Find all School-related info fast with the new School-Specific MBA Forum

It is currently 15 Sep 2014, 23:58

Close

GMAT Club Daily Prep

Thank you for using the timer - this advanced tool can estimate your performance and suggest more practice questions. We have subscribed you to Daily Prep Questions via email.

Customized
for You

we will pick new questions that match your level based on your Timer History

Track
Your Progress

every week, we’ll send you an estimated GMAT score based on your performance

Practice
Pays

we will pick new questions that match your level based on your Timer History

Not interested in getting valuable practice questions and articles delivered to your email? No problem, unsubscribe here.

Events & Promotions

Events & Promotions in June
Open Detailed Calendar

(Un) Scientific analysis of gmatprep

  Question banks Downloads My Bookmarks Reviews Important topics  
Author Message
48 KUDOS received
CEO
CEO
User avatar
Joined: 17 May 2007
Posts: 2995
Followers: 55

Kudos [?]: 434 [48] , given: 210

GMAT Tests User
(Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 05:23
48
This post received
KUDOS
6
This post was
BOOKMARKED
I tried to analyze the gmatprep software to analyze some things I've been always curious about. I've done gmatprep-1 5 times and know the answers to almost all CRs and RCs by heart.

1. What is the importance of the 1st 10 questions?
Experiment : If I mess up the first ten and kick but in the remaining can I reach my normal score (as the OG claims)
Methodology : Guess the 1st ten and focus on the last 27
Results : Q41 with 12 wrong overall.
Analysis: On the outset 41 seems like a decent Q score esp since I got 8 out of 1st 10 wrong. However, I've touched 50 with 12 wrong. This myth has been busted. Get those 1st 10 right, any amount of catch up later will not help. The next experiment will be to get the 1st 20 right and just guess the last 17 to gauge the importance of the last few questions.

2. Improving SC will greatly improve my final score.
Experiment : how important SC really ?
Methodology: I decide to guess all RCs and CRs and only focus on SCs.
Results - Final V score 13 :oops: . Although I got 14/15 SCs correct.
Analysis: This makes sense, since the majority of the SCs tend to appear after Q1-10. I will do another experiment and ignore SCs completely and only focus on the RCs and CRs to gauge their importance in the future.

ps. Suggestions, feedback and further questions on my experiments welcome.

Yup, I was at it again today. Next set of experiments :
3. What is the importance of the last 17 questions?
Experiment : If I get the 1st 20 correct and mess up the last 17 what will happen ?
Methodology : Kick butt in the 1st 20 and guess the last 17
Results : Q49 with 12 wrong overall.
Analysis: Whoa !!! 49 ? with the same number wrong as the 1st experiment ... By Question 20 I had received all the tough quant concept questions that I could have ever hoped for. However, I wasn't expecting a score as high as 49. Although that might be because I got a few guesses on the last 17 right. Even so I'm sure I'd have received 46 or above here. Goes to underscore the importance of those first 10 questions compared to the last few.

4. What is the importance of CR in Verbal ?
Experiment : If I get all the CRs correct and mess up everything else what will happen ?
Methodology : Kick butt in all the CRs and guess everything else
Results : V13 with 29 wrong overall.
Analysis: I think I've busted my own myth that CR is more important than SC in some way. This doesn't seem to be the case. I got 11/11 CRs correct and still could not achieve more than 13. The CRs were different to the ones I usually get (at 40+ level). I encountered 4 of these CRs in the 1st 10 questions.

5. What is the importance of the 1st 5 questions?
Experiment : If I mess up the first five and kick butt in the remaining can I reach my normal potential (as the OG claims)
Methodology : Ensure I get the first five wrong and focus on the remaining 32
Results : Q49 with 7 wrong overall.
Analysis: I, for some reason, thought that the 1st 5 are pretty crucial to the GMAT score. Boy was I wrong. I made sure that I got all first five incorrect (I actually solved them and then picked the wrong one just to be sure). Then I really focused on the last 32. Got a couple of them wrong but over all 30 correct. Final score = 49. Of course I am sure it would be a 50 if I had some more success in the 1st 5, however, this shows that even if you totally stuff up the 1st five, you can still pick it up and reach 49 Q, a most acceptable quant score.

