Retention & Manhattan GMAT Roadmap book question

[rkafc81]

Hi,

I'm really confused as to what study strategy to use after analysing and reviewing a problematic problem in-depth...

I've realised this week that one of big problems is retention... I'm really stuck as to how to effectively retain what I've learnt from reviewing a question, some time after having learnt and understood it... do I redo the question again, or put the whole thing on a flashcard, like the MGMAT GMAT Roadmap book says to do?

i did a practice exam recently and realised I've previously done and reviewed, and was comfortable with at the time, a lot of similar question to the ones I got wrong. I've simply just forgotten one or all of the following:

(a) key takeaways
(b) techniques and backup strategies
(c) how to do those questions

In the MGMAT GMAT Roadmap book, in the flashcards section, it advises making flashcards out of problems that have given you trouble or that you have learned something from, by:

(a) copying the problem (front)
(b) breaking it down and listing different solution methods and
takeaways (back)

is putting a question on a flashcard like that enough to remember, on similar questions:

(a) techniques to solve
(b) content learned
(c) takeaways gleaned from the problem?

my question is, after analysing them properly, should I just put these sorts of problems on a list of problems to review and have a 'review day' every 4-5 days where I redo them again, and if I have a problem still, put on a list to review again?

And at the same time, any key points or takeaways I get from the problem, put onto a "quick", one-liner, flash card? ... but, the problem I find with these sorts of 'one-liner' flashcards though is that they don't provide context[i][/i] to the rule/takeaway/snippet of information that I put on the flashcard. For example, it's all fine and dandy to make a flashcard that says 'the average of a set with an odd number of terms is an integer' but then that fact becomes pretty useless to have learnt over and over using flashcards when a question pops up that indirectly tests this fact...

I think redoing problems again in a database of problems (I generated heaps and heaps of blank word documents, linked them into my Excel error log, and when I mess up a problem or don't get it, then I copy and paste the problem into the word doc, and have another 'cloned' word doc which will contain the answer and solution methods... my idea is to then use Excel to generate random sets of problems using previously messsed up ones, and do these problems until I'm comfortable with them, at which point I will 'close them off' and not look at them again until 1-week before the exam...) is probably quite useful for me, but is there a more effective method ? reviewing a flashcard would be quicker than redoing a problem.., but redoing a problem would:

(a) remind me of what i did wrong
(b) help me get comfortable with the concepts
(c) help me to learn good techniques to solve it and correct bad habits

or is the process of making these sorts of in-depth flashcards enough to remember everything in the problem (including techniques, takeaways etc.) and develop recognition for future, and similar, problems? they seem time-consuming to make.

So in summary, the problem is that after I review a question in-depth, extract takeaways from it and make flashcards from those takeaways, I will likely forget how to do that problem down the track...

I've attached an example of one I made...

thanks!

[rkafc81]

... seems i can't attach the example flashcard...

[StaceyKoprince]

Good question. Different people learn differently. Flash cards work for a lot of people, but they might not work for you / everyone. You can also try something else with flash cards; what I'm about to describe works better for me, personally.

I don't like the "whole problem on a flashcard" thing because then I just feel like I'm trying to memorize the whole problem and (a) that's not that useful, since I won't see that problem on the test, and (b) it's not like I'm actually going to be able to memorize all of these problems anyway!

So what I like to do instead is figure out more specifically: what do I need to learn / do better at for this particular problem?

For example, let's say that the question stem said "y^2 < y" and I spent a minute figuring out that if y^2 < y, then that just means 0 < y < 1, because only fractions between 0 and 1 can make that initial inequality true.

And let's also say that, once I figured that out, I was fine with the rest of the problem - nothing else I have to learn / memorize / practice / whatever.

So what do I do? I make a flashcard that says, on one side, "When I see y^2 < y" and, on the other side, "then I'll write down 0 < y < 1."

That's really the heart of what I need to learn here: I need to know the GMAT code that y^2 < y is just a disguise for a certain piece of information. And so that's what I target in my notes or flash cards. Then I've got something much more specific that I can drill - not an entire problem that I won't see again, but a very specific takeaway that could apply to any problem.

For example, it's all fine and dandy to make a flashcard that says 'the average of a set with an odd number of terms is an integer' but then that fact becomes pretty useless

Totally agree! That's why I also like the "When I see ______" and "I'll think / do _______" format for flash cards. The first side should include actual language or snippets or examples of what I might see on an actual problem.

The flash card becomes almost like a translation tool. Eventually, when a certain translation is so ingrained that you don't miss it any longer, you can "retire" that flash card.

And, yes, I agree that periodically it's beneficial to redo problems. Also, try explaining problems aloud, either to a friend with whom you're studying or just aloud to yourself, but as though you're trying to teach someone else. If you can explain both (a) what to do, and (b) how you know what to do (this part's critical!), then you know that you can probably replicate that thinking in future. If you can't, then you know you've got to dig in more. (And, often, that's about digging in on HOW to know what to do - what are the clues that tell you what step to take or what to try? They're there, I promise - but can you read them?)

