GMAT Strategy

AristotlePrep wrote:1. "The IR section will become increasingly important in the coming admission cycle so ask your students to not take it lightly" - Dr Larry Rudner

2. "The OG does not contain the best questions. Remember that the OG contains questions that have been 'retired' from the test. Do not over analyse the OG" - Dr Larry Rudner
Dr. Rudner's comment is extremely crucial in light of the fact that most students and experts alike regard the OG as the be all and end all of GMAT preparation. We had a long discussion with Dr Rudner offline as well and we'll post more of his views and our conclusions on how to use the OG, especially the explanations contained in the OG, in a comprehensive article in the next few days.

3. Experimental questions on the actual GMAT (Dr. Rudner refers to these as 'operational' questions btw) do not come at the beginning or towards the end of the test. They are typically scattered in the middle. Dr Rudner refused to divulge the number of these questions students are likely to see on the GMAT.

4. Dr. Rudner has finally revealed how the penalty is calculated for leaving questions unattempted at the end of a section. We'll cover this is in a separate article subsequently but needless to say, it is a bad idea to leave questions unattempted in the test.

I'm struck by these comments -- call me crazy, but there is nothing newsworthy or valuable in any of these first four points.

1. Yes -- and when admissions officers also share that commentary, the market will agree.
2. Does anyone truly believe that the OG was made up of the 'best' questions? They're just old questions from the test library -- neither good nor bad... just what they are.
3. We shouldn't be worrying about experimental questions in any form.
4. This is like finding out how painful death will be if your parachute doesn't open when skydiving -- we know it's bad... and ideally, we want to make sure that doesn't happen. DON'T LEAVE QUESTIONS BLANK.

brianlange77 wrote:
I'm struck by these comments -- call me crazy, but there is nothing newsworthy or valuable in any of these first four points.

1. Yes -- and when admissions officers also share that commentary, the market will agree.
2. Does anyone truly believe that the OG was made up of the 'best' questions? They're just old questions from the test library -- neither good nor bad... just what they are.
3. We shouldn't be worrying about experimental questions in any form.
4. This is like finding out how painful death will be if your parachute doesn't open when skydiving -- we know it's bad... and ideally, we want to make sure that doesn't happen. DON'T LEAVE QUESTIONS BLANK.

I don't really agree with your take on these things, besides perhaps your comment about the IR section.

2. I certainly think the OG is made up of the best questions that you can find in a book. There are better questions in GMATPrep and GMATFocus (since those questions are more recent, for the most part), but there is no prep company book which contains better questions than the OG. I don't think you can reasonably say OG questions are 'neither good nor bad'. They were good enough to appear on the GMAT, and real GMAT questions are held to exacting standards - the questions simply have to be good.

What I suspect Lawrence Rudner was cautioning against was drawing specific conclusions about the GMAT from the questions in the OG. So, for example, the content balance in the OG may not perfectly reflect the content balance of the actual test. Or it may be the case that the OG does not give a fair reflection of the difficulty level of current GMAT questions, at least at the higher end of the test. But I can't see how anyone could say the questions are bad.

3. The way in which experimental questions are used is something I think should be of concern at least to GMAT experts, if only so experts can correct the misinformation you so often see about how experimental questions are used. Some people develop test-taking strategies based on a misunderstanding of how experimental questions work. I want to have actual facts, and not just speculation, to use to correct those misunderstandings.

4. First, it's never been a mystery how the penalty for unanswered questions is applied. The OG even explains it; it's a simple proportional penalty. So the OP is quite misleading in claiming that some GMAT secret has finally been revealed. Second, it is simply not true that test takers should never leave questions blank and many sources overstate the importance of finishing the test at all costs. There's an official GMAC research report that discusses this very issue. It turns out that sometimes by guessing randomly at the last few questions, you lose more by getting those questions wrong than you would if you hadn't finished the test at all. Because question selection is unpredictable (sometimes the last few questions will be easy, and guessing those incorrectly hurts a test taker a lot), it's impossible to say, on a particular test, whether guessing randomly just to finish is a good idea or not unless you know the exact characteristics of the questions the test taker will guess at. But for higher level test takers, finishing the test is much more important than it is for lower level test takers, who might actually be worse off by guessing in order to finish.

Dramatic misinterpretation of my first reply in this thread.

RE: point 1 -- Didn't say to ignore IR. Simply said -- when we have a clearer signal that ultimate decision makers are more heavily relying upon IR, the market will respond accordingly.

RE: point 2 -- My comment was simply to say that the problems in the OG are retired test problems. That is fact. They are an important tool in prepping for the exam, but they are not an exam. They're retired problems from the test library. No judgment implied, other than perhaps surprise that students were relying upon the OG as a higher power.

Re: point 3 -- Should have provided more context to my comment. If a student is taking the exam and evaluating their likelihood of getting into b-school on a question by question basis because of their own perceived notion of whether or not a question has become harder/easier -- then that student has been poorly guided in preparing for the exam. IMHO, The best way to think about experimental questions is to know that they exist, and then not really worry about it again. Because there is NOTHING that a student can do about it. And any attempt to 'adjust' will not end well. Just as macroeconomic indicators show us the downside risk in trying to "Fight the Fed", IMHO, there's nothing but downside if a student is trying to develop a strategy 'around' experimental questions. Hence -- we (general student test taking population) shouldn't worry about experimental questions.

Regarding point 4 -- I think we open up a dangerous Pandora's Box there. The guidance I've always given to my students, and basically mirrored out of the GMAC report's conclusion is "Answer the questions to the best of your ability within a reasonable amount of time, to allow ample opportunity to give thought to every question."

No one should disagree with that.

https://www.gmac.com/~/media/Files/gmac/ ... ssWhat.pdf

Yes, my understanding is that leaving questions blank (i.e. with the old strategy of devoting more time to the front) is a bad strategy at any reasonable test score level, and so it would be irresponsible to give any advice other than "Don't leave questions blank."

-t