Many observers have criticized the difficulty of the questions that appear in the OG (the Official Guide). The problem is not so much that the OG items are too easy as it is that the OG, like a traditional pencil and paper test, features items that range across the p scale. The high probability items (the easy questions) are generally of little interest to students who are seeking scaled scores at the far right tail. Suppose we group items into seven difficulty strata with the middle stratum called stratum x. The majority of items that one would encounter on a traditional pencil and paper test would come from the middle strata near x (i.e. x-1, x, x+1). In a traditional test, you would encounter very few questions at the top of the range (x+3) since these questions have undesirable characteristics for most test takers (it would provide low differentiation (discrimination) among low and medium ability test takers). Now suppose we create an adaptive test from this body of traditional questions. The first test item administered to each student would come from the middle stratum (stratum x). If it is answered correctly, we move to stratum x+1 for the second question. If the first question is answered incorrectly, we move to stratum x-1 for the second question. If the first two questions are both correct, we move to stratum x+2. If this questions is answered incorrectly, we move back down to stratum x+1. At the end of the test we would combine the information concerning difficulty and number of questions answered correctly and incorrectly to obtain an estimate of the net number of items this student would have answered correctly on a traditional paper based test (this is the raw score). The estimated raw score on the paper test is then used to read the associated scaled score from the paper test's raw to scaled score conversion table. Hjort

Last visit was: 19 Nov 2025, 09:31

It is currently 19 Nov 2025, 09:31

Toolkit

General GMAT Questions and Strategies

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.

Go to My Error Log Learn more

Request Expert Reply

Confirm Cancel

Nov 20
UCR MBA Impact - Live Q&A with UCR AdCom & Student
07:30 AM PST
-
08:30 AM PST
Learn what truly sets the UC Riverside MBA apart and how it helps in your professional growth
Nov 22
GMAT RC Masterclass
11:00 AM IST
-
01:00 PM IST
Do RC/MSR passages scare you? e-GMAT is conducting a masterclass to help you learn – Learn effective reading strategies Tackle difficult RC & MSR with confidence Excel in timed test environment
Nov 23
Conquer Algebra on the GMAT Focus Edition
11:00 AM IST
-
01:00 PM IST
Attend this free GMAT Algebra Webinar and learn how to master the most challenging Inequalities and Absolute Value problems with ease.
Nov 25
Try OnDemand GMAT Masterclass
10:00 AM EST
-
11:00 AM EST
Prefer video-based learning? The Target Test Prep OnDemand course is a one-of-a-kind video masterclass featuring 400 hours of lecture-style teaching by Scott Woodbury-Stewart, founder of Target Test Prep and one of the most accomplished GMAT instructors.

Back to Forum

Create Topic

GMAT Scoring Algorithm - My observations

1 2 3 4 5 6

amitdgr

Joined: 30 Jun 2008

Last visit: 21 May 2013

Posts: 536

Own Kudos:

3,124

Given Kudos: 1

Posts: 536

Kudos: 3,124

28 Oct 2008, 12:58

Kudos

Bookmarks

spiridon

This is a very contraversial issue, I believe first 10 questions bear more importance tho...

maybe they throw more experimental questions later on to throw u off or something but if u dont do well in the beginning u will have fewer chances to answer harder questions

also, who decides which questions are harder? stats?

for example, someone could struggle with divison or some simple computing crap like that and be very good in geometry and probability but end up never seeing those questions because of getting several wrong answer in the beginning of the test.

GMAC says that GMAT tries to give the test-taker a wide spread of questions (spread in terms of different topics; geometry,arithmetic etc...) I agree what might be tough for one, might be easy for other.

I have noticed on GMATPrep, I get good scores when I have minimum number of questions wrong in first 15.

