Experts' Topic of the Week, 7/7/17: Get 4 new, “fake” GMATPrep tests!

Can somebody please confirm if the repetition bit has been fixed in the new update !

Also, Do the RC questions come together or are questions from a single RC spaced out .

Thanks,
Aspiring Guy

I looked into it a little bit more, and I'm skeptical that they've fixed the repetition issue. I read the update file, and there was no mention of it.

If you're doing an adaptive GMATPrep test (or a real GMAT), you'll typically see 4 RC passages, with a grand total of around 13 questions. On the "fake tests"? Well, that's exactly the problem that we're referring to when we say that the verbal tests won't be perfect. You might see more or fewer than 4 passages, and each passage might have more or fewer questions than we'd ideally like. If you follow the directions in the article precisely, the "fake tests" will generally come pretty close -- but you certainly won't get exactly 4 passages and 13 questions every single time.

Enjoy!

How many questions are there in each RC passage in question pack ? In OG, one RC passage has more than 6-7 questions which makes it slightly tough to solve passages.

Dear experts,

Need one help. In my official practice test I score 730. my GMAT is scheduled in less than 10 days and I am targeting a score of 750. kindly suggest whether I should solve only medium and hard level questions of question pack? Or I should solve easy questions also because most of the initial questions in the actual exam will be on an easier side?

Sent from my XT1033 using GMAT Club Forum mobile app

If you're scoring THAT high, then you definitely don't want to include the easy questions in your "fake tests." The questions will start out somewhere around the medium level, anyway -- so unless you make a bunch of incredibly unnecessary errors, you won't see any truly easy ones on your actual exam. If you want to do those easy ones as a warmup at some other time, that's great. But the "fake test" will feel way too easy if you include the easy ones.

Good luck with your exam, and please let us know how it goes!

I have already solved all the questions in question pack during my first attempt. Is there any way to reset the previous history and solve the questions again as mentioned in this post?

I'm 99% sure that the questions will appear as "unanswered" if you uninstall and reinstall the software. You might also have to re-enter your activation code from GMAC after you reinstall the software.

If anybody has tried anything else, please let us know!

Many Thanks!!

This has worked. Now you can be 100% sure about this

hi GMATNinja

Can we apply tactics on this forum for studying from online q bank of OG18(wiley.com) ?
The RC is culprit here since in OG, lots of passages have 7-9 Qs.

Also since you mentioned : On a typical GMAT exam, you’ll see around 13 RC questions (four passages), 17 SC questions, and 11 CR questions, give or take a few. On the “fake tests” why did we not select say 12 SC or 15 DS instead of selecting ALL in Gmatprep?

Great question! There are a few problems with Wiley's online question banks from the OG (or quant guide or verbal guide or whatever). In no particular order:

• Yup, you'll get however many RC questions appear with the passage in the book -- so as many as seven or eight questions per passage. On an actual exam, you'll see either three or four questions on each passage, without exception.
• Even in "exam" mode on Wiley's website, the question types aren't properly randomized. Last time I tried selecting all of the verbal questions, my exam consisted of nothing but RC questions. On an attempt at a quant exam, I saw 18 PS questions in a row. Not realistic at all.
• You can't actually select exactly 37 quant or 41 verbal questions. The dropdown menus only let you select round numbers of questions, like 30 or 40. (Though you could always just stop yourself after the correct number of questions, so this isn't really a big deal.)
• On quant, Wiley's software doesn't seem to support special characters, like roots or exponents or fractions -- or at least they don't show up properly in my browser. The square root of 2 just appeared as "2" in the questions, with no symbol at all. That's a disaster.

So maybe they'll upgrade the software at some point, but for now, there's no way to turn the OG into anything resembling a legitimate practice test. But that's not the end of the world: there's also value to just practicing OG questions, without time pressure.


You certainly could specify the exact number of questions you'd like of each type, but the problem again is RC: if you don't select ALL of the RC questions of whatever difficulty levels you're using, you're much more likely to have weird stuff happen, like an RC passage with only one question.

