Subscription pricing
I came across a question that said something along the lines of "In a survey 60% of participants said yes and 40% said no" and the conclusion was "This shows more people believe yes than no" (Granted the question was much more nuanced and another flaw was present but thats besides the point of my question).
One of the ACs said "They draw a conclusion about the population in general based on only a sample of the population"
Even though I got this question correct because I saw the other flaw in the argument, the AC above was wrong which confused me because the explanation why it was wrong specifically said "there is nothing inherently flawed about drawing a conclusion from a sample. What would be flawed is relying on an unrepresentative sample".
Does this mean that if it said "We surveyed NYC, here are the results, therefore more people say yes than no" would be correct? or is it incorrect because it included information regarding the sample. In the question above there was no information regarding how the survey was conducted or who was a part of it but, they still made a generalization of the population from whatever sample they surveyed.
1
4 comments
It's hard to know without seeing the question, but if the conclusion doesn't explicitly state that the sample applies to the general population, then it doesn't necessarily have to have a diverse well-randomized sample representative of the general population. For example, if it said something like "this shows that more of these people believe yes than no", it's referencing back to the people in the sample, which is a valid method.
It'd be easier to know if you could reference the pt/question you're talking about.
I assume that the sample was not representative of the whole. Usually, in flaw questions they explicitly say the sample was diverse/representative/random and if they don't, it means there's no reason to believe the sample is representative.
In your example with NYC, it's incorrect if you're drawing a conclusion about a wider group of people (all of America) from just NYC people because no singular city is representative of a whole country.
Also for a weaken question would a correct answer to "In a survey 60% of participants said yes and 40% said no. This shows more people believe yes than no" be an answer that then applies a specific trait to the survey size? Like if an answer choice said "The survey was conducted of only minors" then we would have reason to believe it is not representative of the broader population because it introduces that nuance needed for it to be flawed?