#17 - A 24-year study of 1,500 adults showed that those

[Robert Carroll]

desi,

I think what you say is true, but it looks to me like that's exactly what the explanation in the first post is getting at. So you're definitely on the right track!

Robert Carroll

[desiboy96]

Hello Robert, thank you so much. LR has me very paranoid XD.

[lsatstudent99966]

Hi there,

Regarding answer choice (E), does it matter that we don't know the proportion of smokers and non-smokers in the 12-year study?

I was a little hesitant to eliminate (E) because I think that without knowing the proportion of smokers and non-smokers in the 12-year study, we wouldn't be able to confirm that smoking leads to the discrepancy between the result of the 12-year study and the 24-year study (maybe in the 12-year study, the proportion of subjects with a high intake of beta-carotene supplements who smoke is also much smaller than the proportion of subjects with a low intake of beta-carotene supplements who smoke).

Thank you so much for reading this!

[Jeff Wren]

Hi lsatstudent,

It's true that we would need to know the proportion of smokers to non-smokers in the 12-year study to really feel completely confident that the discrepancy was due to this factor. For example, in the extremely unlikely situation where the smoker/non-smoker percentages were identical to the 24-year study, then the discrepancy would remain a mystery.

However, knowing that the 24-year study has an obvious flaw (not accounting for other important variables such as differences in overall health, smoking, etc. of the groups of participants is a huge flaw) that would likely overstate the benefits of beta-carotene would help to resolve this discrepancy. While we don't know whether the 12-year study commits the same flaw, there is no reason to assume that it does.

Imagine a hypothetical in which we have two different studies that had very different results. One study (the one with the higher numbers) we know was flawed in a way that would lead to higher numbers. We don't know for certain whether the other study was flawed, but there's no indication given that it was. On its face, the fact that the study with the higher numbers was flawed in a way that would make the numbers higher than they otherwise should be would help explain the different results.

[lsatstudent99966]

Hi lsatstudent,

It's true that we would need to know the proportion of smokers to non-smokers in the 12-year study to really feel completely confident that the discrepancy was due to this factor. For example, in the extremely unlikely situation where the smoker/non-smoker percentages were identical to the 24-year study, then the discrepancy would remain a mystery.

However, knowing that the 24-year study has an obvious flaw (not accounting for other important variables such as differences in overall health, smoking, etc. of the groups of participants is a huge flaw) that would likely overstate the benefits of beta-carotene would help to resolve this discrepancy. While we don't know whether the 12-year study commits the same flaw, there is no reason to assume that it does.

Imagine a hypothetical in which we have two different studies that had very different results. One study (the one with the higher numbers) we know was flawed in a way that would lead to higher numbers. We don't know for certain whether the other study was flawed, but there's no indication given that it was. On its face, the fact that the study with the higher numbers was flawed in a way that would make the numbers higher than they otherwise should be would help explain the different results.

Thank you so much, Jeff. This is very clear and extremely persuasive!