Suggested tips to improve from mgmat 670, for next 1 week

[abhishekk810]

Hi,

I'm giving my GMAT exam on the 9th of September. I got a 640 (q47 v31) a month ago on the exam (before that, I averaged 700 on the official mocks, but obviously test day pressure and increased difficulty were something I hadn't prepared enough for)

My verbal has improved, I have scored 670 (q45 v36) and q44 v37) in 2 mgmat mocks recently. Apart from adding some discipline to my sc strategy (as I find I have 1-1.5 minutes to spare at the end of verbal, and >50% of my verbal errors are on SC - I assume it's because I'm rushing through SC, looking at the clock), I'd say my verbal is on course to hit a v39-40 on the actual exam.

Regarding quant, I have scored a q48 and q47 in gmatclub quant cats. All the scores in mgmat and gmatclub quant cats are from this past week.

I'm targeting a 730+ score. Please let me how these scores could translate on the actual gmat, and any additional tips to improve my prep. Let me know if I should share any additional information.

Thanks,
Abhishek

[StaceyKoprince]

A lot of people find our CAT quant to be harder than the real thing. Since you scored Q47 on the real thing, there's a decent chance you can maintain that level.

If you can get V39-40, your overall score would likely be about 700-710. (Which is a great score! I know you want more—but any 7xx score is a great score. )

I'm mentioning that because most people don't see big score differences in a week—usually, what you're scoring about 7-10 days before the test is about what you can expect to score on the real thing. An extra 20-30 points isn't huge—except when you're already over 700. It's difficult to go from 700 to 730 in general, let alone over just one week.

But let's go for it this week and see what happens—if you get a 700 and want to take it again to go for the 730, you'll be in a great position (because you'll already have locked in a 7xx score).

You mention test day pressure and increased difficulty getting to you. For pressure, I'd recommend doing some mindfulness training (which is usually best to start at least 4 weeks before, but let's do what we can in the time you have).
https://www.manhattanprep.com/gmat/blog/2013/05/09/stressed-out-meditate-to-lower-your-anxiety-and-boost-your-gmat-score/

Re: increased difficulty—do you mean that you messed up the timing? (That's what usually happens: People get sucked into really hard problems, spend too long, then have to rush elsewhere and start missing problems they could have done if they'd had enough time.)

If timing was a factor, take a look at this (although, again, this is usually a 4-6 week process to address):
http://www.manhattanprep.com/gmat/blog/2016/08/19/everything-you-need-to-know-about-gmat-time-management-part-1-of-3/

Given what you've told me, I'd go in with a goal of hitting 700 and assume that I'm going to take it again to try to get up to 730. (If you actually do hit closer to 730, great—we'll take it! Just don't put that pressure on yourself for this time. Lock in a solid score that you know you're capable of getting and we'll work from there.)

Use this 2-part series to help you do that:
https://www.manhattanprep.com/gmat/blog/2014/11/07/last-14-days-gmat-part-1-building-game-plan/

And let us know if you have any more questions. Good luck!

[abhishekk810]

Hi Stacey,

Appreciate your response, apologies for the delayed reply. I took the test a couple of days ago, and ended up scoring a 680 (Q47 V36). I'm waiting for my official score report to be published so i can test my hypotheses through the ESR (E.g. assumption that i was at V40 for SC and RC, but a low score CR of V28 due to exam pressure contributed to a massive drop in my verbal score). For starters:
1. The test felt ridiculously tough - Quant kept throwing hard questions at me - right until the very end - 2 out of the last 3 questions were quite hard. I felt i held my own, but I'm disappointed and slightly amused that I've gotten the same Quant score even with improved practise and an improved time management strategy
2. Having said that I suspect i might have made a couple of foolish errors (E.g. calculating variance when the answer asked for SD, with a bunch of raw values provided), hence the test kept giving me opportunities till the very end to improve my score
3. The SC and RC proceeded like clockwork, but my CR felt.... paralysing. I was able to focus and process the information, but my reasoning skills felt sluggish - i guess i was just not test ready for CR.

