GRE Thoughts

Ah, the GREs. The GREs manage walk the frustrating line of being barely important enough to an application to warrant studying for. Many top schools don’t "require" the GREs, but as some websites gently suggest (like Iowa in this link -- scroll down to point 4 under "Other Notes on Admission and Residency Requirements"), scores can often affect funding. And whenever funding dollars are involved, you can be sure that it is important enough to put time and study into. Some insights and strategies I learned when preparing for the GREs:

I took the test on September 13th, and got a 650 on the Verbal, a 680 on the Math, and a 6.0 on the essay section. Verbal, of course, is the big kahuna of the GREs, and really the only score that the schools will care about. My basic strategy for the Verbal portion consisted of two basic rules:

1) Memorize the "most frequent word" lists provided by any of the Barron's/Kaplan/Princeton study books. This was important to my success. Already have a gigantic vocabulary? Great. You'll do wonderfully on the test, and will have nothing to worry about. But for the rest of us, it’s impractical and unrealistic to even attempt to memorize the 3,500 word lists provided by the preparation books. The "most frequent word" list represents a much more realistic and efficient way of targeting words that you'd be likely to see on a test. My prep book's list was a little over 300 words, which I managed to cram in about two weeks before the test. During the test I recognized at least 7-10 words from that list, which was a significant portion of the test.

2) Take as many practice tests as you can. Myself, I took at least 20 practice tests from a variety of sources. I can’t stress how important it was to me to take those practice tests. As much as the ETS would like you to believe, the GREs are not a measure of practical knowledge, but really a measure of well you can take the GREs. The problems on the Verbal section of the GREs, especially the analogy questions, require a large amount of familiarity to have any kind of success. So in addition to the relearning those baseline strategies you’d use for the SATs or ACTs back in high school (eliminating obvious answers, etc.), I was able to improve my pattern recognition of certain kinds of questions through a ton of repetition. Cramming works.

Another tricky aspect of the GREs is the fact that they are Computer Adaptive Tests (CATs) conducted on computers, as opposed to paper tests. What is a CAT? In essence, it’s a test that “adapts” its question difficulty based on how many questions you answer correctly or incorrectly. The more consecutive questions you get correct on a test, the more difficult a test becomes, and vice versa, all to eventually determine your final score. But there's more than that. Since the test uses the thirty questions to zero in on your score, it tends to weight the first 15 questions more heavily than the last 15. For example, before you answer even one question, your initial score, like everyone else's, will be set on the average of the bell curve of all those who've previously taken the test (somewhere around 400-500). If you answer the first question right, the computer will nudge you up into the tier of questions intended for 500-600 scoring folks, and if you were to get it wrong, the computer will shift you down into the tier for the 300-400 folks. As the test progresses, those large point swings will get smaller and smaller as the computer "figures out" the appropriate level of difficulty of questions for you. By the end of the test, a correct question will only be work 10-20 points each. Keep in mind that while my math above is hypothetical (I don’t really know if the point swings are 100 at the beginning or not), the process is the same.

As you can see, the first 10 or so questions are incredibly important to moving into the highest possible tiers, as you can only move your score so much once you get into the later stages of the test. One of the most common criticisms of the GREs is that the test unfairly punishes those who make mistakes early on without much hope for improvement later. But if you go into the GREs knowing this, you can similarly use this to your advantage. With 30 minutes to answer 30 questions, I knew I wasn't going to get every single question right, but I took the extra time on the first 10 questions to carefully answer them and to make sure that I got as many right as I possibly could. In fact, I probably employed this strategy a little too effectively -- at one point I had about 14 minutes to answer the remaining 20 questions -- and had to really rush in order to answer every single question. I had no doubt I probably answered more questions incorrectly on the second half of the test, yet I still got (in my own estimation) a pretty good score. So while it may feel counterintuitive as a test-taking strategy to front weight your time on the test, it really is something worth thinking about.

In my next post I'll go a little bit deeper into the resources and tests I used for preparation for the GREs.