Preparing for 3rd attempt in 6 weeks need verbal help

[Scott.Wachs]

I just finished my second attempt and scored a 580.

44 Math
27 Verbal

This time I'm really at a loss for words. I really thought I was doing ok in the verbal section. My last 5 practice manhattan gmat verbal scores were 37, 33, 36, 33, and 36. The last time I scored in the 20s was on my first attempt which also came as a surprise. My previous score was 42 math, 22 verbal. A few thoughts from today. First off I felt very confident going in. My verbal scores alone made me believe that at least i'd score in the 30s. This was really all I was aiming for. I have never had an issue with timing. I finished the verbal section with at most 1 minute to spare. I never felt rushed during any of the the questions. Where do I go from here. I have roughly 6 weeks. I need a game plan to tackle the verbal section. So far my practice tests really don't measure how accurate my verbal scores on the actual test have been.

Thanks

[StaceyKoprince]

We need to figure out why your practice tests are consistently in the 30s but your real test dropped to the 20s. If we know why, then we can figure out what to do about it.

Read this article and do the analysis described (for verbal), then come back here and post:

http://www.beatthegmat.com/a/2009/10/26/my-score-dropped-figuring-out-what-went-wrong

I'd like you to dig more into the timing, by the way - finishing on time is good, but you could still have a timing problem. I talk to students all the time who think their timing is fine because they finish the section on time, but they're mismanaging their time within the section, spending too much time on certain questions and then having to work too quickly on others in order to make up for it. The ones on which they choose to work too quickly are the ones that they think are easy... and then they start to make careless mistakes due to speed on lower-level questions, which can just kill your score.

Go look through your most recent MGMAT test and check out your per-question timing. (Not the averages - the actual timing on every single question in the section.)

How many times did you go more than 30 seconds over the time you were supposed to spend?

SC - about 60-75 sec; max of 90 sec
CR - about 2m; max of 2.5m
RC - about 2.5m (short) to 3.5m (long) to read; about 1 min for general purpose questions; about 1.5 to 2 for everything else

How many times were you more than 30 seconds UNDER the time you were supposed to spend, and how many of those did you get wrong? (The only good reason to get something wrong quickly is when you knew you didn't know how to do it. Otherwise, you thought you knew what you were doing, and you made a careless mistake.)

[Scott.Wachs]

Hi Stacey,
I wanted to give you a lot of data so you can make a more accurate recommendation. I took the verbal sections of the last 3 exams based on what you wanted. A few things to keep in mind that were in my head and in general my overall strategy. I feel stronger in critical reasoning and therefore am able to take less time to compete these questions. Sentence correction question I tend to take a little more time on. Especially when the underlined passages are extremely long. I'm just not able to complete those long passages as quickly as others. With regards to my reading comprehension, my first 2 tests I rushed through the passages but my last practice test as well as the exam I slowed down a lot. Here is the data I compiled.

Test 1
+30 (Questions that I went 30 seconds over the normal time)
SC: 7
2/7: level 600-700 50% correct
5-7: level 700-800 60% correct

RC: 1
1/1: level 500-600 0% correct

-30 (Questions that I went 30 seconds under the normal time)
CR: 6
1/6: level 300-500 100% correct
3/6: level 600-700 66% correct
2/6: level 700-800 0% correct

RC: 5
1/5: level 500-600 0% correct
1/5: level 600-700 0% correct
3/5: level 700-800 33% correct

Test 2
+30
SC: 7
3/7: level 600-700 66% correct
4/7: level 700-800 25% correct

CR: 1
1/1: level 600-700 0% correct

RC: 1
1/1: level 700-800 0% correct

-30
SC: 1
1/1: level 600-700 100% correct

CR: 3
1/1: level 500-600 100% correct
1/1: level 600-700 100% correct
1/1: level 700-800 0% correct

RC: 4
1/1: level 500-600 100% correct
1/1: level 600-700 100% correct
2/2: level 700-800 0% correct

Test 3
+30
SC: 8
4/8: level 600-700 75% correct
4/8: level 700-800 50% correct

CR: 2
1/2: level 500-600 100% correct
1/2: level 600-700 100% correct

RC: 1
1/1: level 600-700 0% correct

-30
CR: 2
2/2: level 600-700 50% correct

RC: 4
2/4: level 600-700 100% correct
2/4: level 700-800 50% correct

I know this is a lot of data. But i wanted to give you more than enough information just in case my last exam was a fluke. If you need me to change the way I present this info I can move things around for you.

Thanks in advance!
Scott

[StaceyKoprince]

So, let's see.

When you go over on:
CR: 2/3 right
SC: 12/22 right
RC: 0/3 right

And that potentially costs you:
CR: 5/10 wrong
SC: 0/1 wrong
RC: 7/13 wrong

Check my math to make sure I compiled that correctly.

Okay, what are the lessons? It looks like it's not worth your time to go over on RC (the data point is a bit small, but...). It might be worth your time to go over on CR (but not more than 1m over - how long did it take you on those 2/3 that you got right?). SC is a bit of a mixed bag - the percentage is decent, but we also have to look at what that might have cost you.

And going too fast on RC appears to be costing you big time - you got more than half of those wrong.

For the "too fast" category, classify the problems as one of two things:
1) I knew I didn't know how to do it, so I just made a fast guess, or
2) I really thought I knew how to do it and was surprised when I got it wrong.

The ones in the 2nd category should be 75%+ correct. (Again, this is for the "too fast" group - if you really thought you knew what you were doing, so much that you thought you could work that quickly, then you better not get it wrong!)

How many of those CRs and RCs from above fit into category 2? This was the true cost of spending too much time on the others.

Now, on the real test, your verbal score dropped. On the real test, our bad habits tend to get magnified a bit due to stress. So, if you're making some poor decisions about how to allocate 30sec here and there among problems, you're likely to do it even more on the real test. You're going to hang on even longer on some, and then have to work even faster on others, and then the whole effect is going to be worse on your score.

Also, dive into the SC data a bit more - that's obviously where you're spending the bulk of the extra time. Compile a list of problems where you tend to spend extra time. Tally it by content area tested - but not the overall category listed on the test. Look at each question and figure out why YOU spent so much time on it (b/c there are always multiple things tested on one problem). Also tally it by anything else you notice - eg, the whole sentence is underlined.

Are there any patterns in terms of when the extra time spent is valuable (you get it right) vs. when it isn't? Maybe you tend to go over on both modifiers and parallelism, but only parallelism tends to pay off in terms of getting the question right. If you know that, you can make a better decision next time: if it's parallelism, I'll give myself permission to stick with it a little longer, but if it's modifiers, I'm dumping it now.

Also, see if you can figure out the point at which you have narrowed to 2 answers, and then how much more time you take to pick just one. A lot of people lose a lot of time at this stage. The rule: once you've narrowed down to 2, you look at each choice ONCE more. Then you pick one and move on. (It's already 50/50. You're either going to know how to make that choice the first time you compare, or you're going to have to make something of a guess - and that very rarely changes after an additional 30 seconds of agonizing back and forth.)

You mentioned, in particular, needing to take a lot longer when all or most of an SC is underlined. That's because it's taking you longer to figure out what to test / focus on. When all or most of the text is underlined, it's often the case that there are big blocks of words moving around or changing, as opposed to just one word changing back and forth. What you need to do is develop the ability to see the "blocks" - these 5 words go together, the words between these commas go together - so that you can start comparing blocks more easily rather than having to rely on just single words changing. That's going to take some work, but you can do it!