update: 8 IR (same as GMATPrep) and 6.0 AWA--that was quick!
Sorry for the length. Brevity might be the soul of wit, but I crushed the GMAT, so I’m going to pontificate for a while.
Last week, after about 3 months of aggressive prep (which followed several months of half-assed prep), I took the GMAT. I performed at the top of my quant range, and probably a little below the average of my verbal range. Overall Assessment
It’s been stated countless times, but this is a very learnable test. I was a political science major in college, and while I’m by no means quant-averse, I’m far from a natural. I was solid on verbal from the get-go, largely because I studied for the LSAT a few years ago (on a side note, I hope those of you who start at Q50 on your first practice test and only have to work verbal find my practice test scores as frustrating as I found yours
- Manhattan’s GMAT overview book followed by stone-cold Manhattan GMAT
CAT for diagnostic purposes
- MGMAT Quant books with accompanying OG problems by topic
- Sackman’s Total GMAT Math
with accompanying practice problems (only the harder ones since I was later in my prep)
- (same time as Sackman’s book) MGMAT SC
and Powerscore CR
. Sporadically did the last 50 of each type in the OG
- GMATClub tests
every day that I could for 2-3 wks, reviewed MGMAT Advanced Quant
- GMATClub Book just for a new take on the basics
- (important, IMO) the last 2 weeks were ONLY official prep material. GMATPrep tests
every weekend and sets of hard/medium GMATPrep question pack problems every weekday. Takeaways
1. Don’t get too discouraged or encouraged by MGMAT scores. My suspicion is that the GMAT algorithm is far too complex for (with all due respect to MGMAT) a group of test teachers to crack. The writers of the GMAT algorithm probably have countless years of collective experience in the field of psychometrics under their belt.
2. You might not need a prep course. While I spent around $500 on prep materials (in hindsight, I could have easily spent half that), it’s still a fraction of what many courses charge, and frankly I feel a little more proud of my score knowing it was self-study (not to disparage anyone else’s score, this is just my personal take)
3. Prep for IR. No telling how this score will come into play in the future (there are already murmurs of consulting firms looking at it), but if you’re getting a good score in quant, maxing IR should be easy with minimal time investment. Also, don’t forget you can TYPE on the IR calculator. This will save lots of time.
4. Save your GMATPrep tests
until the end. One of the challenges of prep for me was constant doubt that I was progressing because of an absence of reliable progress indicators (which the LSAT has in spades). This is the reason I purchased 4 GMAT Foci. I think I did the right thing, though, because I could be confident my GMATPrep scores were accurate when I finally got them.
5. Get your mind & body right for this. Stay healthy so you can think clearly, and start to normalize your daily routines. Ex: I ate the same things for breakfast for several weeks before the test, drank the same brand of coffee, and took practice tests at roughly the same time I’d take the real thing. Our bodies go thru rhythms throughout the day, so you may as well take it at a ‘high point’ in that rhythm (although adrenaline will help regardless of time of day). Also, I read about a study that correlated creatine monohydrate consumption with enhanced cognitive performance, so I figured why not give that a shot. Hard to say if it had an effect, but I figured ‘why not?’
6. The internet ain’t gospel when it comes to the GMAT. A lot of things spouted as truth are completely unsubstantiated. I challenge anyone to show me a good primary source for the two well-known “facts” of the 80/80 split and the first-ten-questions rule of high scoring. In fact, the latter is directly refuted by papers published on the topic, at least according to an article by MGMAT’s Stacy Koprince.
