#24 - A recent poll of a large number of households found

[Administrator]

Please post your questions below!

[BostonLawGuy]

I got this right on test day, but was still confused.

The stimulus describes the % of households with cats that has a person with a university degree and compares with it with a less % for dogs. The conclusion maintains the % ( people with degrees are MORE LIKELY to live with a cat) so I did not think that this was a numbers to percentage flaw.

It seems logical to conclude that people are 9% more likely to live with cats than dogs. Maybe I am not reading this correctly?

[Brook Miscoski]

Boston,

My reaction to the stimulus is that you can't flip percent relationships like that without knowing how the group sizes compare to each other. The stimulus might show that cat owners are more likely to have a degree when compared to dog owners. The claim that degree holders are more likely to have cats than dogs is an erroneous reversal of that relationship, although it is because of number concepts instead of conditional concepts.

For example, we will use one person to a household to stay sane. Imagine that you have 100 people with cats, and 1000 people with dogs. That is 47 cat people degrees, and 380 dog people degrees. Because there are far more dog people degrees than cat people degrees, having a degree will be associated with being more likely to have a dog than to have a cat.

That is answer choice (B), and it is also one of the standard ways of attacking a stimulus with reasoning that is based on percentages.

[BostonLawGuy]

Thanks for that excellent explanation. I should not have fell for the trap answer, especially having studied the PowerScore course which outlines this flaw.

I am not good with stats, and somehow I have to ignore the fact that even with a much greater number of dog owners with degrees, it wouldn't change the fact that cat owners are still 9% morel likely to have degrees.

[lalalala]

Hi,

I'm still a bit confused on why answer choice E isn't correct? I get how B is correct, but is it because they are not making a causal claim in the conclusion?

Any clarification would be super helpful.

Thanks so much!

[Luke Haqq]

Hi lalala!

You are right to note that the stimulus isn't clearly making a causal argument. That should make answer choice (E) fall out of contention. A stimulus would need to be much more directly making some assumption about correlation implying causality, which this stimulus is missing. If we had a stimulus with a conclusion such as "owning a cat incentivizes people to get a degree"--that sort of claim is more clearly a causal one.

If you're unsure of whether you're dealing with cause and effect reasoning, you might try weakening (or strengthening) it--for example, by coming up with an alternate cause, or by showing that the causal relationship is actually reversed. If you're unable to weaken (or strengthen) the reasoning in these ways, then that can be a good reason to conclude that you may not be dealing with causal reasoning.

[sqmusgrave]

Hi! I see clearly why B is correct, but before I read that I was tempted by A, partly because I don't entirely understand the implications of A but it seems like the subject matter is on point. Could someone explain to me what A means and why it isn't an issue?

Could someone explain a scenario where A would be a legitimate charge against the argument? Thanks!

[Dana D]

Hey SQ,

One good way to test whether (A) is an actual flaw is to see if the addition of that consideration actually fixes the flaw of the argument. We know the flaw is the fact that we're comparing two percentages here but we can't actually tell what they represent, so using that comparison to say those with university degrees are more likely to live in a household with a cat rather than a dog is flawed.

If we added into the stimulus a sentence which addressed the possibility of a household having both a cat and dog, would this eliminate the flaw?

For example, lets say 47% of households with a cat had a degree, 38% with dogs had a degree, and none of the households had both. Or we could say 90% of the households had both. Either way - does this help you draw the conclusion that the author did and say that a cat household is more likely to have a degree? No. If anything, we're left more confused about the impact of cats or dogs on degree ownership.