several people has asked me if it is possible to improve your quant score form 44-46 range to 50+, how this can be done and how much time is needed..... so i gave it some thought....

obviously - everyone is different.... each has different weaknesses and stregths so there is no generic answer to all - but still, i think that there can be some rough guidelines for doing this leap.... here are my thoughts about it.

if you are constantly performing at the 44-46 quant range (previous gmats, trustable prep material) i guess you:
- mastered most of the required skilled
- understand the structure and requirements of gmat math questions

your problems are likely to be in the following areas:
- "silly mistakes" i.e. things that in the pressure of the gmat you do wrong although you know and understand them
- get yourself into unnecessary complications in solving some of the questions.
- timing issues... there are some things that just take you too much time to get right.
- mistakes that are a result of a subtlety of concepts that you werent sure about (which made you both spend time, eventually guessing or answering without confidence, sometime getting it wrong)

the most common prep mistake (in my view...) that people do:
- concentrating on advanced concepts and hard questions. the ROI of this kind of prep is small (it would be higher if you were to advance for 48-49 to 51).

what you should do (again.... my own opinion):
- perfect your basic techniques. for example: solving regular linear, one variable equations should be performed with no mistakes at all, and should take you no more than 20-30 seconds; doing simple arithmetic - again... no mistakes are allowed, do it fast and with confidence... etc...
- boost your confidence. my suggestion - do questions again and again, until you can do, in time, without guessing and providing full answers, until you can do a full test (37 questions of various difficulty) with no more then 2-3 mistakes.
- try to look for the simplest way of solving things. don't "buy" comlex explanations. for 95% of gmat questions there are simple explanations. also, don't sufficient yourself with just one way to solve questions. for many questions there are more than one good way to solve. familiarizing yourself with all of them will help you find the simplest way to do it.
- look for specialized material that concentrates (correctly) on subtleties of concepts, especially the concepts you feel are difficult for you. this is the most difficult part, as it is hard to come across specialized question sets. in fact, i thought of compiling such sets, but unfortunately i don't have much time for that. on the other hand - you can use this forum wisely. if you ask about a subtlety of a concept in this forum i'd probably answer giving 2-3 gmat like questions that uncover it.
- try GMAT Club tests - these are hard math problems designed to make you work hard for every question

again these are general guidelines....
as you can see the ROI (or effectiveness) of a prep method depends on where you are and what you want to achieve.
concentrating on advanced concepts is good for 48->51
the above is better for 44->50
for 36-40 -> 45 you should probably spend more time reviewing concepts that are not clear to you (there are certainly some if you score less than 40). for under 35 you should probably continue review basic skills before moving forward....

any opinions? comments?

Great advise hobbit.

An additional trick that I often use is that after I solve a problem I would try to verify it with a different method, or just see if it makes intuitive sense. When you practise it will help if you could try to come up with more than one solutions.
_________________

Keep on asking, and it will be given you;
keep on seeking, and you will find;
keep on knocking, and it will be opened to you.

for me, a non native english speaker, improving on quantitative was a lot easier and a better ROI than improving on verbal...and it really paid off for me on the final day..

i would recommend focussing on accuracy in the beginning. Do the last 50 OG questions, 10 at a time after thoroughly understanding the concepts. Spend as much time as possible to ensure you get all right. for me, it took almost 2 and a half hours to complete. Once you find, you are getting 9 out of 10 right every time, you can move to practicing on a timed test, and try finishing 37 questions in 60 minutes. ( Keep accuracy at the top of your head, if you have to guess, try to eliminate at least 3 wrong answers.), but move on. This balance between speed and accuracy is the key to a higher quant score.

I concur with hobbit that you may not need Venn Diagram to answer the questions. Also you don't need to know how to calculate std dev but you'll need to know the concept. The concepts of mean, median, range and std dev are often tested together. Just remember that knowing one or a few of these measures may not tell you anything about the others. For example, same mean and a higher range may or may not indicate a bigger std dev.
_________________

Keep on asking, and it will be given you;
keep on seeking, and you will find;
keep on knocking, and it will be opened to you.

Could someone provide specific questions to give us an idea what the questions in the Quant 45+ bin look like and a list of strategies we need to learn to solve these correctly and quickly? Thanks for the thread, it's very illuminating.

how is the problem on the bottom solved without calculation?

I'm not sure we can say one of the roots is 3 if we have x(f(x))=3. The only option I would be able to eliminate would be x=0, since 0 times anything would not be 3. However to sort throught the rest 4 options, I'd still have to break the expression down to x^3(x-3)+2x(x-3)-(x-3)=0.
_________________

Keep on asking, and it will be given you;
keep on seeking, and you will find;
keep on knocking, and it will be opened to you.

Good point. If we limit the solution set of x(f(x))=3 to integers, it is true that x must be 1,3 -1 or -3.
_________________

Keep on asking, and it will be given you;
keep on seeking, and you will find;
keep on knocking, and it will be opened to you.

to my mind, it is easier than it seems...
among the sets of roots the first one is correct...

3 (as a 'free member' of the equation, is not multiplied by x) MUST be DIVISIBLE by the real roots of the equation!

3 does not divide by 0!
Eliminate second choice...
Moreover, 0 is not the root!

I would prefer to use plugging in on these questions