Parallel reasoning/flaw

[wj097]

On Parallel reasoning/flaw questions, I am spending so much time that it almost worth 2 questions. Now I am thinking there gotta be a standard process that eliminates the most off ones really quickly and just focus on one or two. I have come across some prep courses say just go right to the conclusion and match it up and if you find discrepancies like the conclusion in the stimulus say something "cannot" or "will not", but the answer choice says something "must all be" then without even evaluating the flaw just rule it out...and I am very uneasy with this approach. How would MLSAT shed light on this.

Thx

[tommywallach]

Hey WJ,

You could read our Logic Games book to get a more detailed answer, but what you've read is true. Parallel reasoning questions have to match up in ALL major ways. In other words, if the conclusion says "Some monkeys are friendly...", then the parallel argument's conclusion cannot say "All fish are cool" or "No children are smelly." The "some" has to match up.

Basically, you can look for similarities from conclusion to conclusion, premise to premise, and link to link (Link meaning the connection BETWEEN the premise and the conclusion). The link issue is the hardest one, but just the conclusion and the premise matching should get you down to one or two answer choices.

In short, the key is to get a REALLY solid sense of the constituent parts of the original argument, then go hunting. The one thing to be careful of is that the architecture doesn't have to be the same. In other words, your main argument may be structured like this:

Premise. Premise. Conclusion.

But the correct answer may be structured:

Conclusion. Premise. Premise.

The ORDER of the pieces doesn't have to be the same, only the actual CONTENT.

Hope that helps!

-t