6. Improving RC will greatly improve my final score.
Experiment : how important RC really ?
Methodology: I decide to guess all CRs and SCs and only focus on RCs.
Results - Final V score 6 :oops: . Although I got 10/11 RCs correct. Total 28 wrong.
Analysis: This was again, totally unexpected. I genuinely thought that RCs mattered, at least as much, if not more, than CRs and SC. This experiment showed that if anything, RC mattered LESS :shock: When I did the guess work for CRs and SCs at least I was getting a V score in the 10s. I certainly wont take this result as an indication that I can slack off on RCs just yet ..


Quote:
I thought I would add a unique expereince with GMAT Prep.

I took the GMATPrep2 yesterday and scored 3 wrong (all in a row) in the first 10. Questions 8, 9 and 10 to be specific. I then scored another 3 wrong on the next 10 questions and 1 wrong in the last 17.

Questions.............# Wrong
1-10.........................3
11-20.......................3
21-37.......................1

Overal Quant score: 50

I really do not think the first 10 or even 20 are as heavily weighted as most people think. MGMAT seems to follow this belief as well.

Last edited by bsd_lover on 05 Jun 2008, 13:32, edited 2 times in total.
Manhattan GMAT Discount CodesKaplan GMAT Prep Discount CodesGMAT Pill GMAT Discount Codes
1 KUDOS received
CEO
CEO
User avatar
Joined: 29 Mar 2007
Posts: 2593
Followers: 16

Kudos [?]: 189 [1] , given: 0

GMAT Tests User
Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 05:26
1
This post received
KUDOS
bsd_lover wrote:
I tried to analyze the gmatprep software to analyze some things I've been always curious about. I've done gmatprep-1 4 times and know the answers to almost all CRs and RCs by heart.

1. What is the importance of the 1st 10 questions?
Experiment : If I mess up the first ten and kick but in the remaining can I reach my normal score (as the OG claims)
Methodology : Guess the 1st ten and focus on the last 27
Results : Q41 with 12 wrong overall.
Analysis: On the outset 41 seems like a decent Q score esp since I got 8 out of 1st 10 wrong. However, I've eve touched 50 with 12 or so wrong. This myth has been busted. Get those 1st 10 right, any amount of catch up later will not help. The next experiment will be to get the 1st 20 right and just guess the last 17 to gauge the importance of the last few questions.

2. Improving SC will greatly improve my final score.
Experiment : how important SC really ?
Methodology: I decide to guess all RCs and CRs and only focus on SCs.
Results - Final V score 13 :oops: . Although I got 14/15 SCs correct.
Analysis: This makes sense, since the majority of the SCs tend to appear after Q1-10. I will do another experiment and ignore SCs completely and only focus on the RCs and CRs to gauge their importance in the future.

ps. Suggestions, feedback and further questions on my experiments welcome.


Interesting.

Keep posting your findings.
2 KUDOS received
SVP
SVP
User avatar
Joined: 24 Aug 2006
Posts: 2135
Followers: 3

Kudos [?]: 86 [2] , given: 0

GMAT Tests User
Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 05:27
2
This post received
KUDOS
bsd_lover wrote:
I tried to analyze the gmatprep software to analyze some things I've been always curious about. I've done gmatprep-1 4 times and know the answers to almost all CRs and RCs by heart.

1. What is the importance of the 1st 10 questions?
Experiment : If I mess up the first ten and kick but in the remaining can I reach my normal score (as the OG claims)
Methodology : Guess the 1st ten and focus on the last 27
Results : Q41 with 12 wrong overall.
Analysis: On the outset 41 seems like a decent Q score esp since I got 8 out of 1st 10 wrong. However, I've eve touched 50 with 12 or so wrong. This myth has been busted. Get those 1st 10 right, any amount of catch up later will not help. The next experiment will be to get the 1st 20 right and just guess the last 17 to gauge the importance of the last few questions.



I don't think SC is important to a good verbal. It's just that SC is the easiest to improve, therefore increasing your verbal score.

As for getting the first 10 questions right. I have long been of the belief that getting the first 5 or so questions is imperative to achieve a high score. Yes, the myth is fact and gmac's statement that all questions are weighed the same is bull.

btw, I don't think it's the first 10, I think it's more like the first 5-6.