And if you're going to select all of the RC questions, you'll have to do the same with CR and SC, just so that you come close to having balanced representation of the three question types. If you choose ALL of the RC, but just 17 SC and 11 CR, then you'll see much more RC than the other question types. Basically, you'd be asking the software to select randomly from a customized question bank, but the bank would be dominated by RC.

Also, most people will want to pick two or three different difficulty levels for each fake test (either easy + medium, medium + hard, or all three, depending on your skill level), and it's just easier to let the software randomize the difficulty, instead of trying to figure out exactly how many you should choose of each question type and difficulty level to make the test semi-realistic.

I hope this helps!

GMATNinja Sir, can you please tell, how to give myself a score after doing this fake test...?

I am facing a problem in solving RC so please tell me what material should I read to do warm up and also get use to with the difficult languages just like use in RC passages.

Check out the sections on "limitations of fake tests" and "interpreting your fake test results" toward the very end of the article. These aren't actual, adaptive tests (which is why we lovingly call them "fake tests"), so they don't have actual scores, unfortunately.


I'd recommend starting here for general RC advice: https://gmatclub.com/forum/experts-topi ... 41004.html. You'll definitely want to take a look at the links at the end of the article, especially the links about recommended books and periodicals if you need some extra help with your reading skills.

Have fun studying, and welcome to GMAT Club!

GMATNinja

Does it make sense to manually enter data in an error log and then calculate
ability (% of correct answers, but proportional to the level of difficulty assigned
to each question)

See below data used for interpreting:
Total Weight- Is the sum of weight for that particular question type. For example- SC- Parallelism- If you attempted 6 questions. In those 6 questions, you have 2 easy(2 points each. So total is 4 weight), 2 medium(4 points. So total is 8 weight) and you have 2 hard question(6 points. So total is 12 weight).
So the total weight of SC- Parallelism- 24.

Ability- Is how many hard questions you got correct. In the above example, if you answered only easy and medium question correct. Then your ability is 12.

Ability %- its Ability/Total Weight: Based on the above example. Your ability % is 12/24. That's 50%.

Accuracy %- As the name suggest. Its your accuracy. for above example. You got 4 out 6 question correct. So your accuracy is 67%.

I found error log by Gnpth most close here.

I know this is far away from seeing x score out of 800 on screen, but it makes more sense to have some approximation.

I see the appeal of something like that! Since we know that the software doesn't do a great job of randomizing the questions -- even if you select the exact same settings every time -- your system would offer a way to easily compare your results from one "fake test" to the next.

But I'm not sure that you need it to be that complicated: honestly, the biggest thing to worry about is whether you're missing questions that are relatively easy for you. You could potentially see improvements in the hard questions, but if you're still executing badly on easy/medium questions, you'll have a really, really bad time on an actual adaptive test.

Bottom line: if you want to create something that's more consistently quantifiable for these, that's totally cool, but you'll also want to keep a very close eye on the easy and medium errors -- or, more broadly, errors on questions that are easy for you, regardless of whether they're labeled easy, medium, or hard.

I hope this helps!

GMATNinja

I have edited the main post as per shorter format, added revised timing strategies by bb
and a link for the external stopwatch.

An additional query:

You quoted:


Can I revisit the question packs in below form too:
1) If your average verbal score is below 30, select ONLY easy and medium questions, for all three question types (CR, RC, and SC). This step is especially important for the verbal section, so make sure to select the highest number possible in each of the six dropdowns (since the number in hard tabs shall now be 0).

Does this strategy help anyone as much as yours?
I think the biggest advantage of these fake mocks to know time spent per question and how my mock is paced throughout an hour,
an inference I can not deduce from an updated version of exam packs.

Hi All,
I am particularly asking this to moderators and test takers in India, can we get the GMAT-style test booklet and marker which GMATNinja mentions in this post locally here (at a cheaper cost); to buy this set at 17 USD (1,200 INR) doesn't look worth

The reason I want to buy this is I have seen many test-takers (also me) suffer while using this on the test day (MAYBE we are not habituated to use these weird markers and super big booklets)