predisposed to taking both? I'd say they might be doing a decent job of the latter. The variability in translated performance just seems very high for this to be a reliable regression for adcom's purposes.

asimov · Answer

I would take ETS's comparison tool with a grain of salt. ETS is trying to promote GRE, so I would guess it would try to make the scores more comparable.

I entered a GRE score of someone I know who scored 650 on GMAT.
GRE (720V/740Q) - predicted score of 720 with  67.4 Standard Error .
67.4 points either way is HUGE on a 800 point test.

My conclusion is that GRE isn't an accurate predictor of GMAT score.

coaks · Answer

Or at least their sample doesn't tell much about predicted GMAT performance, especially not at the total score level. I'd be curious to see whether Verbal/Quant performances on GRE and GMAT were more reliable.

640GMAT vs 1300GRE

Prep Toolkit

Top 5 GMAT Debrief Videos

615 to 715 in 15 Days (Ria's Strategies)

More than Quant/Verbal Abilities: Speed vs. Accuracy

745 with Only Self-Prep and GMAT Club

From 555 to 765: 210-point Improvement

Perfect 805 Score Debrief by Julia

My Rewards