Good stuff bsd. +1
4 KUDOS received
CEO
CEO
User avatar
Joined: 17 May 2007
Posts: 2995
Followers: 55

Kudos [?]: 434 [4] , given: 210

GMAT Tests User
Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 05:31
4
This post received
KUDOS
Interesting thought kidderek - I'll redo this experiment with 1st 5 wrong to gauge the importance of those.
3 KUDOS received
Senior Manager
Senior Manager
User avatar
Joined: 07 Jan 2008
Posts: 318
Location: Ann Arbor, Michigan
Schools: Ross Class of 2011
Followers: 6

Kudos [?]: 81 [3] , given: 0

GMAT Tests User
Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 05:50
3
This post received
KUDOS
I'm a believer that the first 10 aren't as heavily weighted as some people believe. I think the score decreases (and increases) will readjust by the number of questions the get get correct/incorrect in a row.

So will you get a higher score getting the first 3 right and missing the last 7 than if you miss the first 7 and get the next 30 right - I think so but that's because you don't get as many hard questions correct in a row.

The test becomes very complicated but makes sense that the algorithm will readjust your scores based based on serries of right and wrong answers. If the test "thinks" you are a 40q and it gives you a 42 level question it expects you miss it, if you do, and then miss a 40q it basically starts to revalute your score until it finds the range where you get questions right at and below but not above. If you miss the 40 it might give you another 40 or a 39 but if you miss that one it will start to make larger jumps basically giving you a question it knows you should get right if you then miss a 38 it will start with bigger jumps like 36, 32, 30 - this is why missing a lot of quesitons in a row kills you.

I'd be interested in seeing a test where you miss 10 questions 2-3 from the first 10 (non in a row) and then 6-7 from the rest of the questions with equal spacing between them. I'm sure this scoring algorithm is extremely hard to understand but I'm not sure you can "beat" it.
1 KUDOS received
Manager
Manager
avatar
Joined: 14 Mar 2008
Posts: 127
Schools: Chicago R2
Followers: 1

Kudos [?]: 17 [1] , given: 0

Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 05:56
1
This post received
KUDOS
Excellent idea!!! I'm awaiting final results!!!
1 KUDOS received
VP
VP
User avatar
Joined: 22 Oct 2006
Posts: 1443
Schools: Chicago Booth '11
Followers: 7

Kudos [?]: 149 [1] , given: 12

GMAT Tests User
Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 06:12
1
This post received
KUDOS
It's all about difficulty level. If you get questions wrong in the beginning then your at a lower difficulty level so you are starting off at a disadvantage, you need to get back to those difficult questions and get them right, so maybe the questions towards the end are worth more since you will only have a limited number of difficult questions to answer and if you miss them towards the end your score will decrease signicantly more, but if you get them right the test will view the beginning as an anomoly

solution: get every question right and devote enough time within reason to answer the question. NEVER go over 4 minutes for a question, and only do a handful of questions (maybe 2-3 questions) ovre 3 minutes and maybe 5-6 between 2 and 3 minutes, since you can make up for it if you know you can answer questions in a minutes time. I do think it is worth it to spend 4 minutes on a hard question you know you can solve and just guess on the end. I guessed the last 3 questions in Q and got a 50.(educated 20 sec guesses)
SVP
SVP
User avatar
Joined: 24 Aug 2006
Posts: 2135
Followers: 3

Kudos [?]: 86 [0], given: 0

GMAT Tests User
Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 06:23
terp26 wrote:
solution: get every question right


Alas, if it were only that easy :lol:
SVP
SVP
User avatar
Joined: 30 Apr 2008
Posts: 1893
Location: Oklahoma City
Schools: Hard Knocks
Followers: 29

Kudos [?]: 433 [0], given: 32

GMAT Tests User
Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 06:25
This type of analysis is priceless! bsd_lover, please keep up the testing. If you can post a detailed account of your methodology, others of us can also experiement and provide results to you in a sreadsheet, or screen shot of the right/wrong questions on the review page for the GMATPrep.

I agree with everyone here. The nature of the way CAT works is it's biggest vulnerability. A certain question at one point of the test IS weighted differently than the same question at another point in the test. For example, say Question A is a Level 45 question. It makes sense that this is so, because if you get Question A as the # 27 question once, it has 26 other questions to analyze with your response to that question vs. if you get Question A as the #2 question. Your correct answer to Question A at #2 should impress the CAT more than getting it right at #27, but again, much of how impressed the CAT is depends upon the level of questions received prior to getting Question A.
_________________

------------------------------------
J Allen Morris
**I'm pretty sure I'm right, but then again, I'm just a guy with his head up his a$$.