Founder

Joined: 04 Dec 2002

Last visit: 19 Nov 2025

Posts: 42,385

Own Kudos:

82,117

Given Kudos: 24,107

Location: United States

GMAT 1: 750 Q49 V42 Verified score

GPA: 3

Products:

Expert

Expert reply

GMAT 1: 750 Q49 V42 Verified score

Posts: 42,385

Kudos: 82,117

17 Feb 2009, 15:40

Kudos

Bookmarks

Threads are a bit old but still very good - merged

nitya34

Joined: 04 Jan 2008

Last visit: 06 Mar 2014

Posts: 516

Own Kudos:

4,361

[1]

Given Kudos: 17

Posts: 516

Kudos: 4,361

[1]

24 Feb 2009, 12:51

Kudos

Bookmarks

I believe you will get the worst score when you do consecutive mistakes(it could be anywhere in 1-37 or in 1-41)

Founder

Joined: 04 Dec 2002

Last visit: 19 Nov 2025

Posts: 42,385

Own Kudos:

82,117

Given Kudos: 24,107

Location: United States

GMAT 1: 750 Q49 V42 Verified score

GPA: 3

Products:

Expert

Expert reply

GMAT 1: 750 Q49 V42 Verified score

Posts: 42,385

Kudos: 82,117

24 Feb 2009, 13:05

Kudos

Bookmarks

nitya34

I believe you will get the worst score when you do consecutive mistakes(it could be anywhere in 1-37 or in 1-41)

That's what GMAC seems to say - if 3 questions are missed in a row - the test things that the level was too high and needs to be adjusted down, so the score goes down.

If every other question is missed, that means the test evaluated your performance with maximum precision.

nitya34

Joined: 04 Jan 2008

Last visit: 06 Mar 2014

Posts: 516

Own Kudos:

4,361

Given Kudos: 17

Posts: 516

Kudos: 4,361

25 Feb 2009, 08:02

Kudos

Bookmarks

Yes and I experienced the same when i was running short of time in verbal(mid range) during real exam(I got 590-48,24,4.5) and did consecuting mistakes in CR and almost answered one easy RC at the end in 3-4 mins.
More practice is the only way we can tackle the GMAT monster
practice means we can be familiar with the pattern of Qs(be it DS or CR)

nitya34

I believe you will get the worst score when you do consecutive mistakes(it could be anywhere in 1-37 or in 1-41)

Mark53

Joined: 08 Mar 2009

Last visit: 13 Aug 2010

Posts: 9

Kudos: 0

10 Mar 2009, 02:53

Kudos

Bookmarks

There's still something I don't understand about the GMAT. Does your final score refect your absolute performance, or does it reflect your performance relative to other candidates who took the same test? In other words, can 2 people with the same quant and verbal score have a different score at the end?

As a non American, I really have difficulty in understanding why the rating and the structure of this test has been made so sophisticated. I simply don't see the point in using an adaptive test since the only way to provide an absolutely impartial and fair test is to propose the same questions to all candidates. From the moment when questions are different for each candidate (because they are statistically inferred), there is systematically a bias...

Mark

Founder

Joined: 04 Dec 2002

Last visit: 19 Nov 2025

Posts: 42,385

Own Kudos:

82,117

Given Kudos: 24,107

Location: United States

GMAT 1: 750 Q49 V42 Verified score

GPA: 3

Products:

Expert

Expert reply

GMAT 1: 750 Q49 V42 Verified score

Posts: 42,385

Kudos: 82,117

10 Mar 2009, 16:32

Kudos

Bookmarks

Mark53

You get a final score and a percentile.
The score stays the same always and is directly related to your raw scores (math and verbal). There may be a few minor overlaps in scores since GMAT operates in multiples of 10, but generally a set of math and verbal raw scores will correspond only to one GMAT score.

However, what GMAT also does, it provides with a percentile score that actually changes very slowly year after year. When I got 750 in 2002, it was 99th percentile. Today it is 98th. Here is a link to the document that shows the relationship between raw scores and percentiles - https://gmatclub.com/forum/t75809-frequently-asked-questions-about-gmat. Look for the GMAT Score Guide.

Let me know if this does not do it for you.

Mark53

Joined: 08 Mar 2009

Last visit: 13 Aug 2010

Posts: 9

Kudos: 0

11 Mar 2009, 07:28

Kudos

Bookmarks

Ok, thank you very much bb.

According to you, percentiles are redefined year after year so when someone takes the test and gets a score, he/she won't be compared with the people who took the very same test but with all people who took the GMAT during the year. This looks surprising, because a session may be more difficult than others (for instance when experimental questions happen to be particularly difficult).

