Sal shows how to construct a scatter plot. Created by Sal Khan.
Want to join the conversation?
- what do you do when you have 3 different things to put on a graf like this
- 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.(4 votes)
- 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!(5 votes)
- How do you find the equation for the line of best fit?(2 votes)
- 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!😊(7 votes)
- A scatter plot is just random numbers on a plot.That is what I am learning lmao.(1 vote)
- It is not actually random numbers on a plot. For example, the question in the video shows how each class performed on average. The plot is a good way to visually see which class did the best on average. You can quickly see that Spanish was the highest dot and that Computer Science was the lowest dot. Often times, scatterplots are the best way to see trends. This example did not show any trends; because it was organized quite randomly. If the data was rearranged by least score to highest score on the x-axis, you will be able to see an upward trend on which period performed the best. Statistics are actually much more useful in real life than some of the other math you learn. Although you weren't asking a question, I hope this still informs you on why scatter plots exist :)(5 votes)
- If you have a bunch of random dots everywhere and then some clusters in some random places,what would you call that?(3 votes)
- 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.(1 vote)
- How do you know where to plot the points?(2 votes)
- Use the information on the table as X and Y coordinates -- if there is no key, which there should be, the independent variable goes on the X-axis, and the dependent on the Y-axis. Then plot the respective X and Y coordinates.(4 votes)
- What do you do if there are two points on the same spot on a graph?(2 votes)
- Are scatterplots just like graphs?(1 vote)
- Yes, scatterplots show raw data, and the directions and flow in them allow us to see trends and make predictions.(4 votes)
- Which (in your opinion) is the best graph to generally use?(2 votes)
- 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.(2 votes)
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.