Q4 - Seventy-five percent of dermatologists

[Nina]

I chose B during my timed practice, but i remember i was hesitating between B and C, and during review i still cannot eliminate C quite comfortably.

What if this survey was conducted among dermatologists that were less qualified to evaluate skin cream? would their recommendation become less reliable?

Thanks a lot!

[tommywallach]

Hey Nina,

Happy to look at this one for you. But let's take the whole shebang:

Conclusion: Use Dermactin!
Premise: 75% of dermatologists prefer Dermactin to other creams.

(A) brings up other types of doctors, which doesn't make a whole lot of sense, as dermatologists are skin doctors. Anyway, it wouldn't make the sample unrepresentative.

(B) CORRECT. 75% is a high number, but not if they only talked to 4 dermatologists!

(C) might be a good error if it said "ignores the possibility that some dermatologists are less qualified than others to evaluate skin cream." I think this is what you think it said, Nina. But actually, it accuses the argument of presuming that some dermatologists are better than others. But the argument doesn't do that. It weighs all dermatologist opinions equally.

(D) does not describe an error in the argument. The opinions of consumers are not referenced, only the opinions of dermatologists.

(E) is something the argument does. The conclusion only relates to those who "need a skin cream".

Hope that helps!

-t

[magnusgan]

Hey guys, really had doubts over (B) because I didn't think that a small sample size could be a valid flaw.

Taking the example above where only a handful of dermatologists (4) were surveyed, if the 4 were like the most influential/most respected/most up-to-date/blah blah dermatologists on Earth, wouldn't a small sample be sufficient?

I was looking for an answer which said that the sample was unrepresentative, so I picked (A) even though it wasn't great. Just thought it was better than (B).

[sujin91]

Hello,

So can we safely assume that every time we encounter percentage statement in an argument, we need to know the entire population (in number) ? I feel like I have encountered percentages in an argument in the past (in LR) without the entire population number, and it was not an issue.

When we see a percentage of let's say people agreeing on something, we need to see the number of people who were asked this question in order for the argument to be valid?

Please let me know!

Thanks,

Sujin

P.S i am taking the JUNE LSAT (next monday). I'd appreciate it if you could get back to me asap.

[raziel]

Most of the time when there is a conclusion drawn from a survey you can expect that it has multiple errors, not just the one mentioned in (B). Even for this argument there are multiple errors.

We know that a claim made in the argument is that "it is the best skin cream on the market". The evidence? 75% of dermatologists SURVEYED liked it best. I think this is what makes this argument hard since it is hard to tell what exactly its conclusion is. But this is typical of other passages with Advertisement authors. But once you have the evidence and the conclusion, you can readily spot multiple errors.

In this case, some errors can be: (1) Assuming that what dermatologist say they like best is what is best for the consumer (2) Assuming that the dermatologists prefer the cream because it has desirable qualities (maybe they get a bigger cut from the company) and most obviously (3) assuming that the sample is representative of all dermatologists (notice it just says "survey", it doesn't tell us whether it was a good survey at all).

So the assumption is that the survey is a good indicator of the qualities of the cream. Answer choice (B) points out one of the ways in which the survey could be bad, but it could also have pointed out other flaws in the survey like saying "overlooks the possibility that the survey is not representative". This directly attacks the assumption that the author made in the passage. Because it ruins the any positive conclusion made about the cream, you know that this is your answer.

Answer choice (C) attributes an assumption to the passage. If you try negating it, "it is not true that...some less qualified" or "some are not less qualified...". It doesn't really ruin the conclusion, so you are better off picking (B)

[Alvanith]

Maybe this is a late reply to the last post, but I do not think the conclusion of the argument is Dermactin is the best. Although the stimulus says the advertiser tried to ensure the Dermactin to be the best product, but the conclusion is just to recommend Dermactin.

I think the core is:
P: 75% of dermatologists surveyed prefer Dermactin to other brands.
C: use Dermactin.

Thoughts:
1. looks like the advertiser is assuming 75% approval from the dermatologists surveyed is good enough to recommend the product. The problem is maybe the other brands all get more than 75% approval from dermatologists! In that case, to recommend the product because of the 75% approval seems inadequate.
2. surveyed? This could be problematic. Are the people surveyed representative enough to warrant a recommendation? We don't know.

Answer choice B is attacking the representativeness of the people surveyed.

P.S. I am not sure whether the advertiser is assuming the dermatologists surveyed prefer Dermactin for its desirable qualities. Maybe the advertiser is implying to the consumers that the dermatologists' preference is based on the quality of the product, but the advertiser may not necessarily believe his recommendation is based on the dermatologists' preference for the quality. Does implying necessarily mean assuming? The advertiser here is just recommending the product because of the 75% approval from demonologists surveyed. Am I overthinking here? Somebody please help me! Thanks in advance:)

[christine.defenbaugh]

Some really excellent thoughts here from a number of people!

First, I'd like to clear up the issue of 'small sample size' vs 'representativeness' that both magnusgan and sujin91 both raised.

A small sample can, theoretically, be perfectly valid - so long as it is representative. The problem, realistically, is that the smaller a sample size becomes, the less likely it is to be truly representative at all. If we somehow knew for sure that 4 particular dermatologists were a perfect reflection of the dermatologist population at large, then surveying exactly 4 would be acceptable - but how in the world would we ever know that?

So, the practical result of all of this is that as sample sizes go toward the very small, the reliability becomes more and more tenuous.

In this case, some errors can be: (1) Assuming that what dermatologist say they like best is what is best for the consumer (2) Assuming that the dermatologists prefer the cream because it has desirable qualities (maybe they get a bigger cut from the company) and most obviously (3) assuming that the sample is representative of all dermatologists (notice it just says "survey", it doesn't tell us whether it was a good survey at all).

Raziel's got a killer breakdown of some of the flaws running around in this argument.

However, I have to agree with Alvanith that the conclusion here is not that Dermactin is the best, but rather that "if you need a skin cream, use Dermactin". The difference there is subtle, but significant.

P.S. I am not sure whether the advertiser is assuming the dermatologists surveyed prefer Dermactin for its desirable qualities. Maybe the advertiser is implying to the consumers that the dermatologists' preference is based on the quality of the product, but the advertiser may not necessarily believe his recommendation is based on the dermatologists' preference for the quality. Does implying necessarily mean assuming? The advertiser here is just recommending the product because of the 75% approval from demonologists surveyed. Am I overthinking here? Somebody please help me! Thanks in advance:)

I understand what you're saying here, because we're pretty skeptical of this smarmy advertising guy as a general matter. In fact, if we met this guy in real life, we'ld probably be pretty skeptical of whether he even believes his own conclusion.

But we can't dismiss an assumption on the LSAT because we think the author is just lying about the conclusion - that way lies madness. We have to assess what the author would have to be assuming in order to claim their conclusion in good faith.

So, let's try negating this concept - if the dermatologists "preferred" Dermactin based on something other than its quality or functionality (say, based on it's fun sounding name!), then this would destroy the potential usefulness of this evidence in supporting the conclusion that you should use Dermactin! This totally destroys the argument being put forward.

In other words, what you are characterizing as 'implying' is really what you see as the author (advertiser) hoping his *listeners* assume from this statements. In this scenario, the advertiser doesn't believe his own conclusion - but he does hope his listeners will.

All in all it's less mind-bending to just assume that authors are making arguments in good faith, and that when they are wrong they are flawed rather than lying. 'Never attribute to malice that which is adequately explained by stupidity' is a good mantra for the LSAT.

Let me know if that helps a bit!