7. On timing: if you’re at a high-enough level on quant, you know your limitations. I always did my catchup around question 25, making sure I had 2 min/question for the last 12. Some folks will tell you to absolutely guess and move on at 2:30, but I never timed each question that accurately. Also, since I was ready to punt questions I had no idea on, I could confidently take my time on hard ones in the first 20 questions or so and know I’d have at least one punt-worthy question in Q 20-30. It’s a risky strategy, but so is every other strategy. Materials Used
1. MGAT Prep books (quant)
-- a lot of praise is doled on these books, and for good reason. They’ll show you how to solve any problem you may encounter in the prep. In my opinion, though, the quant books don’t provide sufficient depth in their coverage of concepts. I personally found their treatment of combinatorics to be oversimplified. On the other hand, their tabular treatment of overlapping sets was really useful. All-in-all, worth going through in their entirety, but don’t be afraid to supplement with material that is less “gimmicky” in its methodologies (e.g., the GMATClub book).
2. MGMAT Archer
– this was OK. It was really helpful being able to see explanations for almost every question right in one spot, but the statistics seemed almost useless. For example: they plot your performance on a time over accuracy chart, with “green zones” at 1:30 and 60%, respectively. I asked if these #s meant anything at all, and they stated they were just indicators that you were doing well—which to me is another way of saying they’re arbitrary. I think this application has potential, but as it stands, it only risks giving you a false sense of security (or worse, insecurity), because the metrics don’t correlate at all to your test-day performance.
3. MGMAT Tests
– I think I’m one of the outliers in their test taker data set. I have a habit of getting bogged down on problems, but I’m quick enough on actual GMAC-written problems that I can often make up for this. With so many hard problems on the MGMAT tests, my weakness was being blown open and bogging down my score. This probably helped improve my performance, but it took a good deal of confidence in my own abilities to ignore the MGMAT scores and press-on with my practice. In all, these were most useful for enhancing timing & endurance.
4. Jeff Sackman’s Total GMAT Math
—I did this after the MGMAT books
and the OG Problems b/c I wanted to make my fundamentals bulletproof. I liked Sackman’s explanations of math concepts because he made no effort to dumb it down. I didn’t like, however, the amount of errors in his practice problems and the fact that this publication now seems to be no longer supported. I emailed Sackman about an error and never heard back, so caveat emptor. I also purchased his problem sets, but ended up not using them in favor of the GMC tests.
5. GMAT Club Test
—problems with score accuracy as you can see from my score log, but these questions were as close an analogue to the actual hard GMAC-written questions as I was able to find. While MGMAT seems to introduce difficulty thru computational rigor and complicated problems, the GMATClub tests
focus on a kind of conceptual elegance that really challenges your understanding of a topic. Their numprop questions are second-to-none, which is vital considering how many high-level GMAT questions come from that category. Take as many of these as you can.
6. GMATClub itself
– invaluable for explanations to questions from other sources. I didn’t randomly browse questions, though, for fear of seeing a question from GMATPrep. As I mentioned, I was diligently saving GMATPrep till last, and I didn’t want to piss that away. I found that, eventually, I was just scrolling down to Bunuel’s explanations. Others were often helpful, but his always seemed to be the explanation that put the discussion to rest. Succinct & smart, almost without exception. I have no idea if this person is paid by GMATClub or not, but I can say that without a doubt, the majority of whatever revenue GMATClub made from me was because of Bunuel.
– These were OK. The timing was weird b/c of question counts and there was no geometry. GMATFocus gives you an 80% confidence interval for your quant score, and my CIs were pretty consistent (albeit very wide). This was helpful for putting my mind at ease—it said to me in fairly certain terms that I was capable of performing where I wanted to, at the 48/49 level.
8. GMATClub book
-- I wish I would have read this before Sackman’s. It was similar in that it was no-nonsense, but it was free, so if deciding between those two sources, go with GMATClub’s. The formula for determining the sum of a geometric progression with |r| < 1 actually helped me on an exceptionally difficult real GMAT problem, and something as plain-and-simple as that appeared in no other source (or if it did, I didn’t notice it). Formulas are tools, and they’re far superior to gimmicky tricks—GMATClub’s book realizes this and employs it fully in their material.
9. Powerscore CR/MGMAT SC
-- I grouped these together because they’re highly recommended here and elsewhere, but honestly I don’t have much to say. I seldom used their tactics and typically just went with my ear and/or the gut notions I honed on the LSAT.
