Welcome! Check out our free B-School Guides to learn how you compare with other applicants.
Login or Register

Educated Guess

This topic has 5 expert replies and 5 member replies
ArpanaAmishi Rising GMAT Star Default Avatar
Joined
06 Jun 2010
Posted:
93 messages
Followed by:
1 members
Thanked:
2 times
Educated Guess Post Mon Jun 20, 2011 2:08 am
In an article I read that we should collect all the strategies that talks about how to guess intelligently, So opening this thread....request to share the experience on guessing...it should include all GMAT sections .

Experts are also highly welcomed.

I am still in observation mode , so nothing to share.

Need free GMAT or MBA advice from an expert? Register for Beat The GMAT now and post your question in these forums!
arun@crackverbal Really wants to Beat The GMAT!
Joined
10 Jun 2011
Posted:
223 messages
Followed by:
53 members
Thanked:
44 times
Post Mon Jun 20, 2011 9:43 am
"Guessing" is a science rather than an art. It is perfected by practice.

Here are some "generic" tips:

1. In Sentence Correction when stuck between 2 answer choices - pick the shorter one.

2. In Reading Comprehension if you are confused on a "main point" answer pick the one which has the topic/keyword that appears in the 1st few sentences.

3. In Critical Reasoning when stuck pick the one which repeats words from both the premise and the conclusion.

I could go on but it will be of little use Smile These are things you should learn using your own experiences. That is the best way to apply an educated guess.

Arun

_________________
Free Ebook on GMAT | GMAT Scoring, Study plan, top study mistakes etc
Download here: http://www.crackverbal.com/downloadebook.htm

Register for a free Online GMAT Class with us:
http://www.crackverbal.com/GMAT-Live-Online.htm

GMAT/MBA Expert

Brian@VeritasPrep GMAT Instructor
Joined
03 Jul 2008
Posted:
1025 messages
Followed by:
211 members
Thanked:
671 times
GMAT Score:
750
Post Mon Jun 20, 2011 11:10 am
Good question - and if I can add anything here I'd challenge you through practice tests and homework sets to develop a personal strategy for WHEN you should make an educated guess. The real usefulness of an educated guess on the GMAT is to improve time management, so in this same vein you can benefit from knowing when it's a good idea to make an educated guess and when it's not.

For example, my colleague David@VeritasPrep insists (and I believe him) that taking a full minute to narrow the answer choices down from 5 to 3 is almost always a waste of time. You're only taking your odds from 20% to 33% and you're using a full minute to get there - that minute could be used on another question to get you almost all the way to a correct answer on a question that you KNOW how to do.

Another consideration - the GMAT is a lot more sophisticated than is the SAT or ACT, so the test is pretty good at hedging against the educated guess. You may actually find that a completely blind 20% guess that takes 5 seconds has just as high a probability as a calculated 45-second "educated" guess, and saves you plenty of time to use to get another question right.

Which isn't to say that educated guessing can't be helpful, but I think it can be important to be strategic about when you spend the time and when you don't. If you're struggling enough with pacing that educated guessing is going to be a part of your strategy, you may want to give yourself a fixed number of "just plain guesses" in 20 seconds or less that you consider assets, like those Lifelines in Who Wants to Be a Millionaire. Say you realize that you'll probably run around 5 minutes short, plan on taking 3 "passes" that save you those 5 minutes, and use those "passes" when you realize within 15-20 seconds of looking at a problem that you probably won't have a shot at solving it correctly in 2-2.5 minutes' time.

Like Arun said, a lot of this comes from experience, so taking practice tests and getting a feel for how much time you need to save, which question setups tend to take you the most time, etc. can be extremely helpful for this kind of strategy.


And rest assured...there are a few educated guessing techniques that can help, but I'd push you even more toward truer process-of-elimination type strategies that aren't purely hunches. For example, on a lot of math questions you can use properties of numbers to at least eliminate a few answer choices (do you know that the answer MUST BE divisible by 3? Or even? Or end in 0? If so, that's not a guess...you've definitively narrowed down the choices). Or you can often look at an involved fraction and at least determine that the denominator will be larger than the numerator, so you're looking for something less than 1 in that case. If you can use time to DEFINITIVELY eliminate answer choices, that time used is more valuable than time spent trying to come up with an educated guess. And, ultimately, you need to have your own pacing strategy to determine whether you should spend that full minute trying to come up with a guess or to just "punt" right away...or whether you should have enough time to finish solving that question even if it takes a while. The key, like Arun said, is practice and experience.