Noting the above points, please let me know your suggestions (if you have any) for a self-diagnosis, and the best way to retake the test in the shortest possible time (2-3 weeks). Below are my thoughts regarding the same:
1. Go through each OG and Quant Review question in detail - categorize questions you're unable to solve within 1.5-2 mins by topic - sub topic, similar to how Manhattan does it. Lay out the key rules/tips for each topic and list out possible strategies required (e.g. SETS: piecemeal venn diagram method for 'only' questions, direct formula application, double set matrix method, etc.) - the goal is to have a set of strategies in my arsenal to apply based on the information given, aimed at reducing the time spent doubting/second guessing my approach
2. Repeat the same process by going through my error log for Manhattan Quant CATs - and then extending this to the 7-8 GMATCLUB CATs i've taken in the past. This should not only help familiarise the mistakes i'm prone to, but also systematically address issues at a sub-topic level by ingraining the most effective methods for a given problem type
3. Verbal - Review my approach for CR questions (based on the ESR) and work on my approach in weak areas

I'm currently waiting for to access my ESR, hope it comes through in another couple of days. In the meantime, I'll take a more leisurely look at the links you've shared in your response and look forward to your thoughts on the above. Couple of last points to note:
1. How do you recommend I diagnose my current quant level (been scoring 44-46 on 4 manhattan quants)?
2. Would you have any further suggestions that i may be missing out on?

Thanks and Regards,
Abhishek

[StaceyKoprince]

I'm glad you locked in that 680 score. Nice work.

You mention wanting to retake in 2-3 weeks. Is your goal still 730? A jump from 680 to 730 in 3 weeks is...ambitious. Most people would not be able to do that. Or is your goal more like a 700 now? That's more reasonable for a 3-week timeframe.

I would like to see what your ESR says, of course, before we know for sure what your plan should be. But if we assume that CR was the problem on the verbal side, I'd like to know more about how you were falling short there. There are multiple areas that can be problematic—you weren't correctly identifying the question type; you weren't correctly identifying the conclusion or you misread how the argument worked; you misread or misunderstood an answer choice; you fell for a particular kind of trap answer; etc.

We can't know from the official test questions, of course, so I'd want you to go look at recent OG CR problems you've done and analyze what went wrong.

Likewise, I want to see what your ESR says in terms of the specific areas (if any) that were more problematic on the quant side. If that shows that there are certain holes in your foundation that are holding you back, then obviously you'll want to work on those areas.

And to your comment about making some careless mistakes—yes, that can really hurt your score, especially when you are already at the top end of the scoring range. A few careless mistakes can easily drop you a point or two on the 2-digit scale. So part of the task is figuring out how to work systematically in a way that allows you to minimize careless mistakes. Part of that has to do with how you write stuff on your scrap paper, organize your work, etc—but part of it also has to do with NOT allowing yourself to get too sucked into really hard problems. Even if your time management is okay, you're still using a lot of mental energy on those really hard ones—and that leads to mental fatigue, which leads to careless mistakes on easier problems. If you had to choose between getting a hard one right and an easy one, I'd choose the easy one every time. If you think about the range of questions that you earn / are given on the exam, missing a problem at the easier end is far more detrimental to your score than missing a problem at the harder end.

Finally: GMAC is about to publish a new Advanced OG—300 hard problems (150 Q and 150V). The publication date is Sep 24th. So that could be a good source of new hard quant problems to work on, but it would mean waiting a little longer to take your exam. (Then again, if you do still want the 730...then I think you're likely to need more time anyway...so then this would be an appropriate resource for you on the quant side.)

The ESR is usually available about 3-6 days after you take the exam, so post the data when you have it. In the meantime, since you feel pretty sure that CR was an issue, dig into recent OG CRs (or do more) and analyze what I described above to try to pinpoint where / why you're struggling more on those. And then that "why" will help us to know what you need to do to improve.

[abhishekk810]

Hi Stacey,

Thanks for your continued correspondence. There was a technical glitch due to which my ESR was unavailable till now, the issue has been rectified and below is the ESR (i guess i do not have upload permissions, please provide the same if you'd like a copy of the ESR for easier visual analysis - and more specifically, a clear picture of 'Average Difficulty'). I also have my previous ESR (of a 640 score, please let me know if you'd like that uploaded as well to know the progression of my strengths and weaknesses).

I have underlined my weak areas as i saw fit - I would love your opinion on the same + strategies to strengthen each weakness

My ESR: Q 47, V 36, IR 7
Quant:
Score_Percentile - PS 44_49, DS 49_70 (Arithmetic 58, Algebra/ Geometry 56)
Tme Mgmt - Overall 1:58, PS 2:06, DS 1:48, Arithmetic 2:13, Algebra/Geometry 1:39
Skills - Geometry 0%, Rates/Ratio/ Percent 100%, Value/Order/Factors 44%, Eq/Ineq/Alg 80%, Counting/ Sets 71%
Performance Correct - Q1 (100%), Q2 (71%), Q3 (14%), Q4 (57%)
Difficulty Progression - Medium - Medium High - High - Medium High
Quarter wise time spent - 2:01, 2.56, 1:33, 1:22

