If you're seeing this message, it means we're having trouble loading external resources on our website.

If you're behind a web filter, please make sure that the domains *.kastatic.org and *.kasandbox.org are unblocked.

Main content

Data collection and conclusions — Harder example

Video transcript

- [Narrator] A polling agency recently surveyed 500 adults who are selected at random from a large city and asked each of the adults whether they supported a new federal policy. Of those surveyed, 42% responded that they support the new policy. Based on the results of the survey, which of the following statements must be true? So, pause this video and see if you can figure this out. All right, now let's work through this together. So, let's look at the statements. So, statement one, of all the adults in the city, 42% support the new federal policy. Let's see, they took a sample of 500 adults who are selected at random and 42% of them supported the policy. So, of all the adults in the city, 42% support the new federal policy. Well, we don't know that for sure 42% would be a pretty good estimate of it based on the sample. So, I'm feeling a little bit queasy about choice one, let's try to see choice two, If 500 different adults selected at random from the same city were surveyed, 42% of them would respond that they support the new federal policy. While I'm feeling queasy about that as well because once again, it's likely to be close to 42, but it could be 43%. We can might've just gotten lucky with the number of people who supported the federal policy it could be a lot lower or we could have gotten unlucky and it could be a lot higher. So, I'm feeling queasy about that one as well. So let's see, if 500 adults selected at random from a different city were surveyed, 42% of them would respond that they support the new federal policy. Well, that one feels actually the hardest one to believe, because now you're looking at a completely different city that it could have very different views, I don't like that one either. So actually, I like none of these choices, you don't know just based on the sample of 500 and you got 42% of them supporting. You don't know that every time you get a sample that you're gonna get exactly 42. You also don't know that it's exactly 42 of the entire population in the city. And you definitely don't know that if you took 500 adults at random from another city, that it would be 42%. So, I would say none.