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

Constructing a scatter plot

AP.STATS:
UNC‑1 (EU)
,
UNC‑1.S (LO)
,
UNC‑1.S.1 (EK)
,
UNC‑1.S.2 (EK)
CCSS.Math:
Sal shows how to construct a scatter plot. Created by Sal Khan.

Want to join the conversation?

  • hopper cool style avatar for user Rabbit
    why do you have to the score on the y axis? Couldn't you put on x? Could some explain it for me?
    (11 votes)
    Default Khan Academy avatar avatar for user
  • orange juice squid orange style avatar for user Kaiya Dahlberg
    what do you do when you have 3 different things to put on a graf like this
    |------------hight
    |
    |------------hours
    |
    |------------days
    (9 votes)
    Default Khan Academy avatar avatar for user
  • piceratops seedling style avatar for user questioningeverythingiknow
    If you have a bunch of random dots everywhere and then some clusters in some random places,what would you call that?
    (2 votes)
    Default Khan Academy avatar avatar for user
    • cacteye blue style avatar for user Jerry Nilsson
      I don't know if there's a name for it, but the clusters suggest that some 𝑥-values are more frequent than others, and for those 𝑥-values some 𝑦-values are more frequent than others.
      These clusters have a greater impact on the regression than the surrounding dots, but since you say the clusters are also randomly strewn we should still have a weak linear regression.

      Comparing people's heights to the number of shoes they own could potentially produce a pattern like this, with one cluster forming around the intersection of the average female height and the average number of shoes per female, and another cluster around the intersection of the average male height and the average number of shoes per male.
      (8 votes)
  • aqualine ultimate style avatar for user Jaanvi
    I know how to construct a scatter plot but, I have no clue how to "make appropriate scatter plots" I keep getting it wrong. Im not sure how to do that.
    (3 votes)
    Default Khan Academy avatar avatar for user
    • primosaur seed style avatar for user Ian Pulizzotto
      In a good scatterplot, the points make good use of the space on the coordinate grid (for example, the points are not all “bunched up” in a small portion of the grid). Also, the independent variable should be on the horizontal axis, and the dependent variable should be on the vertical axis.

      Have a blessed, wonderful day!
      (4 votes)
  • purple pi teal style avatar for user Lily Gorum
    How do you find the equation for the line of best fit?
    (1 vote)
    Default Khan Academy avatar avatar for user
    • leafers sapling style avatar for user green_ninja
      Try to eyeball a line that goes through the "middle of all the points", drawing it on the graph. Once you've done that, find the slope using the rise and run of the points on that line. Locate the y-intercept as well. Finally, arrange the data into y = mx + b form.

      Hope this helps!😊
      (6 votes)
  • aqualine ultimate style avatar for user guyawesome356
    I think her experiment was bad because some classes are harder than others

    I would have done worse in chemistry than math no matter what periods it was

    if she checked all the first periods and all the second periods and all the third period etc.. it would have been accurate thought
    (3 votes)
    Default Khan Academy avatar avatar for user
  • female robot grace style avatar for user Katie
    Are scatterplots just like graphs?
    (1 vote)
    Default Khan Academy avatar avatar for user
  • winston default style avatar for user kimdz1
    how do you do a 2 way plot
    (3 votes)
    Default Khan Academy avatar avatar for user
  • winston baby style avatar for user 🎁StoppyPoppy🎁
    Which (in your opinion) is the best graph to generally use?
    (3 votes)
    Default Khan Academy avatar avatar for user
    • duskpin sapling style avatar for user Kaileigh Row
      That depends on what your scenario is, and what you want to show and find out. For example, if you wish to show individual results on a class's math test, use the scatterplot. If you want to predict profit for your company, use a line graph. The pie chart would work nicely for showing how much of your sales were in a product or group.
      (1 vote)
  • male robot hal style avatar for user prehistoric-rishi
    I know this is a bit irrelevant, but does anyone know how to get a black hole badge?
    (3 votes)
    Default Khan Academy avatar avatar for user

Video transcript

Aubrey wanted to see if there's a connection between the time a given exam takes place and the average score of this exam. She collected data about exams from the previous year. Plot the data in a scatter plot. And let's see, they give us a couple of rows here. This is the class. Then they give us the period of the day that the class happened. And then they give us the average score on an exam. And we have to be a little careful with the study-- maybe there's some correlation depending on what subject is taught during what period. But let's just use her data, at least, just based on her data, see if-- well, definitely do what they're asking us, plot a scatter plot, and then see if there is any connection. So let's see. On the horizontal axis, we have Period. And on this investigation, this exploration she's doing, she's trying to see, well, does the period of the day somehow drive average score? So that's why Period is on the horizontal axis. And the thing that's driving is on the horizontal, the thing that's being driven is on the vertical. So let's plot each of these points. Period 1, average score 93-- right over there. Period 6, 87. Oh, that's not the right place, and then we can move it if we want-- 87, right over there. Period 2, 70. Period 4, 62-- right over there. Period 4 and 86, that's right over there. Period 1, 73. Period 3, average score of 73 as well. Period 1, 80, average score of 80. And then Period 3, average score of 96. So there we go. And it doesn't really seem like there's any obvious trend over here. So let's make sure that we got this right. And we did.