_________________
Brian Galvin
GMAT Instructor
Director of Academic Programs
Veritas Prep

Looking for GMAT practice questions? Try out the Veritas Prep Question Bank. Learn More.

Get 5 Brand-New GMAT Practice Tests from Veritas Prep for Just $49! Veritas Prep’s computer-adaptive GMAT practice test are powered by artificial intelligence, guaranteeing that you see the most realistic GMAT questions and most accurate scores. Get Your Tests Now.

GMAT/MBA Expert

Ian Stewart GMAT Instructor
Joined
02 Jun 2008
Posted:
2178 messages
Followed by:
279 members
Thanked:
988 times
GMAT Score:
780
Post Thu Jun 23, 2011 3:52 pm
You'll read a lot about guessing strategies in various prep books - some prep companies seem to focus more on guessing strategies than on actual content. I've done a lot of research into the effectiveness of some of the guessing strategies I've seen in print, and I've come to the conclusion that if you don't understand a question, there is no way, based just on the structure of the answer choices, to guess any better than randomly. To list some of the guessing 'strategies' I've seen in print:

* in Quant PS, I've seen many books suggest you try to identify 'twin' answer choices, and guess one of those. For example, in a question like:

At an academic conference, the ratio of the number of men to the number of women was 2 to 3. What percentage of people attending the conference were men?

the reasoning is that the answer choices will include not only 40%, the correct answer, but also the 'opposite' of the correct answer, 60%, which represents the percentage who are women. Thus if there are only two 'twin' answer choices, guessing one of those increases the likelihood of guessing correctly from 1/5 to 1/2, or so these prep books say.

Well I looked through every PS question in both official guides to identify questions with 'twin' answer choices, and worked out how many right answers you'd expect to get if you always guessed a 'twin', and how many right answers you'd expect to get if you simply guessed randomly. It turned out that you'd actually expect to get slightly *more* correct answers by guessing randomly than by guessing a 'twin', though the difference was so small that it was statistically insignificant. In any case, if you're spending time looking for 'twin' answers, you're just wasting your time.

* In Quant PS questions which begin "which of the following..." some books suggest that E is more likely to be correct than any other choice, so you should guess E, the reasoning being that the question designers want you to do the maximum possible amount of work, so they want to force test takers working from A through E to work through as many answer choices as possible. I looked at every official 'which of the following' question I could find (there aren't that many, so I was looking at 31 questions in total), and A or B were correct 10 times, while D or E were correct 12 times - early answer choices were correct essentially just as often as later answer choices. So this kind of 'strategy' is pretty much pointless.

* In Sentence Correction, I've seen some people recommend guessing the shortest answer choice if you need to guess randomly, the reason being that good writing economizes on language, so wordy answer choices are normally wrong. A while ago I looked at every official SC question in the Official Guide to see how often each answer was correct based on the length of the answer choice. I don't have that data in front of me, but if you go only by the lengths of each answer choice, there is only a minuscule advantage to choosing short answers over long ones - if I recall correctly, a shorter answer choice had about a 23% chance of being right, and a longer one about a 17% chance of being right (20% of course being the chance a random answer is right). The result wasn't statistically significant either, so it may just have been coincidence. In any case, if guessing short answers is helpful on GMAT SC, it's only helpful by a negligible amount.

Those are just three examples of guessing 'strategies' I've read about, and which may sound plausible in theory, but which are almost completely useless in practice. The strategies I know of which *are* useful all require some understanding of the content of the question. Estimation, for example, can be helpful in many types of Quant problems. Number Theory can also often be used to eliminate answer choices. Often some answers are completely illogical and can be eliminated immediately (e.g. if a question asks for the remainder when you divide something by 4, the answer can only be 0, 1, 2 or 3, so you can eliminate any other answer choice).

_________________
Nov 2011: After years of development, I am now making my advanced Quant books and high-level problem sets available for sale. Contact me at ianstewartgmat at gmail.com for details.

private GMAT tutor in Toronto

Thanked by: ariz, chieftang
Post Thu Jun 23, 2011 6:53 pm
Well, with Arun and Ian and Brian already on this one I have to jump in!