Founder

Joined: 04 Dec 2002

Last visit: 19 Nov 2025

Posts: 42,385

Own Kudos:

82,117

Given Kudos: 24,107

Location: United States

GMAT 1: 750 Q49 V42 Verified score

GPA: 3

Products:

Expert

Expert reply

GMAT 1: 750 Q49 V42 Verified score

Posts: 42,385

Kudos: 82,117

11 Mar 2009, 10:45

Kudos

Bookmarks

Mark53

Experimental questions do not count for your score, though you are correct in the sense that they may eat up more of your time than others, which reconfirms the strategy that one should not spend more than 2.5 mins on a question. At the same time, the questions you see, may be completely different from the questions presented to your neighbor, esp. if you have a different level of difficulty (or could be different questions for the sake of being different), so comparing people from the same session does not really help since the test is adaptive.

As to the tests, they are and have been consistent, despite a number of claims, and have provided a good level of measurement of one's skills.

jimjohn

Joined: 19 Aug 2007

Last visit: 25 Apr 2014

Posts: 108

Own Kudos:

238

[1]

Posts: 108

Kudos: 238

[1]

26 Apr 2009, 22:19

Kudos

Bookmarks

Guys, in the GMAT, are the first 10 questions more important than the others? I remember taking a Princeton sessio last year where they said so but Manhattan GMAT says this is not the case. Any ideas?

I just took a practice GMAT from the practice tests mba.com sends out and I noticed that even though I got 15 wrong on the math section and only 5 wrong on the verbal section, my math overall score was higher. heres te breakdown of which questions I got wrong:

math:
9,14,18,20,22,26,28,29,30,31,33,34,35,36,37
overall math score: 46

verbal:
3,14,24,28,41
overall verbal score: 42

the only thing i can think of is that because i got an early questio wrong in verbal that really hurt me.

any thoughts? thanks!

icandy

Joined: 05 Jul 2008

Last visit: 15 Apr 2011

Posts: 621

Own Kudos:

2,157

Given Kudos: 1

Posts: 621

Kudos: 2,157

04 May 2009, 20:17

Kudos

Bookmarks

jimjohn

Whew!. You got only 1 correct in the last 10 Q questions and still got a 46 overall.

How ever, you got 5 incorrect all over the V section and they are spaced all over the place.. You still get 42. Lets say all the 5 Q's you got wrong are not experimental, You still should have answered 25 Q's correct out of 30 and you get a 42 in V. In V, the gains are more for every correct Q. I dont get it. Really crazy! Is this ETS PowerPrep or the latest GMATPrep?

Thanks

icandy

Joined: 05 Jul 2008

Last visit: 15 Apr 2011

Posts: 621

Own Kudos:

2,157

Given Kudos: 1

Posts: 621

Kudos: 2,157

06 May 2009, 17:54

Kudos

Bookmarks

k, I know that I am going to cause a stir for the nth time. But I will do so as I am starting to feel that GMAC has got me confused more than enough.

In page 17 of OG11 Myth vs Fact, GMAC goes on to say

It is true that the CAT algorithm uses the first 10 questions to obtain an initial estimate of your ability.

There is more but given that 1/4th of the questions do not count to the score and the first 10 are used to estimate the ability of the test taker, can we safely infer that the first 10 does not contain any questions that are not counted towards the test. IIRC, Stacey Koprince of MGMAT interviewed some one from GMAC and the gist of the story was that Not all questions in the final 5 are dummy questions. In a quant section this essentially boils down to 8-9 non counted questions from 22 questions. That is one Q out of every 3 does not count?

It is obvious that many test takers do run out of time in the end and GMAC's purpose of gauging the difficulty of a Question will not be served if the last few Q's are randomly answered. It is in the interest of GMAC to put as many Questions as possible into the middle 20 to get a gauge.

How ever, I might be wrong here as GMAC might take into consideration the time taken to answer a Question & location of the Question in the test (ie 35th out of 37) to count that particular test taker's answer in calculating the difficulty of the question. What might be a last Q for one test taker might be a middle Q for another test taker?

Has any one ran GMATprep a few number of times to see how things work out? Does the idea/rumor that the first and last quarter of the test contain almost no non counted questions and most of the non counted questions are in the second and third quarter of Questions has some legs?

All said, I am not saying that take 30 min for first 10 questions as many test prep companies say. I would still go with the 20 min for the first 10. How ever, I would prefer shrinking the time for 2nd and 3rd quarters to make sure that I have 2 min per Q on the last 10. Most of the times, it is the final Quarter of Q;s that most test takers, including myself, fall short on time.

