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

Analyzing mosaic plots

We can use mosaic plots to draw conclusions about relationships between two categorical variables. Created by Sal Khan.

Want to join the conversation?

  • blobby green style avatar for user calum.a.swain
    this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell this is hell
    (0 votes)
    Default Khan Academy avatar avatar for user

Video transcript

- [Instructor] We're told that administrators at a school are considering a policy change. They survey a group of students, staff members, and parents about whether or not they agree with the new policy. The following mosaic plot summarizes their results. Which of the following statements can we justify from the mosaic plot? So pause this video and try to figure this out on your own. Pick which of these statements can be justified and there could be more than one based on this mosaic plot. All right, now let's work through this together. So before I even look at the choices, let me see if I can interpret this. So this mosaic plot, what it does above and beyond a segmented bar chart, is it gives us the width that shows us how many, for example, students versus staff versus parents were sampled or surveyed. And it looks like more than half of the people surveyed were students. And then staff and parents seem similar. In terms of who is agreeing with the policy, so that's that light blue color. It seems like students are not very likely to agree with the policy. It looks like staff is very likely to agree with the policy, that the bulk of staff is agreeing with it. And parents are kind of on both sides. So let's see which of these statements are consistent with what we just looked at. Parents were the least likely to agree with the new policy. No, that's not true. The least likely to agree with the new policy that's students, right over here. They were definitely the least likely. The lowest percentage of students are agreeing with the policy. So I don't like that choice. More than half of the total responses came from students. And that does look like the case. 'Cause if you view this entire width as the total responses, it looks like the student width right over here, that is more than half of it. It looks like it's about 50 something percent or even 60 percent. So I like this choice right over here. And then last but not least, there were more total "No" responses from students than from staff and parents combined. So let's look at the no responses from students. No, that's that darker blue color. The no responses for students is this area right over here. And then the nos, so I can, maybe I'll shade all of that in, and then the no responses from staff and parents combined, that is this area right over here. And it does look like indeed the total no responses from students, this area is much larger than this area right over here. So I like this choice as well.