Data collection and conclusions — Basic example
About this video
A survey draws 120 people at random from fans of an American film and finds 70% like an Indian remake. The video shows that the conclusion applies only to the group that was sampled. Claims about all moviegoers, or about Indian moviegoers, would need a sample drawn from those groups.
You'll be able to
- identify the population a sample actually represents
- rule out conclusions about groups that were never sampled
- explain what sampling design a claim would need
Key words
- Target population
- The group a conclusion is about, which must match the group the sample was drawn from.
- Sampling frame
- The list or group from which the sample is actually drawn.
- Generalization
- Applying a sample result to a larger population than the one that was sampled.
Transcript
- [Tutor] A film historian selected 120 people at random from a group of people who indicated that they liked a certain American movie. The 120 people were shown an Indian remake of the movie and then asked whether they liked or disliked the remake. Of those surveyed, 70% said they liked the remake.
Based on the survey results, which of the following conclusions about the remake is valid? Pause this video and see if you can work through this before we do it together. All right, now let's work through this together. So choice A says, most people who like the original movie will like the remake.
So can we make that conclusion? So let's see. They selected 120 random people from a group who indicated that they liked the movie. Okay. So these are, this is a random sample of the folks who liked the movie, and we can see that 70% of them liked the remake. So yes, it does seem that most of the people who like the original movie will like the remake based on this sample of 120.
So I'm liking choice A, actually. Let's see, choice B, most people who dislike the original movie will dislike the remake. Well, can we make that conclusion? Because we haven't sampled people who dislike the movie, and because we haven't sampled people who dislike the movie, we haven't asked them whether they like or dislike the Indian remake.
So we can rule this one out. So A is still holding up. At least 70% of people who watch movies will like the remake. Well, no, we can't make that conclusion because they didn't take a hundred, they didn't take a sample of 120 people from people who watch movies, they took 120 person sample from people who liked the original American movie.
So you can't make the 70% claim about all people who watch movies. So let's rule that out. And choice D, at least 70% of people who watch Indian movies will like the remake. So that would be a different if to make this conclusion, you would have to randomly sample people who watch Indian movies and see what percentage of them like the remake.
That was not what was done here. So we could rule that one out as well. So we are liking choice A.
Video © Khan Academy, used under CC BY-NC-SA. Watch on Khan Academy. Puzzicle is not affiliated with Khan Academy.