Verbal
Score_Percentile - CR 31_57, RC 37_78, SC 40_87
Time Mgmt - CR 1:59, RC 1:54, SC 1:18
CR - 75% Analysis/ Critique, 25% Construction/Plan
RC - 80% Inferred Idea, 80% Stated Idea
SC - 85% Grammar, 80% Communication
Performance Correct - Q1 (62%), Q2 (86%), Q3 (86%), Q4 (62%)
Difficulty Progression - Medium - Medium - Medium - Medium
Quarter wise time spent - 2:48, 1:36, 1:12, 1:03

A summary of the test experience -
Quant: I started well enough, confident in my attempts at the tougher questions. However, in the second and third quarters, I felt the difficulty constantly increasing - I felt myself get carried away with certain questions and was trying to catch up from the second quarter itself. My issue was that a lot of questions felt hard - so i didn't have a clear strategy on which questions to leave, thereby wasting further time with my indecision and wondering which questions to invest my time in (the impact of this can be clearly seen in Q3, where i consistently faced high difficulty questions and my poor time management up until then meant that i did not have the focus required to tackle these questions, running short of time)

Even if your time management is okay, you're still using a lot of mental energy on those really hard ones—and that leads to mental fatigue, which leads to careless mistakes on easier problems. If you had to choose between getting a hard one right and an easy one, I'd choose the easy one every time. If you think about the range of questions that you earn / are given on the exam, missing a problem at the easier end is far more detrimental to your score than missing a problem at the harder end.

Your point above strikes a cord with the dilemma i was faced with during quant - I'm at odds as to how to guard against the same.

Verbal: I spent a little longer on the initial RCs and CRs in Q1 - but felt i was in control wrt to my time management. The disappointing bit is that in spite of this - my accuracy was poor on this section. I believe an RC in the last quarter that was attempted with less time may have led to a drop in my RC score. Also, throughout the test, my CR skills failed to come to the fore - I attempted each question rather numb and dumbly, in spite of a percentile of ~75-80 during practise.

We can't know from the official test questions, of course, so I'd want you to go look at recent OG CR problems you've done and analyze what went wrong.

I've analyzed my CR mistakes from 3 official mocks i'd taken 5-10 days before the exam - it appears my comprehension fell short, in this order
1. Overall stimulus comprehension - Line of reasoning used
2. Details/specific words
3. Answer choice analysis

Likewise, I want to see what your ESR says in terms of the specific areas (if any) that were more problematic on the quant side. If that shows that there are certain holes in your foundation that are holding you back, then obviously you'll want to work on those areas.

Would love your insights on pthese specific problem areas in quant.

You mention wanting to retake in 2-3 weeks. Is your goal still 730? A jump from 680 to 730 in 3 weeks is...ambitious. Most people would not be able to do that. Or is your goal more like a 700 now? That's more reasonable for a 3-week timeframe.

My goal is still an ambitious one, a 730 ideally. However, I am currently not working and can peacefully put in 5-6 hours a day at the least. I haven't yet booked my date. I also plan to purchase the New OG Tough questions, so thanks for that information. Currently:
1. For Quant, I'm analyzing each mistake i've made/any question i'm not immediately comfortable with in this order (GMATClub CATs - Manhattan CATs - Official Mocks)
2. For Verbal_CR, I'm working on each question type (1. Approach 2. A 5-10 Q test of easy questions (~90% accuracy is a success) 3. A 5-10 Q test of Hard questions (~70-75% accuracy is a success)
3. For RC, I PLAN to add a new strategy based on advice I've seen online - Immersion and an almost dialogue-like method to really comprehend the tone and flow of the passage

Big Questions

1. Could you please elaborate on what classes of questions fall under Construction/Plan? Do you have any tips on how to improve my accuracy on 'Construction/Plan' questions?
2. Could you please elaborate on the specific classes of questions that fall under Arithmetic?
3. Do you have any specific tips on how to improve Problem Solving accuracy, Geometry accuracy and time take for Arithmetic questions?
(I believe i do not fully utilize the power of the answer choices as well as PLUGGING techniques - I stick to other approaches. This could be a possible improvement avenue as well)
5. I'd like to keep my reach at Q50, V42, with Q49, V41 as my realistic limit. Keeping in time my full commitment to preparation for GMAT - What would be a realistic date to schedule my next attempt?

Again, thank you for the positivity (helped immensely to continue believing in myself) and support (helped to break down my arrogant assumptions and instead work pragmatically towards piece-wise improvement).

