April 23, 2009

My Misconception About the GRE

So, I took the GRE and, as Maggie put it, "welcome to your first hoop." It's a relief having the test behind me and I'm happy enough with my scores that it's also a confidence boost. You could say it has dissipated my diffidence.

Among all the practice testing, I noticed something and I thought I'd pass it along.

I used a prep book (Kaplan), and I also used the software that ETS (Educational Testing Service) sends you when you register for your test. These include many practice problems plus full-length timed practice tests. The practice problems are great for getting used to the types of questions you'll be asked. The full-length timed tests are good for getting a feel for how you'll manage your time, but they also provide a dynamic that is missing in paper tests: they simulate the adaptive nature of the GRE.

When I was in college, my test-taking skills taught me not to spend too much extra time on harder questions, but rather to answer them, make a note of them, and then return to them once I'd completed the test. This allowed me to answer all of the questions I was most confident about and use remaining time to strengthen my answers on the more challenging questions. On the GRE you can't do that. You aren't allowed to revisit a question.

In my early diagnostic runs through the timed test, I found myself running out of time. This upset me a great deal, because unanswered questions are scored as incorrect responses. I thought I had a problem with my time management, and I complained to Maggie and to coworkers that this was a real problem in my preparation.

So, I adjusted my approach and "rushed" through certain challenging questions, making educated guesses so I could answer more questions. My new approach had me finishing the test with time to spare! However, the adaptive practice test was now telling me that my score was really lousy. I was getting a few questions wrong, but not that many. However, my score was tanking. What happened?

A couple of smart people told me they hadn't finished the quantitative portion of the GRE. I didn't know their scores, but I figured they did pretty well, since they got accepted to good schools. I stopped worrying about finishing the test and instead concentrated on (surprise!) answering the questions correctly. As time began to run out, only then did I start rushing and making educated guesses to try to complete the test.

My score improved by a lot. Why? Because every time you answer correctly, the test adjusts to you by giving you a harder question. Answer incorrectly and you get an easier question. But the key is that the harder questions are worth more toward your score. It turns out they are weighted quite heavily. It is in your best interest to get into harder questions as early as possible. Many people give this recommendation, but you can see it in action if you vary your approach on the practice tests. The difference for me was dramatic. How dramatic? On quantitative I could finish the practice test, get 1/4 of the questions wrong and end up in the 33rd percentile: somewhere in the low 500s. Or, I could start strong, and get nearly 1/2 the questions wrong/unanswered and end up in the mid-high 600s, 150 points higher and in the 60th percentile.

Even though I'd been given this advice while I was preparing -- spend extra time on the early questions to gain access to harder questions of greater weight -- I wasn't sure how to reconcile it with my aversion to not completing the test. If you are preparing for the GRE, you may be much quicker at math than I am and this may not apply to you. The practice tests will tell you. But definitely spend time trying to get those early questions correct. If you are worried about not completing the test, try it both ways on the practice software. See how long it takes you to be fairly certain you're getting those early questions correct.

The downside is that answering correctly gets you harder and harder questions, so you might start to feel like you're not doing so well on the exam as the challenge increases. But take the hard questions as confirmation that you're doing well. That should bolster your confidence.

If you're interested, you can even download the ETS Power Prep software for free here. You can try practice questions and three full-length adaptive practice tests which will give you estimated scores. Heck, do it for fun! I found the whole test mostly enjoyable (even the harder math problems) except for reading comprehension where I hated having to scroll the boring passages back and forth to scan for details so that I could answer questions that relied heavily on shades of meaning and degrees of correctness. Bleh. I prefer reading on paper where I can mark up the text with notes.

A final note on actually taking the test: I took my test at the Prometric site on Post Road in Warwick, RI. I dislike constrained situations, but the friendly professionalism of the staff put me at ease so that I was able to focus on the test.

Posted by James at April 23, 2009 1:36 AM
Create Social Bookmark Links

That seems the "smart" way to administer and score a standardized test.

I haven't taken a standardized test since I was applying for flight school when I was in ROTC, and it wasn't a very academic test at that. The Navy based my postrad acceptance on my shipboard performance evaluations and my overall undergrad record. So I've never had to face the dreaded GRE.

