- Tue Sep 12, 2017 8:02 am
#39615
Tanush - if I had to guess I bet the answer to your first question is, While it's possible that the survey was unrepresentative, the stimulus gives us no reason to think that it's "unlikely" that it's unrepresentative. Usually, when unrepresentativeness is the flaw, the stimulus flags it in some way by mentioning (however subtly) either the number surveyed, which is usually obviously too small, or the group surveyed, which is usually a subset (not in numbers necessary but in kind, as in, only freshmen students were surveyed) of the appropriate, larger group. Neither is indicated here.
I think the testmakers are looking for you to see that, while the students may have generically responded that they prefer a better food provider (who doesn't want better food?), they might respond differently if they knew what the choices actually were. Given that there's only one other choice, and we know that many of them will have direct experience with that choice because Hall was the provider last year, their response would be much more informed and the results might be different. Although it's possible, as you say, that all of the students would prefer Hall, it's also possible that students who said they'd prefer a change generically would nonetheless prefer no change if Hall was the known and only alternative.