Ian - I certainly agree that it is difficult to do better when guessing then the statistical odds. When I teach the LSAT, which is a paper test of course, the recommendation to the students is that if they have to do any guessing they should not waste even a moment looking at the questions, but rather just mark their answer sheets with whatever letter has appeared the least in that section. Of course we cannot do this on the GMAT, you must mark the screen and that means the question will be there but the principle can be the same.

The research Brian mentioned above is the result of some mathematical probabilities. Basically, if you take the experimental questions into account as well as the fact that you might guess correctly, a random guess has more than a 35% chance of either not counting, or you guessed correctly. So if you spend 15 seconds and randomly guess there is a 35% chance the question is not held against you (either you guessed right or it was experimental).

Now what if you take several minutes and narrow the question down to two choices? Now at this point you are not only hoping to have it not count against you - you are looking for some credit for getting it right, after all you spent say 2.5 minutes to narrow it down. Sorry to disappoint, but the odds are only 50 - 50% that you have guessed correctly of the two and there is around a 20% chance that the question is experimental, so you have somewhere around a 45% chance of getting a correct answer AND getting credit for it if you are guessing out of 2 answers.

It hardly seems worth it. In my book, any guessing is guessing. You want to have the greatest number of questions that you are CONFIDENT about. So when you can get one right do that and if you have to guess, guess early.

As to the Sentence Correction...
I have always said, "when you are down to two choices AND all else fails (grammar, logic, specificity, active voice, the way it sounds even) THEN Choose the shorter of the two. I always have some student want to choose the shortest of the 5!! As Ian mentioned this is not the path to victory! Do not just pick the shortest out of 5.

_________________
David Newland, JD, MA, MAE
GMAT Instructor, Tutor, and Author
Veritas Prep

Join Veritas 2010 Instructor of the Year Matt Douglas for G - MATT Mondays

Visit the Veritas Prep Blog

Try the FREE Veritas Prep Practice Test

Thanked by: ariz
Get 5 Brand-New GMAT Practice Tests from Veritas Prep for Just $49! Veritas Prep’s computer-adaptive GMAT practice test are powered by artificial intelligence, guaranteeing that you see the most realistic GMAT questions and most accurate scores. Get Your Tests Now.
ArpanaAmishi Rising GMAT Star Default Avatar
Joined
06 Jun 2010
Posted:
93 messages
Followed by:
1 members
Thanked:
2 times
Post Fri Jun 24, 2011 12:34 am
Thanks everyone....please keep it going ....

Just one thing the indend of this thread was not to do blind guess, rather how to choose one when choices are already narrowed down to 2 or 3 and we don't have any clue further.

GMAT/MBA Expert

Ian Stewart GMAT Instructor
Joined
02 Jun 2008
Posted:
2178 messages
Followed by:
279 members
Thanked:
988 times
GMAT Score:
780
Post Fri Jun 24, 2011 12:56 am
David@VeritasPrep wrote:
Now what if you take several minutes and narrow the question down to two choices?
I think this is the conclusion we all agree on: if you're guessing, guess *quickly*. That said, if by spending 15 seconds you can eliminate 2 or 3 wrong answer choices, that is surely time well spent. If you'd need to spend 2 minutes to do that, that's a huge waste of time, since 2 minutes spent elsewhere on the test will get you a right answer a lot of the time, not 1/3 or 1/2 of a right answer.

Just to give a couple of examples of what I mean, opening OG12 to a random page, if I look at Q214 in the PS section, this question asks for the probability three different things all happen. If you know the basics of probability, you can rule out the first two answer choices almost immediately, because they are far too high (and you can be pretty certain C is wrong as well). Or in question 212, the radius of the circle is clearly smaller than k, but not by much - answer B almost has to be right, and you certainly wouldn't guess anything but B or C if you do a 10-second estimate. In each case you can eliminate 2 or 3 answer choices within 15 seconds, but you need to understand something about the question. Now, it doesn't happen all that often that you can eliminate so many answer choices so quickly, but when it does, it's helpful to take advantage of the fact.