Posted by: Bull at April 23, 2009 10:09 AM

I faced it before its adaptive days and I breezed through it. I may take a stab at the adaptive samples for curiosity's sake. Adaptive does seem to be a much better approach. I'd love to see it refined to the point where it does not count any questions you did not answer, but rather assesses what "level" you got to and answered consistently. It seems to me that you shouldn't need the whole time to accurately ascertain someone's ability.

Posted by: briwei at April 23, 2009 10:45 AM

I think I actually took the GREs at the same time Brian did (Clark U ~1988-89?) and it was the old style test.

I've always been a fast test taker. I was always one of the first people out of the room. I either know something or I don't and don't waste time struggling with questions. I do remember being surprised that I didn't finish the GRE (at least I think it was the GRE) because there were so many questions you couldn't finish it. I do remember being surprised at how well I did considering how many blanks I left (especially on the chemistry test, I think I answered like 1/3 of the questions and still scored highly enough to get me into 3 forensic chemistry programs.

"I may take a stab at the adaptive samples for curiosity's sake" Brian you are a sick man ;)

Posted by: B.O.B. (bob) at April 23, 2009 1:13 PM

They've also added a writing component, which is human-scored. Two critical essays, basically.

Posted by: James at April 23, 2009 1:36 PM

I don't agree with the "not finishing" part, Bri. I think somebody who can do these psycho math problems quickly definitely has an advantage over someone who can't because they have all of the relevant facts so ingrained that the math is just natural for them. That person is a better mathematician, IMO, than somebody who has to think for a bit (as I do).

I took the paper version as well. I didn't score nearly as well on the practice test going through one problem at a time (trying to simulate test conditions) as I did on my original GRE, but I haven't tried the adaptive computer test. If I ever decide to apply for another graduate program, I guess I'll have to.

I don't like the adaptive test because I really like being able to go back, but if you truly can spend a little more time on problems (like James, I was trying to pace myself), then maybe it's okay and maybe my score would be closer to my original score. I would really not like getting a quant score below the 700's!!

Posted by: Maggie at April 23, 2009 3:41 PM

That was the first time I scored lower on the verbal than I did on the quantitative, which surprised me. But the reading comprehensions were the ones that annoyed me (and I completely finished the verbal).

My guess is that the adaptive test is timed because 30 or 45 minutes gives their method a good read on your skill level. The test is guessing what your score should be and then seeing if you can answer a question at that level. Then it readjusts. The reason it can't just keep giving you harder and harder questions is that you can make careless errors, so it wouldn't necessarily be able to tell what level you belong on.

I don't enjoy taking the adaptive test, but I think it's probably an improvement over the paper test, especially in evaluating more skilled participants. Paper tests are likely to lack challenges that differentiate more skilled students.

Posted by: James at April 23, 2009 4:43 PM

I can't say about the GRE, but I know that I was annoyed by the reading comprehension stuff in the SATs (and also a sample MCAS I took out of curiosity a few years ago). I try to read the questions first and then read the sample with those questions in mind, but that still does nothing to improve the incredible dullness of the sample and the fact that you're trying to find out something that you don't care about at all.

Posted by: Julie at April 24, 2009 11:21 AM

The prep book says that they intentionally try to find dull writing and then compress it to make it as dense as possible. It shows.

Being able to read the questions first would be great. I wanted to read the sample questions first, too. However, on the adaptive GRE, you only see one question at a time.

You are encouraged to take notes on the passage, but my notes were rarely helpful. By the time I took the test, I was writing up pretty decent notes. None of the questions I got had anything to do with what was in my notes. Often the questions were about statements the author would likely agree with, the tone of the piece, and things that had to do with specific wording, which did not make it to my notes.

Posted by: James at April 24, 2009 11:29 AM

If I couldn't go back to look at the passage, I'd be screwed for sure. I only take notes on things that seem important to me. I don't have a good memory for specific wording - I translate everything into Julie language as I read - and I don't care whether the author would agree with something or not. :)

Posted by: Julie at April 24, 2009 12:09 PM

So would I; luckily, they leave the passage there for you on the GRE.

Posted by: James at April 24, 2009 1:23 PM

Copyright © 1999-2007 James P. Burke. All Rights Reserved