Look forward to your insights and suggestions!

Best,
Abhishek

[StaceyKoprince]

Ok, some very clear patterns in that data.

Quant
Q1 was excellent—100% right and avg time of 2m. But then you got sucked in. Definitely spent too much time in Quadrant 2 (Q2) and had to rush Q3 and Q4 as a result. You did lift quite high in Q2/Q3, so we should expect Q3 % correct to go down a bit...but it really went down. With only 14% correct, it's likely that you missed some lower-level questions there that dinged your score. You also picked up the % correct in Q4, but you would have dropped enough in Q3 that you likely dropped below your "true" scoring level, so in Q4 you were just trying to make up ground again.

Overall: The good news is that this trajectory shows that you likely have the capability to pick up at least another point or two just by making better decisions about where NOT to spend your time and mental energy. If you can get really strategic about cutting yourself off when appropriate, then yes, you won't lift quite as high in the middle—but you also won't drop as low. You'll have a steadier performance that you can carry through to the end of the section, and that's crucial, since where you end is what you get on the GMAT.

Content-wise, it's a little frustrating because we don't know how hard the problems were—it could be that Value/Order/Factors is at 44% because it just happens that you only got really hard problems in those areas. Still, though, we have to work with this—so this stuff, along with Geo, needs some attention. (Note: When you download your ESR, there is a set of FAQs that you can also peruse, and those FAQs include more details about exactly what they mean by "Value/Order/Factors." So check that out—I suggest taking screen shots. (Last I checked, I could only see those FAQs in the window while I was looking at the ESR—it wasn't just a public FAQ page. But maybe that's changed now.)

And then, if you want to see specific examples of what any categories mean, open up your OG and look at the labels at the beginning of problem solutions. They use those labels when classifying the problems.

There is also a bit of a trend in the overall data. You're spending more time on average on PS, but PS performance is lower—so that extra time isn't paying off. First, remember that consciously every time you're tempted to toss more time to a PS problem.

Also, you're spending more time on Arithmetic. So all of that together tells me that you're possibly struggling a bit to do some arithmetic-type calculations on paper (and you have to do more calculations "to the end" on PS). So you need to learn more shortcuts / test-taking strategies for avoiding tedious math. (Also: Is it possible that long problems in story form are slowing you down? What I call WoT—Wall of Text—problems are somewhat more likely on PS and on Arithmetic.)

Verbal
Yep, as you suspected, CR was the one pulling you down here. I'm seeing the same time management issue—too much time spent early, rushing late. The impact is similar but not quite the same, because Q is your stronger area. Basically, the test starts us in the middle of the difficulty range—which, on Q, is clearly easy for you. So you have a nice string in which you get yourself up into the harder difficulty range and then you start hitting "too hard" problems and allowing yourself to get sucked in.

On V, though, your skills are not quite as high, and so you start to hit those "too hard" verbals in Q1 rather than in Q2.

The key takeaway here: If you're doing well (for you) on the test, you're *always* going to hit those "too hard" questions somewhere. The question is just where—you'll hit them earlier in a section that's less strong for you and a bit later in a section that's stronger. Next, it's *always* possible to let yourself get too sucked into those hard questions—with the consequence that you spend too much time and mental energy early on, and then you run low on time / energy late in the section, your score falls, and where you end is what you get.

So you are literally training yourself to *avoid* getting sucked in when you hit the "too hard" ones. It's always going to happen. It's like training yourself to say No sometimes at work—you have to / want to sometimes! It's always going to happen!

My issue was that a lot of questions felt hard - so i didn't have a clear strategy on which questions to leave

This is something you study just like you study how to answer questions.

There are three broad categories:
Immediate bail #1: Within 15-30 seconds, I see that this is something I've previously decided I hate and don't want to do. For me, if I see cylinders or combinatorics, I guess my favorite letter *immediately* and move on. I don't even look to see if it's an easier one—I don't care.

Immediate bail #2: Within 30-45 seconds, I see too many "this is annoying me" details in the problem. It's got 4 variables AND it's got both an inequality and an absolute value AND it's got roman numerals? Forget it. Too annoying. My rule is that if I see 4 annoying traits, I definitely don't do that problem, and if it's got 3, there's a decent chance I won't do it but I'll base it on how annoying I think those things are—three mildly annoying traits and maybe I'll do it. Three pretty strongly annoying traits and I'm out.