David@VeritasPrep wrote:
The research Brian mentioned above is the result of some mathematical probabilities. Basically, if you take the experimental questions into account as well as the fact that you might guess correctly, a random guess has more than a 35% chance of either not counting, or you guessed correctly. So if you spend 15 seconds and randomly guess there is a 35% chance the question is not held against you (either you guessed right or it was experimental).
I think the rest of this post might only be of interest to David and other GMAT specialists here:

I can see where you can get a figure around 35% (1/4 of the time the question doesn't count, and (3/4)(1/5) = 3/20 of the time the question counts and you guess correctly), but I don't think, in practice, it's the right number to use for most test takers. People tend to guess at questions which are hard for them (or at least ones which appear hard at first glance). The difficulty level of experimental questions covers the full range from easy to hard - unlike the questions which count, experimental questions do not adapt to the test taker's level. If a test taker is near the top level of ability (a 49/51 test taker, say), then the only questions she will consider guessing at are those that are at the highest difficulty level. Those are almost never going to be experimental questions, because most experimental questions will be well below the 49/51 level.

I just did a quick thought experiment to estimate how much time is worth spending to eliminate wrong answers. My assumptions might be bad, but I looked at things this way: eliminating two wrong answers is worth about 1/3 - 1/5 = 0.13 correct answers. Since test takers normally get 2/3 of the questions they spend time on right, 2 minutes is worth 2/3 - 1/5 = 0.47 correct answers (notice that, according to this methodology, an answer which is 100% certain to be correct is worth a bit more than 3 minutes of time, which seems close to right, but may be too high - certainly a certain correct answer should be worth more than 2 minutes). So 0.13 correct answers is worth about 33 seconds of time. That is, if you can eliminate two wrong answers in 30 seconds, it's probably time well spent. If you're only going to eliminate one wrong answer, it isn't worth spending more than about 11 seconds to do so - in that case, guessing randomly is probably the best idea. And if you might get to a 50-50 situation, it's worth spending some time to do so.

_________________
Nov 2011: After years of development, I am now making my advanced Quant books and high-level problem sets available for sale. Contact me at ianstewartgmat at gmail.com for details.

private GMAT tutor in Toronto

Post Fri Jun 24, 2011 7:26 am
Good stuff Ian!

So for those at the top range we can agree that they are more likely to be penalized for guessing, (which makes sense we don't want people getting 90% + by clever guesses) while someone in the middle ranges may be more in line with the figures that I mentioned. It all makes good sense because presumably your high scorer will not be randomly guessing very often!

Ian I think that your figure of 30 seconds to eliminate two wrong answers is a good one. I would make that trade. Certainly in data sufficiency and sentence correction in particular this is almost always possible.

Back to ArpanaAmishi
- You are bringing us back to the area of educated guessing - not random guessing. Actually as Ian has stated probably the biggest takeaway is not to spend too much time guessing.

Frankly, if you are looking to score well you should not do too much guessing. I am really only in favor of any guessing for three reasons:

A: You have looked at a Quant problem for a given period of time, 1 minute say, and you have no idea how to solve the problem. If it is Data Sufficiency see if you can evaluate either statement. If problem solving see if anything can be eliminated. Answer quickly from the possible choices and move on.

This is the guessing that may actually benefit some people - because it eliminates a bunch of forced guessing later. If you are one who often runs out of time you should pay attention to this.

B: You are way behind on time and you need to do something. This is where my discussion of random guessing really comes in. If you have limited time at the end of the section you are better off to just randomly guess (sacrifice) some question or questions so that you can address the others fully. People who try to "halfway" address all of the questions - say 10 questions in 5 minutes usually end up missing most of those.

C: When you have done all you can and there is literally no way for you to decide between the remaining choices. I think that this is what you are really asking about in the posting. This situation should be pretty rare. You should hard work to conclude most questions by choosing an answer for a reason.

As to the various question types, here are some thoughts:

Never randomly guess at Data Sufficiency or Sentence Correction.

1) Data Sufficiency - you might not even understand what the question is asking, but you can often evaluate at least one of the statements. And, in fact, many data sufficiency questions quickly come down to just two options. Such as, a specific number question where each statement clearly gives multiple options so you know that A B and D are off of the table. In most data sufficiency when you are down to two or three choices is where the real work begins.