10. MGMAT Advanced Quant
-- a nice summary of the brute-force tricks documented all over these forums. By the time I cracked this book, I’d already read most of the tricks elsewhere, but it was a nice consolidated list.
11. GMATPrep Question pack
-- This was awesome. I found the “hard” problems to be just that, and the software ain’t bad either. The verbal was also very difficult, which more accurately reflected my GMAT experience. I had very little trouble with the GMATPrep test verbal questions and timing on the verbal section, but on test day, the questions more closely resembled question pack 1. This meant I was faced with timing problems that I’d not encountered in verbal prep. I think this cost me a V45 and therefore a 760, but I’m still super grateful for my score.
12. GMATPrep Tests
-- Okay, some asshole wrote somewhere that the two tests in this software draw from completely different banks. With that in mind, I took the first test, then reset to re-take it before taking the second test. When I finally got to the second test, there were several repeats from the first take of test 1. This really upset me because I only got one honest-to-goodness assessment out of this software. When reviewing tests, I noted how many I got right that I’d originally gotten wrong, and made sure to take the requisite 1.5-2min to offset any influence repeats might have on my timing.
13. Thursdays With Ron Reruns
-- these were kind of helpful, but I didn’t like the pace. I think a written article with the same explanations & problems could have been just as useful w/o requiring 1.5 hrs of focused attention. Watched maybe a half-dozen of these.Test Day
Got there just in the nick of time, slugged half-a-cup of coffee and set-in for the warmup sections. Didn’t do great on timing AWA, didn’t get to proofread last paragraph. IR went well at first, but I got way bogged down on one question and lost track of time. I eventually made up for it, but that could have turned out very, very badly. I really believe timing on IR is one of the biggest challenges.
Quant was easy & quick for the first 10-15, and I took the time to double-check. Good thing I did, b/c I caught a few dumb errors. Started hitting questions I just had no clue on, and punted them when that fact dawned on me. Got to the last question with 2-3 min left. I was worried b/c the last 2 questions were so easy I triple-checked to make sure I wasn’t misunderstanding the question, but as it happens, my worries were unfounded. Lots of numprop questions, a couple geometry. Only one combinatorics problem that I can recall, and it was very simple. No probability if I’m remembering correctly.
Test day was the hardest verbal section I’d ever done. A few RCs were baffling and many CRs and SCs had me staring at them for a while. I never got in the habit of punting a verbal question, and that made my time extremely tight this go-around. Finished on-time but was a little strapped towards the end.Things I Should’ve Done Differently
1. Done all OG questions at once instead of breaking out by subject.
2. Not bought any of Sackman’s stuff. Again it wasn’t bad, there are just cheaper/better alternatives.
3. Not reset question bank after taking GMATPrep test 1.
4. Set a test date earlier to focus my prep.
5. Scoped the test center. I’m military currently stationed in Central America, and I was almost late to my appointment b/c of unexpected traffic that I would have expected had I scoped it out. Further, if I’d have realized my test computer had a Spanish language keyboard, I could have been less reliant on typing in the IR calculator.
6. Maybe dialed down the caffeine a little for the real thing. I was fried when I hit verbal, which had never been a problem before, but the verbal was exceptionally challenging, so that may have bit me.
7. At least TRIED an AWA section. I went into this one cold and found I was more strapped for time than I ought to have been.
8. I should have spent some time practicing creating test scenarios for DS number property questions. Often I’d find myself staring at a problem trying to crack it in my head, and by the time I realized I had no idea, I could have easily just tried all relevant test cases and arrived at an answer. Basically, don’t be too proud to try brute-forcing a problem right away.
File comment: Just a quick visual i slapped together comparing my progress in the diff't diagnostics. I only included the mode of my GMATFocus CI with an arbitrary date
Score chart.PNG [ 10.49 KiB | Viewed 929 times ]
File comment: A log of every practice test/test section I took
Capture.JPG [ 101.85 KiB | Viewed 975 times ]