PT59.S2.Q20 - quality control investigator

Comments

• TinaTheLlama Member

  October 2017 edited October 2017 91 karma

  Ah, this one was a killer. I'll paste my analysis below. (its against the rules to post LSAT questions so I'm only including answer choices and my analysis)

  Breakdown:
  1 supplier
  various manufacturing locations
  field inspectors [at/went to?] locations and sent back samples of products
  20% of these samples were found defective
  our contract says the supplier must not let defects account for more than 5% of production
  therefore,
  supplier has breached contract

  A. bases its conclusion on too small a sample of items tested by the laboratory
  NO.
  The sample size is not discussed. It might be a healthy sample size (e.g. M&Ms - easy)

  B. presumes, without providing justification that the field inspectors were just as likely to choose a defective item for testing as they were to choose a non-defective item
  NO.
  If the inspector is not biased, such a presumption would be that 50% are defective and 50% are not defective?

  C. overlooks the possibility that a few of the manufacturing sites are responsible for most of the defective items
  NO.
  If the 20% is comprehensive then maybe all the defects came from one place and skewed the inspection results.
  Suppose there are four sites and three send 25 samples each and one sends 60. Suppose only one out of each of the first three is defective but 24 out of the 60 from the 4th site are defective. Over all that is 20%. Something’s going on at the 4th site. And maybe that’s why they sent more samples, because they know something’s up and thought that the defects would be a smaller percentage of a larger sample. However, remember that the conclusion is that the supplier breached contract: not the manufacturer. Regardless of how the dispersement of defects across manufacturers occurs, its still 20% for the supplier.

  D. overlooks the possibility that the field inspectors tend to choose items for testing that they suspect are defective
  Possibly.
  But then we are not supposed to assume a field inspector is defined as one without bias and there is nothing in the argument to suggest that the field inspectors could have a bias.
  Suppose that the inspectors DO choose items they suspect are defective. Is their suspicion well founded? What if these are some kind of rocket canister or other part of a rocket and the inspector has no knowledge of rockets and is choosing samples based on their weight. This may have nothing to do with whether the sample is actually defective or not.

  E. presumes, without providing justification, that the field inspectors made an equal number of visits to each of the various manufacturing sites of the supplier
  NO.
  Their making an equal number of visits to each site could only more remotely have to do with these results. What about the number of samples? And what if they are visited an unequal number of times but the same number of samples are taken?
• 1 Like

  TheDeterminedC Member

  October 2017 edited October 2017 1025 karma

  I just did this PT a few days ago

  The field inspectors presumably were out to attain an unbiased random sample that would allow for the conclusion to be drawn. The argument's entire conclusion relies on this sample not being biased. But this argument is flawed precisely because they failed to take into account that their sample showing 20% of defective products was actually an accurate representation of the products they were sampling.

  20% of these samples were found defective
  therefore,
  supplier has breached contract

  Read this low resolution summary by @TinaTheLlama and it will really shed light on where the support is coming from. Because of their sample, they conclude X. Answer choice D simply states that they failed to take into account their only support (the sample) might be biased. It makes perfect sense when you piece out the structure.

  Here's an analogous argument:
  The board of directors on education were sent to public high schools to evaluate the amount of extracurricular involvement with children. They interviewed a few students at each of the high schools and found that 40% were involved in extracurriculars. Since this district's public school funding contract requires that 50% of all the students must be involved in extracurricular activities, these public schools have breached their contract.

  Flaw: The argument overlooks the possibility that the board of directors on education tend to choose students for an interview that they suspect were not involved in extracurricular activities.

  The entire validity both these arguments rely solely on these studies and their soundness.
• dcdcdcdcdc Alum Member

  January 2018 edited January 2018 382 karma

  Not sure the earlier response ruling out C is solid.

  Here's the original:

  C. overlooks the possibility that a few of the manufacturing sites are responsible for most of the defective items

  NO.

  If the 20% is comprehensive then maybe all the defects came from one place and skewed the inspection results.
  Suppose there are four sites and three send 25 samples each and one sends 60. Suppose only one out of each of the first three is defective but 24 out of the 60 from the 4th site are defective. Over all that is 20%. Something’s going on at the 4th site. And maybe that’s why they sent more samples, because they know something’s up and thought that the defects would be a smaller percentage of a larger sample. However, remember that the conclusion is that the supplier breached contract: not the manufacturer. Regardless of how the dispersement of defects across manufacturers occurs, its still 20% for the supplier."

  So to start we have 3 factories providing 25 items and a fourth providing 60, for a total of 135 items. One defective items was found at the first three sites and 24 at the last site for a total of 27 defective items. This results in a 20% defect rate for the four factories (27/135 = 0.20).

  I'm not sure why the distinction between supplier and manufacturer is acceptable or relevant. First of all, we don't know whether the supplier is or is not the manufacturer. However, even if we assume them to be separate, there is still the issue that we could have the above sample of 20% for these four sites, yet have many OTHER sites that the supplier uses that result in an aggregate defect rate that is below 5%. In this case, the supplier is NOT providing a 20% defect rate, which is exactly why C is tempting. It is not true that, as given above "regardless of how the disbursement of defects across manufacturers occurs, it's still 20% for the supplier."

  Why is the issue I've highlighted here not a flaw in the argument? Couldn't it be that the sample that supports the claim doesn't show an overall defect rate, but rather one input for the overall defect rate?