On a side note, I am doing the GMATClub challenges and I am able to complete 32-33 Q's of Quant. Given that these are tough Questions, Do you guys think I will have time issues with Quant section? I am getting to the 38 range on verbal tests (MGMAT/GC). What kind of time strategies are you guys using? I have learnt to shed my ego on Quant. I broke my Quant into 10 min intervals 5 Q each and the last 2 get 5 min (ideally). The reason I chose smaller intervals is because it is very easy to loose time on a single Q in Quant. I am trying to do some thing on scratch paper unlike thinking on a V question during which sense of time can be easily lost. I broke my Quant into 4 pieces (11 + 10 + 10 +10 ) with almost equal times give or take a min or two around 19 minutes. What I have noted is I am getting to the last set of 10 Questions any where between 14-16 min and that is killing 3 Questions. I also noted that some times, I waste time on a V question making sure that I did not screw up with a silly mistake. Thoughts fellas?

Founder

Joined: 04 Dec 2002

Last visit: 19 Nov 2025

Posts: 42,385

Own Kudos:

82,117

Given Kudos: 24,107

Location: United States

GMAT 1: 750 Q49 V42 Verified score

GPA: 3

Products:

Expert

Expert reply

GMAT 1: 750 Q49 V42 Verified score

Posts: 42,385

Kudos: 82,117

06 May 2009, 22:50

Kudos

Bookmarks

You are bringing up good points and of course stirring the pot.

I do think you are after something here though as I always thought the middle portion of the gmat (questions 15 - 30) are filler questions that seem to be average difficulty even for higher scoring test takers. I got 49 on my math and I thought the questions that were in the middle of the test were fairly straightforward. Their level actually scored me and made me paranoid as I thought I was failing. I do have to say that through the entire exam I felt as though I was failing and for about half a second considered canceling my score (very briefly) but curiosity got the best of me.

gmatter10

Joined: 11 Nov 2009

Last visit: 24 Dec 2009

Posts: 6

Own Kudos:

Given Kudos: 1

Posts: 6

Kudos: 4

26 Nov 2009, 15:15

Kudos

Bookmarks

I am going to have to say I am very relieved to see a set of GMATPrep scores where the number of questions answered incorrectly on Quant were significantly higher than Verbal but the raw score on Quant was higher than the rawscore on Verbal.

My own experience with GMAT Prep 1 is that - I got 12 out of 37 questions incorrect on Quant and got a raw score of 48 and 8 of these 12 questions were on the second half of the test!

I got 7 of the 41 questions on Verbal wrong and the raw score was 40 with 4 out of first 15 being incorrect.

My mind was reeling with the mind-boggling scores until I saw a post on similar experience on this thread.

Thanks for sharing the Prep1 score breakdown.

osbornecox

Joined: 04 Dec 2009

Last visit: 26 Jan 2016

Posts: 373

Own Kudos:

Given Kudos: 4

Schools: Johnson '14 (A) INSEAD '14 (A) Booth '14 (M) Oxford '14 (A) Kellogg 1YR '13 (D)

WE:Consulting (Consulting)

Schools: Johnson '14 (A) INSEAD '14 (A) Booth '14 (M) Oxford '14 (A) Kellogg 1YR '13 (D)

Posts: 373

Kudos: 80

23 May 2010, 05:09

Kudos

Bookmarks

In my experience, there are THREE KEY guidelines to follow:
- First 8-12 questions REALLY set the tone
- For the remainder, try to get on AUTOPILOT mode (right/wrong/right/wrong is MUCH MORE PREFERABLE than wrong/wrong/right/right/right)
- FINISH THE TEST.

This for quant.

For verbal, you just have to get as many right as possible, period. I remember in a practice test I didn't make my first mistake until question 18, made only two more mistakes, and I got a 45.

dilipthelip

Joined: 22 Jan 2011

Last visit: 26 May 2011

Posts: 2

Kudos: 0

15 May 2011, 03:32

Kudos

Bookmarks

Hi hijort,
I got a clear idea abt the scoring mechanism in GMAt:).
Thanks so much

hitmoss

Joined: 05 Sep 2009

Last visit: 17 Jul 2015

Posts: 5

Own Kudos:

Given Kudos: 4

Posts: 5

Kudos: 15

15 May 2011, 07:49

Kudos

Bookmarks

It's really good to ready every single post!
Many Thanks to Hjort and everyone who has contributed here.

