#20 - Quality control investigator: Upon testing samples

[Emily Haney-Caron]

Hi Basia,

A prephrase for this question might be, "There is some reason that the 20% defective in the samples is not representative of all of the items manufactured." D provides that reason - if the field inspectors only send in items they think might be defective, then it makes sense that far more than 5% of the items they send in actually are defective!

C, on the other hand, is irrelevant. It doesn't matter much which site is responsible for the defective items; the supplier can't have more than 5% defective regardless of site.

Hope that helps!

[eober]

Hi,

Why did we eliminate answer choice B? Doesn't B say something very similar to answer choice D? What makes it different from D and makes it incorrect?

Thanks!

[Adam Tyson]

Actually, B is the opposite of the correct answer, D. D is saying that inspectors are substantially MORE likely to select a defective item, while B says the likelihood is the same for either. B eliminates the possibility of bias in the selection process, and strengthens the argument, while A introduces that possibility and exposes the flaw.

[nutcracker]

Hi Dave,

After reading through the previous posts, I would like to add a follow-up question concerning answer choice (B).

The way it is phrased, "the field inspectors were just as likely to choose a defective item for testing as they were to choose a non defective item", gave me a strong (although faulty) impression that it is describing a random, unbiased sample selection, which would make it a correct answer.

I would like to know how one should go about expressing that random, unbiased sample selection idea using this "just as likely to pick A as to pick B" format, since I believe that I have seen something like this before, but had a hard time coming up with an answer. Thank you very much!

[Eric Ockert]

Hey nutcracker!

Perhaps something like:

"presumes, without providing justification, that the field inspectors were not significantly more likely to choose a defective item for testing than they would through a random sampling of items"

or

"presumes, without providing justification, that the field inspectors were just as likely to choose a defective item as would be expected through a random sampling of items"

And that is the issue. The likelihood the inspectors chose a defective sample matters, but only as it compares to what you would expect from a random sample of items, not as it compares to the likelihood of choosing a nondefective item.

Hope that helps!

[nutcracker]

Thank you! That's the thing I was racking my brain for

[akanshalsat]

For this one my question is - that if D were correct------> even if the inspectors were choosing items for testing that they suspect are defective --> and those items were defective ---> then who cares? they did their job and defective items were found (a lot of them); if they picked out items that were NOT defective, that wouldn't change the fact that many items made WERE defective (they just werent picked) so i'm confused how D is correct?

[James Finch]

Hi akanshalat,

This is an interesting question, because it's asking us to come up with a reason that the company may not have violated the contract. Because the conclusion in the stimulus is certain--they definitely violated the contract--what's required here is only to make it possible that they didn't violate the contract, not prove definitively that they didn't. Answer choice (D) does this by allowing for the possibility that the sampling was biased, testing only those products that looked defective, which means that it wouldn't be representative and may not accurately reflect the true defect rate.

Hope this clears things up!

[silent7706]

Hi,

While the question stem states this is a flaw question, i feel that we are better off approaching it with a mindset of weaken question. It occurs to me that for flaw questions after #15, their "flaw" are generally much more subtle. I feel that we are better off focus on answers that weaken its conclusion instead of identifying any obvious flaws in the stimulus. I wonder what's your thoughts on this?

Thanks in advance.

[James Finch]

Hi Silent,

By definition, a Weaken question is inherently flawed; the flaw is what allows us to weaken it. There's a great deal of overlap logically between the two types of questions, and many stimuli are written in such a way that they could be either type of question, and many Flaw (and Weaken) questions are based on faulty assumptions, as this one is. Ideally after studying the PowerScore approach to flaw questions, you'll be equipped to handle those stimuli that have logical gaps just as easily as those that commit a clearly recognizable fallacy.

Here we only have a survey as evidence for a conclusion, so we need to attack the survey in a logically valid manner. Since we don't know the sample size, (A) would be out; what we really need to know is whether the sample is representative or not. The stimulus assumes that it is, but we don't know that to be true. Only (D) gets to this idea, making it correct.

Hope this helps!