Get the best GMAT Prep Resources with GMAT Club Premium Membership

1 KUDOS received
Director
Director
avatar
Joined: 18 Feb 2008
Posts: 509
Location: Kolkata
Followers: 5

Kudos [?]: 58 [1] , given: 66

GMAT Tests User
Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 09:30
1
This post received
KUDOS
I had earlier told bsd reg my verbal exp and would like to share what i feel reg gprep after giving it 3-4 times...yes i agree in quants first ten questions are important...but also that one should not get say 4-5 answers consecutively wrong....i got 46 in quant with 14-15 questions wrong...yes...out of it,i got 7 correct in first ten....but then the questions i got wrong were never at a stretch 4plus or so...was always say 2 wrong one correct.....again wrong...5 correct...wud request others to observe too..after all we r helping each other...
Reg verbal,completely agree with bsd....sc however good u r,take that adv to give more to cr,rc...bsd,try this,get 10 out of 15 sc wrong but 80 % rc,cr cprrect specially Rc...may be jus one wrong....since u remember all the answers,try it and dude check your verbal score....i am sure 35 plus....keep updating... :lol:
Director
Director
avatar
Joined: 18 Feb 2008
Posts: 509
Location: Kolkata
Followers: 5

Kudos [?]: 58 [0], given: 66

GMAT Tests User
Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 09:33
of course +1 for you is by default. :-D
Senior Manager
Senior Manager
User avatar
Joined: 18 Oct 2007
Posts: 449
Location: USA
Schools: Tepper '11
Followers: 5

Kudos [?]: 56 [0], given: 2

GMAT Tests User
Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 09:47
We need someone who has taken the gmatprep so many times, they know almost all the answers.

then they can intentionally get questions wrong, or get questions right, and see what score comes up. No guesswork involved.

anyone?
Senior Manager
Senior Manager
User avatar
Joined: 07 Jan 2008
Posts: 318
Location: Ann Arbor, Michigan
Schools: Ross Class of 2011
Followers: 6

Kudos [?]: 81 [0], given: 0

GMAT Tests User
Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 09:49
I wouldn't discount SC I got from a 32 to a 40 just by fixing that portion of my test. Remember the algorithm isn't concerned with question type so it expects you to be able to get question correct at certain levels if you are 40+ in both RC and CR (like I was) but you have problem with 35-38 level SC you won't be able to build that score.

Same goes with DS and PS.
1 KUDOS received
Intern
Intern
avatar
Joined: 07 Apr 2008
Posts: 23
Followers: 0

Kudos [?]: 6 [1] , given: 0

Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 10:21
1
This post received
KUDOS
This is a great idea, bsd_lover. I can't wait for your future mythbusting analysis.
_________________

Head-to-Head Analysis: Veritas Prep versus Manhattan GMAT

1 KUDOS received
Director
Director
User avatar
Joined: 14 Oct 2007
Posts: 759
Location: Oxford
Schools: Oxford'10
Followers: 13

Kudos [?]: 175 [1] , given: 8

GMAT Tests User
Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 10:47
1
This post received
KUDOS
Kudos bsd!

btw could it be true?? is this u below?

Image


:lol:
2 KUDOS received
CEO
CEO
User avatar
Joined: 17 May 2007
Posts: 2995
Followers: 55

Kudos [?]: 434 [2] , given: 210

GMAT Tests User
Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 13:14
2
This post received
KUDOS
tarmac I am as good as it gets. I've taken the gmatprep about 5 times and my scores range from 770 to 790 on a consistent basis (and no its not because I am actually that good ...)

Tarmac wrote:
We need someone who has taken the gmatprep so many times, they know almost all the answers.

then they can intentionally get questions wrong, or get questions right, and see what score comes up. No guesswork involved.

anyone?



buffdaddy, your photo is missing my trident, my pointy tail and my horns.
1 KUDOS received
CEO
CEO
User avatar
Joined: 17 May 2007
Posts: 2995
Followers: 55

Kudos [?]: 434 [1] , given: 210

GMAT Tests User
Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 13:18
1
This post received
KUDOS
lsguy, Once I retake the test with CR and RC individually (and perhaps combined), this will become a lot clearer. of the 15 SC questions I did get, most were somewhat *easier* (as scientific as that sounds) than the ones that I get when I hit 40+ in V, so you may be correct.

