OG Strategy

[Schrute Beets]

What is the best use of the OG books? Do most people break them up into sets of 40 questions and take them timed?

Also - Does anyone know if the free Manhattan GMAT CAT exam includes the two 30 minute AWAs?

Thanks!

[lunarpower]

What is the best use of the OG books? Do most people break them up into sets of 40 questions and take them timed?

bad idea, because the problems are ordered by rough level of difficulty.
this means that if you did the first forty problems of a section on a timer, you would very likely end up with a lot of surplus time, and would get almost all the problems right. on the other hand, if you were to do the last 40 problems of a section on a timer, you would find that a very difficult and disheartening exercise indeed.

if you have a theoretical idea of your possible score (i.e., from taking a gmatprep test, or one of our practice tests), then you could use a linear approximation* to try to guess which 40 problems come closest to your adaptive difficulty level, but that's more trouble than it's worth.

instead, what you should do with o.g. problems is do them in SMALL SETS (say 5-7 problems), with the appropriate time limits. that should be good enough practice for time management. but you are right that you should ALWAYS practice with a stopwatch.

if you have our strategy guides, they break the o.g. problems down into SUBJECT AREAS. for REVIEW purposes, this is better than random blocks of problems, because you can scan all the problems in a given topic together, hunting for commonalities that you wouldn't ever see by doing individual problems.

* i.e., the middle of the section = approx. 500's, etc.

Also - Does anyone know if the free Manhattan GMAT CAT exam includes the two 30 minute AWAs?

yes.

[II]

instead, what you should do with o.g. problems is do them in SMALL SETS (say 5-7 problems), with the appropriate time limits. that should be good enough practice for time management. but you are right that you should ALWAYS practice with a stopwatch.

if you have our strategy guides, they break the o.g. problems down into SUBJECT AREAS. for REVIEW purposes, this is better than random blocks of problems, because you can scan all the problems in a given topic together, hunting for commonalities that you wouldn't ever see by doing individual problems.

* i.e., the middle of the section = approx. 500's, etc.

You can break down the problems in the OG guide (roughly into 4 sections) as follows. Take the PS Quant problems:
PS Questions 1 to 62: Level 300-500
PS Questions 63 to 124: Level 500-600
PS Question 125 to 187: Level 600-700
PS Questions 188 to 249: Level 700+

You can also do this for the other sections: DS, SC, RC, and CR.

I agree that it makes sense to practice problems by subject area (e.g. Divisibility and Primes Number properties), since this gives you a chance to recognize commonalities for this question type.
However, the real test will not present you questions like this ... they will all be random. You have to also learn the skill of doing random questions and being able to classify them quickly and recognising the approach to take in solving them.
Also practicing small sets of question (5-7 questions) is good, but not for building stamina/focus. I suppose this is where the practice tests come in ...
As you may already know, the GMAT is a computer adapter test ... it adapts to your skill level, and aims to present you with problems at the top of your skill level. This will make the test seem difficult for everybody, even the 700+ scorers. So you have to feel comfortable with feeling uncomfortable in the test, and again, this comes from (in my opinion) taking FULL practice tests (completing all sections: AWA, Quant, Verbal, all in one 3.5 hour sitting).

There is no magic one approach in preparing for the GMAT ... it is all based on you as an individual.

Hope this helps, and good luck.

[lunarpower]

excellent advice in the post above.

i wanted to comment on the following, though:

You have to also learn the skill of doing random questions and being able to classify them quickly and recognising the approach to take in solving them.

by 'classifying', what you probably meant - and what you certainly should mean - is classification by CONTENT AREA and STRATEGY. this type of classification is essential because it's absolutely necessary if you are to complete the problems within the allotted time; the better you can classify problems by content type and strategy, the more quickly you'll be able to figure out a strategy (and the less time you'll waste).

you should NEVER, EVER, attempt to approximate the difficulty level of a problem ('300-500', '500-600', etc.) in mid-test. to do so is pointless and distracting; you should just solve each problem one at a time on its own merits.

the rest of the advice in the post is splendid.