Pharmacist: A large study of people aged 65-81 and suffering from insomnia showed that most of insomnia's symptoms ar...

Andrew on April 15 at 03:46PM

Control group vs unrepresentative sample

This AC made me nervous because the problem with the conclusion is that it needs a control group in a younger cohort. There's no indication that sample is unrepresentative of the cohort its studying which is what the conclusion is approximately about. I can see how you can call the sample unrepresentative of the conclusion if the conclusion extends beyond the studied cohort but a reasonable person could also confine the conclusion to that cohort. How can I be more confident in the correct AC?

Replies
Create a free account to read and take part in forum discussions.

Already have an account? log in

Emil on April 16 at 09:07PM

Hi, I think there's an even greater issue, here. It's not only that we lack a control group, because having a control group would still fail to solve for a major flaw. The argument is trying to prove that within and individual's lifetime, they would tend to produce less melatonin as they get older.

Observing young people and old people at the same time would not prove this, since it would fail to control for possible confounding factors (e.g., WWII led to less melatonin, smartphones lead to less melatonin) that impact one age cohort but not the other. To prove a decrease over time you would actually need to look at a panel of the same individuals across time periods.

Note, of course, that the right answer doesn't reflect this flaw either. Ultimately there are usually multiple ways to articulate the flaw, and frequently more than one flaw, as both the flaw you pointed out here and the one I pointed out are issues. As long as an AC hits at a flaw, even if it's not the one you anticipated, it is correct.

Moreover, my favorite way to be confident in an AC is to be confident the other 4 are wrong. A-D are just not what the author did, and when I have 4 ACs that are clearly wrong, and one that is weird, unexpected, but I could make a case for it being correct, I can feel confident picking it because I know I hate the other 4

Andrew on April 17 at 02:11AM

Thank you Emil