I would not refer to this as any sort of guessing - but rather as just being smart. Take what the question gives you. At this point, having gotten the choices down to this level if I had any time I would not want to guess, but to finish the problem.

2) Sentence Correction. There are often differences in the answer choices that can be exploited and so you can eliminate two or three choices right off the bat, like data sufficiency. It is worth the time to do this since these are usually based on solid grammatical differences, so that you can confidently eliminate the plural verb when a singular verb is needed, etc.

As mentioned if you are stuck between two choices do not take too much time debating. Look for difference between the choices, try them in the sentence to see if one sounds better, do whatever you can to choose the one that is best. AND THEN if you find that they are completely tied I just say choose the shorter, because it is a bit more likely. If there is any reason to prefer the longer one though you should choose that.

3) For Reading comprehension Do Not Guess. (How does that sound?)

As Ian has mentioned in other postings you do not want to get these questions wrong as they can be below your level and can hurt you. Also, these answers can often be found in the text.

Rather than guessing you should take a moment and try to answer correctly.

For main idea questions I am a big fan of the last paragraph as this is where the author usually lets her opinion come through. Others mention the beginning of the passage.

4) Critical Reasoning - Again, do not guess. (I am sorry to say it but there should not be two choices that appear to be so equal that you must guess).


5) Problem Solving - this is the type where you might do the most guessing. Of all the question types this is the one that you are most likely to look at and have no idea how to begin or even to quickly eliminate choices (after all even the most difficult verbal question begins with you reading something).

As to "educated guessing" this is a matter of eliminating any choices that you can.

_________________
David Newland, JD, MA, MAE
GMAT Instructor, Tutor, and Author
Veritas Prep

Join Veritas 2010 Instructor of the Year Matt Douglas for G - MATT Mondays

Visit the Veritas Prep Blog

Try the FREE Veritas Prep Practice Test

Thanked by: limaki
Get 5 Brand-New GMAT Practice Tests from Veritas Prep for Just $49! Veritas Prep’s computer-adaptive GMAT practice test are powered by artificial intelligence, guaranteeing that you see the most realistic GMAT questions and most accurate scores. Get Your Tests Now.
ArpanaAmishi Rising GMAT Star Default Avatar
Joined
06 Jun 2010
Posted:
93 messages
Followed by:
1 members
Thanked:
2 times
Post Fri Jun 24, 2011 9:58 am
I am noting down each and every important point and after compilation will share it for everyone use...

More suggestions are welcomed ....I think test takers should also share thier views/experiences....

ArpanaAmishi Rising GMAT Star Default Avatar
Joined
06 Jun 2010
Posted:
93 messages
Followed by:
1 members
Thanked:
2 times
Post Wed Jun 29, 2011 3:38 am
Didn't see any further movement...Please share your experiences and I am sure it would be beneficial for forum.

arun@crackverbal Really wants to Beat The GMAT!
Joined
10 Jun 2011
Posted:
223 messages
Followed by:
53 members
Thanked:
44 times
Post Fri Jul 01, 2011 12:52 am
I think everyone has said whatever that had to be said. I really don't see anything left Smile

This is perhaps one of the best threads on BTG and I have already recommended this to a few of my students!

Arun

_________________
Free Ebook on GMAT | GMAT Scoring, Study plan, top study mistakes etc
Download here: http://www.crackverbal.com/downloadebook.htm

Register for a free Online GMAT Class with us:
http://www.crackverbal.com/GMAT-Live-Online.htm

Best Conversation Starters

1 mukherjee.tanuj3@... 53 topics
2 lukaswelker 41 topics
3 candygal79 32 topics
4 shibsriz@gmail.com 25 topics
5 gmattesttaker2 18 topics
See More Top Beat The GMAT Members...

Most Active Experts

1 image description GMATGuruNY

The Princeton Review Teacher

156 posts
2 image description Brent@GMATPrepNow

GMAT Prep Now Teacher

134 posts
3 image description David@VeritasPrep

Veritas Prep

52 posts
4 image description ceilidh.erickson

Manhattan GMAT Teacher

46 posts
5 image description Bill@VeritasPrep

Veritas Prep

44 posts
See More Top Beat The GMAT Experts