Parallel the reasoning--strategy? Forum

[mmk33]

Some parallel the reasoning questions (a lot of them) are easy to diagram or they are easy to conceptualize and compare to the answer choices. I'm reviewing two questions from PT #60 (section 1 17 & 21) and I got both of them wrong because I couldn't diagram them effectively. And between the two questions I probably used up five minutes of test time. Looking back at them it seems like boxing qualifying words in the stem ("not" "some" "all" "most") would be a fast and efficient way to knock out wrong answer choices, and potentially you could get them correct simply by eliminating answers that lack this structure. Has anyone tried this method? I am not advocating doing this on all parallel the reasoning, but for the very difficult ones unless you are an incredibly efficient diagrammer isn't this a reasonable option (almost akin to answering the first question in each LG section by elimination).

[glucose101]

While diagramming aids when there's a clear conditional relationship, on the others, especially the flaws, you should really be holistically pinpointing the flaw, and then picking the corresponding answer.

[ahnhub]

What usually works for me is to generalize the stimulus in a linear way, and then compare the answers in a linear way. This requires you to reorganize the key facts of the answers in the same time sequence as the stimulus, though. The advantage to generalizing the original point is that you don't mix up facts in your head--you always have an abstraction to compare to.

[bp shinners]

I am not advocating doing this on all parallel the reasoning, but for the very difficult ones unless you are an incredibly efficient diagrammer isn't this a reasonable option (almost akin to answering the first question in each LG section by elimination).

It's a perfectly acceptable shortcut, but like all shortcuts won't guarantee you a correct answer 100% of the time. However, I tell my students that, should the stimulus have 'most', 'all', 'some' statements, or others with definable logical force, throw out the answers that don't follow that pattern and look at the others. You can usually get it down to 1 or 2 with just that, and then it's easier to compare just those two.

You should also be breaking them into parallel and parallel flaw questions. The parallel questions need the arguments to be parallel in nearly every way - quantification, logical force, number of premises, etc... The parallel flaw questions just need to have the same flaw, regardless of number of premises.

[Want_My_Life_Back]

The first step I take in attacking these questions is identifying the conclusion and then eliminating all ACs that don't come to the same conclusion. Usually the structure/strength of 3 of ACs allow them to be easily eliminated since if the conclusions aren't parallel there is no way that the arguments taken as a whole can. This allows you to save time normally spent trying to follow the logic behind each AC. Once you've narrowed it down to two then take a more holistic approach.

[NiccoloA]

Guess and comeback if I have time.


I hate the last parallel reasoning question. The language is always so convoluted and when you have only a few minutes left... Yeah... I'm not dealing with that crap, honestly.

[Manhattan LSAT Noah]

http://www.top-law-schools.com/forums/v ... 6&t=165741

[lrslayer]

i think i stumbled onto something today and i'm sure its been said but wanted to share.
think of a parallel question in the same way you would a global accounting logic games question (you know, the first easy knock out question)
when i do it like that, I simply go through the and take first the conclusion, if the ac conclusion is any different i knock it out.
then, take each premise sentence and do the same, one at a time, knock out ac's.
i actually find that when doing this method its very fast and effective. hope that helps!

[Geetar Man]

I am not advocating doing this on all parallel the reasoning, but for the very difficult ones unless you are an incredibly efficient diagrammer isn't this a reasonable option (almost akin to answering the first question in each LG section by elimination).

It's a perfectly acceptable shortcut, but like all shortcuts won't guarantee you a correct answer 100% of the time. However, I tell my students that, should the stimulus have 'most', 'all', 'some' statements, or others with definable logical force, throw out the answers that don't follow that pattern and look at the others. You can usually get it down to 1 or 2 with just that, and then it's easier to compare just those two.

You should also be breaking them into parallel and parallel flaw questions. The parallel questions need the arguments to be parallel in nearly every way - quantification, logical force, number of premises, etc... The parallel flaw questions just need to have the same flaw, regardless of number of premises.

Very good recommendation!

[suspicious android]

You should also be breaking them into parallel and parallel flaw questions. The parallel questions need the arguments to be parallel in nearly every way - quantification, logical force, number of premises, etc... The parallel flaw questions just need to have the same flaw, regardless of number of premises.

This is really good advice.

Also keep in mind that a lot of answers (usually 2-3) can be eliminated because they don't have parallel validity. That is, if they ask for a parallel flaw, you can eliminate any answer choice with a valid argument.

[bp shinners]

Also keep in mind that a lot of answers (usually 2-3) can be eliminated because they don't have parallel validity. That is, if they ask for a parallel flaw, you can eliminate any answer choice with a valid argument.

But not vice versa - we've been seeing more and more parallel questions that are flawed (devious, I say!). For a parallel question, make sure you determine validity, because it might be invalid.

However, a Parallel question with a flaw doesn't become a Parallel Flaw question. The answer still has to contain that flaw, but it also has to parallel the rest of the question.

In other words, Parallel questions have a higher burden for a correct answer (all elements must be parallel, including the validity); Parallel Flaw questions have a lower burden for a correct answer (just the flaw has to match up; the other elements of the stimulus can be different). For a Parallel Flaw, however, note that for many flaws, the elements can't differ radically, or they wouldn't be committing the same fallacy.

[suspicious android]

But not vice versa - we've been seeing more and more parallel questions that are flawed (devious, I say!). For a parallel question, make sure you determine validity, because it might be invalid.

Definitely. It used to be everyone could count on using the question stem to determine the validity of the arguments in parallel questions, it makes sense that LSAC stopped writing them like that.

However, a Parallel question with a flaw doesn't become a Parallel Flaw question. The answer still has to contain that flaw, but it also has to parallel the rest of the question.

I haven't noticed this difference, interesting.