Educated guessing ... Any specific techniques ?!

[II]

Hi There,

Would be keen to learn from the rest of this forum ... what have been your experiences, techniques when coming across a difficult question and having to make an educated guess.

What techniques have you used to make educated guesses ?

Thanks.

[lunarpower]

whoa there, that's an awfully general question.

let's start by narrowing it down a little bit...

MATH: PROBLEM SOLVING

ESTIMATION
some problems can be solved by good old-fashioned estimation. here are the general qualities of such problems:
1) they have numerical answer choices that are spread fairly far apart.
2) they are accessible to estimation in the first place.

by (2) i mean that there has to be some intuitive way of figuring out the size of the quantity requested in the problem. so if it's a geometry problem, for instance, with a nice accessible diagram (whether one that's on the screen or one you drew yourself), guess away. if it's a combinatorial problem (how many teams of six can be picked from the following thirteen people?), then there's probably no real way to get a handle on the size of the answer.

if you decide that you can estimate the answer to a problem, then the process is usually fairly self-explanatory, but here are some tips:
* if you have to estimate two or more quantities to get an answer, then try to estimate the quantities in offsetting ways. for instance: say you're trying to estimate the product of two quantities. then if you slightly overestimate one of the quantities, then it's better to slightly underestimate the other quantity (so that the two errors cancel each other out, at least to some extent).
* take advantage of mental math shortcuts. if you have to divide by 5, accomplish this by multiplying by 2 and then divide by 10. if you have to subtract 97, then subtract 100 and add back 3. etc.

PICKING NUMBERS
if you ever have undetermined quantities in a problem - which usually take the form of variables that appear in the problem but also in the answer choices, but can also take more subtle forms such as 'the volume of a container' or 'the total number of students in a class' - you can usually just pick numbers for those quantities, reducing the problem to arithmetic. this isn't so much a guessing method as an alternative solution method, but it's still worth mentioning. this is a common theme in prep books; in particular, it's illustrated in detail in our equations and inequalities guide.
you have to pick numbers intelligently, though. for instance, if a problem mentions flasks that have, respectively, 1/3 and 1/5 the volume of a large container, then 15 would be a smart choice for the volume of the container, whereas 100 would not. on the other hand, if the problem dicusses percentages of the container's volume, then 100 would be a smart choice while 15 most likely would not.

DATA SUFFICIENCY
the best guessing advice here is to note that many data sufficiency problems have one statement that definitively qualifies as 'the easy statement' for that problem. in other words, there will be one statement for which the sufficiency issue is much, much easier to resolve than for the other statement. for instance, the problem might ask for some combination of x and y, or ask whether x is greater than y, and then statement (1) might just say 'x > 0' (which is clearly insufficient, because y is not mentioned at all).
if statement (1) is the easy statement, then you can eliminate A and D if it is insufficient, and B, C, and E if it is sufficient.
if statement (2) is the easy statement, then you can eliminate B and D if it is insufficient, and A, C, and E if it is sufficient.
almost all 'easy statements' are insufficient; it's extremely rare for the gmat to include a problem on which one of the answers is obviously sufficient (unless the problem is quite easy, in which case you won't be wondering about how to guess on it!)


SENTENCE CORRECTION
make sure that you read through answer choices vertically: i.e., don't read through entire answer choices (at least not at first). instead, glance down through the answer choices, noting differences among them. best-case scenario is you'll see something that's a clear decision between two grammatical opposites (like singular/plural, active/passive, or present tense/past tense); in this case, you should be able to kill off one of the two options.
if you have to guess, try to find at least one or two of these 'splits' and decide accordingly. also, do a quick scan of the choices for redundancy and wordiness - two concepts you'll grow to recognize, if you do enough diligent work in the OG - and you should be able to knock off your share of answers.

CRITICAL REASONING
these vary by question type, but the strongest guessing technique i've seen under strictly limited time is quickly eliminating any choices that are irrelevant or that overstep the bounds of the question. there are almost always some choices that clearly have nothing to do with the question, or that are way too strong (or weak) to be the correct answer. go for those first.
also, this is universal advice for critical reasoning whether you're guessing or not, but ALWAYS find the conclusion of the passage to help you understand it (hence an 'educated' guess). you won't get anywhere with any passage unless you understand its logic, and you won't understand the logic unless you can find the conclusion.

hope that helps... there's a lot more to say, but this should be a good start (and i'm out of time for the evening).

cheers

[II]

Hey there ... thanks for the detailed response ... much appreciated.

[arora007]

thanks Ron!!

[David@VeritasPrep]

Ron really covered this one! I just had one thing to add...Remember that "Educated Guessing" is still guessing. You do not want to spend too much time on it as I will describe below.

First let me say that Sentence Correction and Data Sufficiency (as Ron has described those techniques above) can usually be narrowed down in less than a minute so those are good candidates for educated guessing. For example, on Data Sufficiency, the actual problem could be written in a foreign language (with the statements in English) but if statement number 2 simply said "x is a positive number" we would know that this is not sufficient and have eliminated choices B and D. So you should really never blindly guess on Data Suff or Sent Corr as you will usually be able to quickly eliminate - even if you do not fully understand what is going on!!

Critical Reasoning as Ron has discussed above is susceptible to the quick elimination of irrelevant choices. So this can be a pretty good candidate for educated guessing. Not as good as Data Suff or Sent Corr because you generally have to understand what is going on in Critical Reasoning to be able to eliminate, but it should be quick.

Reading Comprehension is different situation since you will have read the passage, so guessing should not be too frequent but it similar to critical reasoning in that elimination is usually possible.

That leaves problem solving. The difference between problem solving and the other types is that it is really hard to do a problem solving problem part of the way. If the answer choices are "51, 52, 53, 54, 55" you are not going to have much luck doing this part way. It is often all or nothing - you either get all the way to the answer or you might as well not have spent the time.

So that brings us to educated guessing on problem solving.

It may seem like there is a huge difference between guessing from 5 choices and guessing between 2 or 3 but there is really not such a big difference. Trust me on this one...the presence of experimental questions means that you always have less than a 50-50 chance of getting credit for the correct answer if you guess between two equally likely choices.

The reason that I brought this information to your attention is for one simple reason, not to tell you to blindly guess, or to somehow say that you should not eliminate choices and estimate as Ron has described, but rather to emphasize that you should not Spend Too Much Time Making "Educated Guesses." Depending on how much time is a factor for you, it is probably not worth devoting 90 seconds just eliminating answer choices. Even if you get it down to just 2 the odds are not in your favor and you might be able to do an entire question later in those 90 seconds. However, if you can use the techniques above and narrow things down quickly, in say 30 seconds, then it would be worth it.

Just something to think about.