Long way to go!

saeedt

Joined: 06 Feb 2011

Last visit: 28 Mar 2012

Posts: 170

Own Kudos:

Given Kudos: 57

Affiliations: University of Tehran

Location: Iran (Islamic Republic of)

Grad GPA: 4

Concentration: Marketing

Schools: BU

GMAT 1: 680 Q45 V38 Verified score

GPA: 4

WE:Marketing (Retail: E-commerce)

Schools: BU

GMAT 1: 680 Q45 V38 Verified score

Posts: 170

Kudos: 47

18 Oct 2011, 09:00

Kudos

Bookmarks

Hey everybody, I hope this thread is still active. I have one wonder: I am weak at CR and strong at SC and RC. What would happen if i skip almost all CR and correctly answer all SC and RC? Imagine my band rises to 90 by correctly answering the first three questions which are SC. Then I guess one CR and again correctly answer 4 RC. Now my band must be in upper 90%. If this trend continues to the end of the exam, my band must remain in top 80%. Is it possible? My concern is this: will GMAT algorithm be allowed to give me more than usual CR when it realizes that I am skipping all CRs? What do you think about this strategy? Of course, pay attention that I mean educated guesses on CR in less than 1 minute. Hopefully I can correctly answer a few couple of CRs too.

I gave my first CAT two days ago. In the first half of Verbal, I almost correctly answered all RCs and SCs, but wrongly answered most CRs. And the point is that I messed the SCs and RCs in the second half because I had wasted much time for CR in the first half. Imagine the catastrophe: wasted much time on CR and did most of them wrong.

Founder

Joined: 04 Dec 2002

Last visit: 19 Nov 2025

Posts: 42,385

Own Kudos:

82,117

Given Kudos: 24,107

Location: United States

GMAT 1: 750 Q49 V42 Verified score

GPA: 3

Products:

Expert

Expert reply

GMAT 1: 750 Q49 V42 Verified score

Posts: 42,385

Kudos: 82,117

18 Oct 2011, 11:16

Kudos

Bookmarks

Also, see this one: how-does-the-gmat-algorithm-work-120297.html

runitback

Joined: 19 Oct 2011

Last visit: 18 May 2020

Posts: 14

Own Kudos:

Schools: HBS '17 Stanford '17 LBS '17

GMAT 1: 790 Q51 V50 Verified score

GPA: 3.93

Schools: HBS '17 Stanford '17 LBS '17

GMAT 1: 790 Q51 V50 Verified score

Posts: 14

Kudos: 9

27 Oct 2011, 17:42

Kudos

Bookmarks

For test takers who don't miss any questions in the early going, since they will consistently receive hard questions (which probably take longer to solve), won't they be more pressed for time than those who are receiving slightly easier questions? Just wondering how GMAC takes this into account, since it obviously doesn't make any sense to miss questions on purpose just so that you'll get easier questions in the future and not run out of time hahah

1 2 3 4 5 6

Moderator:

MBAACCOUNT2025

General GMAT Forum Moderator

444 posts

Prep Toolkit

Announcements

Master in Management, Finance, Business Analytics and more Master’s Programs!

Live Profile Evaluation Chat Session with ARINGO

Wednesday, November 19, 2025
09:00 am PDT | 10:30 pm IST
Location: GMAT Club Chat

Top GMAT Focus Debriefs

	Magoosh increases your score, guaranteed - 15% off ALL Magoosh GMAT plans with code GMATCLUB15 + Free GMAT Club Tests.
	Manhattan Prep: $79 Bonus Score Higher on Your GMAT Exam. Guaranteed.
	Target Test Prep - Save up to $325 $200 Discount + $79 Bonus GMAT Club Tests
	e-GMAT: 60% Off on GMAT Focus Get to 735+ from any starting point
	Experts' Global - ~~$1720~~ $960 End-to-End GMAT Prep + App Support: 150+ Videos, 2000+ Questions, 15 FLTs +
	GMATWhiz - $299 ~~$499~~ Experience the course through a free trial.
	GMAT Club Tests - Free Included with every course purchase of $149+ View all deals

GMAT Scoring Algorithm - My observations

Prep Toolkit

Top 5 GMAT Debrief Videos

615 to 715 in 15 Days (Ria's Strategies)

More than Quant/Verbal Abilities: Speed vs. Accuracy

745 with Only Self-Prep and GMAT Club

From 555 to 765: 210-point Improvement

Perfect 805 Score Debrief by Julia

My Rewards