Non-immediate bail: I think I can do the problem. I try the problem. I'm ~2-2.5 minutes in (on a 2-minute-average problem) and...things just aren't happening the way I thought they would. I'm lost. Or I got an answer that isn't among the answer choices. Or I find myself thinking anything along the lines of "But I just...." or "I should be able..." I just studied this. I should be able to figure this out. If I'm thinking that...then I do not actually know what I'm doing, so I should actually just guess and move on.

In response to your plan:
1. For quant: The analysis you describe is good. You need to add two things:
A) Easier / dirtier / quicker ways to get to the correct answer letter. Note that I said "answer letter" and not "answer." It actually doesn't matter whether you solve exactly to the text of the answer. All that matters is that you solve enough to know which answer letter is the right one, or which 4 are the wrong ones. PS is not actually all that different than DS in that way—people just get sucked into solving for the specific value or expression because they see it sitting there in the answers.
B) Explicit study around how to recognize when NOT to keep going with a problem

2. CR: I would do these in more mixed sets, in two ways. First, the CRs should be chosen randomly, so that you don't know what type you're about to have to solve. Second, I would mix these up with SC and RC for two reasons: (A) there may be mental fatigue / concentration issues around having to jump back and forth among question types, so I want you to practice in test-like conditions, and (B) you do want to keep up your SC and RC skills too, of course, so this is a good way to do that.

3. RC: I like what you've described.

In response to your big questions:
1. As I mentioned above, take a look at the ESR FAQ description and the OG classifications to see what kinds of things they classify as construction / plan. It doesn't map exactly to our classification. And for tips to improve, you're already starting to do that by analyzing why / how you feel short on your recent problems.

From your description, I would guess that you are sometimes failing to follow the overall line of reasoning and/or missing certain important details or the significance of those details? And then that would of course lead you to fall for a trap answer. Plan-type arguments are, at heart, about someone claiming that a certain sequence of events (a plan) will achieve its goal (the goal of a plan is the equivalent of the conclusion of a more traditional argument type). It's important to lay out the plan in chronological order: First we do A, that leads to B, and that leads to C, and finally that leads to D. My assumptions are that every step of the plan is going to work just the way I say it will, with no unexpected outcomes or other things happening. Is it possible that you're struggling with these just because you're treating them in the same way as more classic arguments and not as plans? Let me know what you think.

You also mention details sometimes being the issue—a lot of times they will qualify something (many, most, etc). Do you find that you're making errors based on either missing those words entirely or misinterpreting? (eg, they say many people will X, but you remember that as most or pick an answer choice that only works if the argument said most...but you didn't notice the discprepancy between most and many?)

2. As mentioned, above.

3. Quant: Yes, this is what I was talking about above. You 100% need to take advantage of the various strategies, as well as the excellent clues provided by the answer choices on PS. I'd focus your studies here—this is likely enough to get you the extra 2-3 points you're seeking. One question: How do you feel about geometry? If you dislike it overall, it could be enough here to make sure you know the basic formulas and rules so that you can answer any easier ones you get, and then just decide you're going to bail right away on the hard ones. There aren't usually more than 4-5.

[4. You didn't list an item 4. ]

5. It's good that you have more time right now to study—though I also want you to be careful not to do too much and just tire yourself out mentally. It's also the case that it's going to take some time for you to change that school mindset that you're still using a little too much. Your more realistic goal represents an increase of 2 points on Q and 5 points on V. I think the Q alone could be within your grasp in a few weeks—depending on how stubborn your brain gets on wanting to solve the quant —but V might take a bit more time. I'd probably give it more like 4 to 6 weeks.

Note: They are about to change their policies on rescheduling exams, starting on Sep 26th (in 3 days). See here for details:
https://www.mba.com/frequently-asked-questions/gmat-reschedule-cancel-price-changes

[rohitv68]

[quote deleted by Stacey]

Good Post! Thank you so much for sharing this pretty post, it was so good to read and useful to improve my knowledge as updated one, keep blogging

[StaceyKoprince]

Welcome to the forums!

Please only quote specific parts of posts when you have a question or comment about that part. Please don't quote an entire post—it takes up too much space and makes it difficult to scroll through posts. Thanks!

[abhishekk810]

Hi Stacey,

Hope you're doing well.

I gave the GMAT yesterday, and I'm elated to share that I got a 730 (q49 v40) this time round. Sheer hard work and perseverance, aided by your invaluable support!

Big thanks to you for your patient and detailed responses to my queries!

Cheers,
Abhishek

[StaceyKoprince]

That's exciting! Congratulations!!

If you have a little time to post about what you think helped make the difference for you in going from 680 to 730? I'm sure others would be very appreciative.

Good luck with applications!