Finally, are you noticing any patterns re: what you tend to forget? Certain content areas, certain types of manipulations or shortcuts or ways of thinking, etc? If so, then perhaps you didn't dig deep enough in terms of understanding WHY the problems work the way they do and HOW you know to take each next step - so maybe that's where you need to start again. :)

[rkafc81]

thanks for your quick reply Stacey!

ok, so here is an example of doing a flashcard in the way you mentioned --> I did a rates problem and I messed it up because there was a units conversion required in it that I wasn't comfortable with... so I came up with this flashcard:

Q. when I see a rate question where the given units are diff. from the desired units, I should ____ ?

A. either convert into desired units FIRST, or perform the conversion AFTER setting up the calculation


so hopefully next time I see a question like that, I'll remember this and not get it wrong.


I recently got a (Data Sufficiency) question wrong because that question asked about the cost per unit of something, and I didn't think that the required information to solve that was to find out the total cost and the # of units. So my flashcard for that was:

Q. Cost per unit = ?
A. total cost / total # units

... is this the type of thing you had in mind in your reply to my post?

in response to your question - the types of things I normally forget are techniques and/or strategies which are specific to unfamiliar problems or learning (and then knowing) what to do when I see certain problem types that I'm not comfortable with... I'll analyse the problem like you've suggested in the "how to analyse a practice problem" article, learn the technique, try it out on the problem, think I've mastered it, and say to myself "ah ok I get it now, this makes sense", and then may not see a problem like that for a while. But the next time I see a similar one some time down the track, I'll get it wrong again because I've forgotten the required technique(s)/strategies to do it or perhaps I still wasn't 100% comfortable with that problem type and techniques required to solve it... This happens a lot :( and this is why I'm wondering about redoing questions over and over again that I wasn't comfortable with/went overtime/got wrong.


thanks again

[StaceyKoprince]

Q. when I see a rate question where the given units are diff. from the desired units, I should ____ ?

A. either convert into desired units FIRST, or perform the conversion AFTER setting up the calculation

Good! Push it further. how are you going to decide whether to do it FIRST or AFTER? Are there certain times when it's better to do it one way vs. the other? If so, what are the ways / clues?

I'm not telling you that one way is better than another because it does depend on your own strengths and weaknesses - you have to figure that out for yourself. eg, if you're more comfortable with a certain measure, and the problem starts out with the one that you're NOT comfortable with, that'd be a clue to convert first. Alternatively, if converting first gives really ugly numbers, that'd be a clue to convert later.

Those are the kinds of things that you want to think about beforehand so that you already know how you want to make the decision. Then you don't have to think as much in the moment. :)

So, yes, you're on the right track, just push it further. I think this will help address the difficulties you described in a couple of ways:
(1) you'll be forcing yourself to think through "what are the clues that tell me to think a certain way / solve a certain way / go down a certain path" so that you'll be more likely to recognize those same clues in future
(2) you'll be setting yourself up to quiz yourself periodically so that you don't forget this stuff as much when it's been a long time between problems; you might even add a reference to the problem # on the back of the flashcard (as in, see example here: OG PS #134) and you can refer back to that problem or try it again in future as you see fit

And, yes, I also think it's a good idea to redo problems after a while, just to see how things go. Don't expect to get them 100% right. Your brain isn't a computer. :) You just want to see improvement, that's all.

[rkafc81]

thanks Stacey, that's great! (thanks for the fast reply too)

so, another question directly related to reviewing problems: what do I do to learn the ins and outs of a troublesome problem? often I find I'll move on after spending sometimes up to 30 minutes reviewing a problem, and I still don't 100% get it... so basically, how do I really learn a problem, especially when it's troublesome for me?



thanks!

[StaceyKoprince]

When something is giving you that much trouble, look for outside resources to help. Search for it on forums to see what others have said about. Ask a question yourself (make sure to follow forum guidelines when posting). If it's an OG problem, look in our OG Archer program (if you have access to it) - we have our own explanations, either written or video, for many of the problems.

If you know others who are studying, get together once or twice a week and each bring 4 or 5 problems that have been driving you crazy. Share and brainstorm better approaches. (If you can, find someone who is stronger in your weakness areas and vice versa.)

You can take this all way up to hiring a private tutor - though of course that's very expensive. Depends what kind of goal you're driving toward and how much time you have to do it. Using the forums is a slower process, but it's free. :)

[rkafc81]

thank you Stacey, you've been very helpful.

I think that's probably all the questions I have for now, I just have to convert the takeaways from this thread into actions!

[rkafc81]

actually Stacey I have one more question on this topic...

so, this exam is all about really knowing the underlying concepts , and practicing questions helps you to learn them (after already learning the content in general). My question is - as it is sometimes hard to find lots of questions to practice for a particular topic in order to learn those concepts, how can I best cover all the concepts I need to know on a certain topic by just doing a limited set of questions (rather than dozens and dozens of questions of that type).

thanks!

[StaceyKoprince]

Actually, the test is really testing you on how you think. The concepts themselves are just a tool for that.

So, in fact, it's very inefficient to DO dozens and dozens of problems, vs. doing a few and really analyzing those problems to death. :)

Read the "how to learn" section of this:
http://www.manhattangmat.com/blog/index ... an-part-1/

Follow the links in that section - they'll show you how to analyze / dig in and really learn from these questions.