Another aspect that I found interesting was that once I stuffed up the 1st 10 in Q, the level of Q questions became a lot easier (obviously). However, they were not as ridiculously easy as I expected them to be. I was exposed to many questions that I had never seen before. Quite a few of them had little tricks and I had to pay a lot of attention to detail to what the questions were asking. But there is no way the algorithm would adjust me to the regular level (of around 49) after the 1st 10 were gone.

lsuguy7 wrote:
I wouldn't discount SC I got from a 32 to a 40 just by fixing that portion of my test. Remember the algorithm isn't concerned with question type so it expects you to be able to get question correct at certain levels if you are 40+ in both RC and CR (like I was) but you have problem with 35-38 level SC you won't be able to build that score.

Same goes with DS and PS.
Senior Manager
Senior Manager
User avatar
Joined: 18 Oct 2007
Posts: 449
Location: USA
Schools: Tepper '11
Followers: 5

Kudos [?]: 56 [0], given: 2

GMAT Tests User
Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 13:31
bsd_lover wrote:
tarmac I am as good as it gets. I've taken the gmatprep about 5 times and my scores range from 770 to 790 on a consistent basis (and no its not because I am actually that good ...)


Oh okay. One suggestion I have though: I think you should either be intentionally getting the answer wrong, or trying to get it right, no need to answer randomly.
25 KUDOS received
CEO
CEO
User avatar
Joined: 17 May 2007
Posts: 2995
Followers: 55

Kudos [?]: 434 [25] , given: 210

GMAT Tests User
Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 15:18
25
This post received
KUDOS
Yup, I was at it again today. Next set of experiments :
3. What is the importance of the last 17 questions?
Experiment : If I get the 1st 20 correct and mess up the last 17 what will happen ?
Methodology : Kick butt in the 1st 20 and guess the last 17
Results : Q49 with 12 wrong overall.
Analysis: Whoa !!! 49 ? with the same number wrong as the 1st experiment ... By Question 20 I had received all the tough quant concept questions that I could have ever hoped for. However, I wasn't expecting a score as high as 49. Although that might be because I got a few guesses on the last 17 right. Even so I'm sure I'd have received 46 or above here. Goes to underscore the importance of those first 10 questions compared to the last few.

4. What is the importance of CR in Verbal ?
Experiment : If I get all the CRs correct and mess up everything else what will happen ?
Methodology : Kick butt in all the CRs and guess everything else
Results : V13 with 29 wrong overall.
Analysis: I think I've busted my own myth that CR is more important than SC in some way. This doesn't seem to be the case. I got 11/11 CRs correct and still could not achieve more than 13. The CRs were different to the ones I usually get (at 40+ level). I encountered 4 of these CRs in the 1st 10 questions.

Last edited by bsd_lover on 05 Jun 2008, 15:32, edited 1 time in total.
2 KUDOS received
SVP
SVP
User avatar
Joined: 30 Apr 2008
Posts: 1893
Location: Oklahoma City
Schools: Hard Knocks
Followers: 29

Kudos [?]: 433 [2] , given: 32

GMAT Tests User
Re: (Un) Scientific analysis of gmatprep [#permalink] New post 05 Jun 2008, 15:28
2
This post received
KUDOS
bsd_lover

This is an awesome experiement! You should get some major kudos for this.
_________________

------------------------------------
J Allen Morris
**I'm pretty sure I'm right, but then again, I'm just a guy with his head up his a$$.

Get the best GMAT Prep Resources with GMAT Club Premium Membership

Re: (Un) Scientific analysis of gmatprep   [#permalink] 05 Jun 2008, 15:28
    Similar topics Author Replies Last post
Similar
Topics:
3 Analysis of new GMATPrep software, errors found mmapplebeck 0 04 Apr 2012, 12:35
GMATPREPDS:UN ventivish 4 17 May 2008, 20:18
Experts publish their posts in the topic Un#$#%% belieavble...... aviroop 17 20 Apr 2008, 00:46
DS - UN gk3.14 1 17 Dec 2006, 13:34
UN Security Council ronybtl 12 01 Nov 2005, 05:44
Display posts from previous: Sort by

(Un) Scientific analysis of gmatprep

  Question banks Downloads My Bookmarks Reviews Important topics  

Go to page    1   2   3   4   5   6    Next  [ 115 posts ] 

Moderators: bagdbmba, WaterFlowsUp



GMAT Club MBA Forum Home| About| Privacy Policy| Terms and Conditions| GMAT Club Rules| Contact| Sitemap

Powered by phpBB © phpBB Group and phpBB SEO

Kindly note that the GMAT® test is a registered trademark of the Graduate Management Admission Council®, and this site has neither been reviewed nor endorsed by GMAC®.