Scatter Plots Video Lecture Transcript
This transcript was automatically generated by Zoom, so there may be discrepancies between the video and the text.

15:33:31 Hi! Everybody! Welcome back in this video. We're finally gonna start making some charts and tableau with the simple scatter plot.
15:33:39 So in general, tableau works with dragon drop. So we kind of showed this often earlier videos.
15:33:47 But now we're gonna get more explicit about that.
15:33:48 So remember, you have your data panel on the left hand side.
15:33:52 You can drag and drop the desired variables where you'd like onto axes or into the rows or columns or the color, size and shape.
15:34:00 So I think it's going to be helpful just to see this in action as opposed to hearing me talk about it.
15:34:04 Just talk about it. So go ahead if you want to follow along and open your tableau public app, either online or the desktop version.
15:34:12 If it's online, you're gonna have to drag and drop the Galton dot Csv data from the data folder of the repository.
15:34:21 So I'm gonna click on text file and click on the Gulf and data.
15:34:26 That's what we'll be using. Okay. So then you'll see this.
15:34:30 Galton is already within our little data area up here.
15:34:33 So we're good to go. So I'm gonna go to my worksheet now that I'm in the worksheet, I'm gonna demonstrate dragging and dropping.
15:34:41 So again, in this video, my goal is going to be to make a scatter plot and for me, it's gonna be a scatter plot of the height against the father's height.
15:34:49 So I'm gonna put my height here so one way I can do it is to drag and drop the height onto my this side.
15:34:57 And then the fathers, height onto the horizontal area which it's not letting me do.
15:35:03 So once. Since. It's not letting me do that in here.
15:35:06 What I will do instead is, bring this over to the columns, bar at the top.
15:35:11 So this will ensure that father is placed on the horizontal axis which represents our columns and height, is placed on the vertical axis which represents our rows so that's one way to do it.
15:35:26 Before we get further another way to do it is, I can click, and I can add it to the sheet, and I can click, and I can add it to the sheet.
15:35:37 Okay. Another way, you can do it. So I just says a note.
15:35:42 If you want to get rid of variables as you can drag and drop them off as well, so we can go ahead and click and highlight.
15:35:50 Both of these. Okay? So I'm highlighting father.
15:35:52 And height, and I can go to show me, and I can click on this little icon here.
15:35:57 So show me, is it nice? What show me does is it shows you all the various plot types you can make for the things that you've highlighted so I've highlighted 2 measures and one of the things that I can make with 2 measures is a scatter plot so i'll
15:36:14 click on that, and here they flipped them so what I can do instead is flip them back to what I want.
15:36:22 So that's a bunch of different ways. We can do things so as you may recall.
15:36:29 I said that measures which these are their continuous measures continuous because they're green measures because they're down here.
15:36:38 Measures are aggregated by default, and so, instead of seeing the nice scatter plot, we might expect of all the different observations, height against all the different observations, height against all the different observations, height against all the different observations, fathers, height you're
15:36:49 seeing the of the different observations. Fathers, height, you're seeing the sum of the height against the sum of the father column.
15:36:52 So one thing we could do is keep this.
15:36:58 Sometimes this is useful. It's not gonna be useful for us ultimately, but with this you can add variables to the marks.
15:37:06 So for instance, maybe I'll add the sex variable to color.
15:37:10 And now you can see the different aggregated variables for the different sexes.
15:37:15 Here you're seeing that the sum of all the heights of the men the male children, are larger than the sum of all the heights of the female children.
15:37:27 Does that mean anything, you're not necessarily right? Maybe there are just more male children than femal children.
15:37:34 Okay. But in general this isn't gonna be useful for our particular plot.
15:37:38 So how do I get rid of this aggregation?
15:37:41 So you're going to go up to your toolbar up top, click on.
15:37:45 And now it'll see aggregate measures.
15:37:48 Is, has a checkmark next to it. So what we're gonna wanna do is click on that.
15:37:53 And now, if we go back and check aggregate measures, does not have a checkmark next to it anymore.
15:38:01 And so we can see all the individual plots. Now, one thing you might notice is that they're all kind of buged up here, whereas I might normally expect from my experience with Python right, that the axes would adjust so we need to change the axes by hand so you can do that by
15:38:16 double clicking, and this will pop up where you can edit the access.
15:38:21 And so the reason that it's not like this is because by default the axis are suggested to include 0.
15:38:28 So I'm gonna unclick that. And now you can see that it's adjusted, and then I'll double.
15:38:33 Click, my vertical axis and unclick and include 0.
15:38:37 And so now this plot looks a little bit more like the one that I might have expected from when I was working in Matt Plotlib.
15:38:44 Okay, so let's make sure that we covered everything.
15:38:49 Covered show me covered, adjusting the axes. Okay? So the next thing I wanna talk about is the analytics panel.
15:38:58 So this is the data panel or tab. Let's talk about the analytics.
15:39:05 So we click on the analytics panel, and we can add a couple of things.
15:39:08 So the first thing we might be interested in adding is a reference line.
15:39:12 So we'll click on. We can drag and drop reference line here, and then we'll do a line.
15:39:20 And so, for instance, maybe what we want to do is the some of father which would give us the average value.
15:39:31 So it's still saying some of father, even though it's gonna be father, remember, aggregation is not enabled.
15:39:37 But for some reason, when you pull this up, it says some, so what this will do is this, will draw a reference line of the average height of let's double check.
15:39:48 I believe this should be the average height. So and then this is the average father's height. Okay?
15:39:54 So now we have reference lines for that we can remove those cause.
15:39:59 I don't particularly want them, but I'm just showing you that you can do that.
15:40:03 Another thing you can do with a reference line. Another thing we can do with the reference line is, let's go ahead and add a new one.
15:40:12 You can go ahead and do create a new you can go ahead and do create a new parameter and you can just type in like parameter. One.
15:40:22 And maybe my new one will be like 60. You can just set it to be 60. Okay?
15:40:28 So now there's just a line drawn at 60 and then here it's still drawn at the average value.
15:40:36 So we can drag these off. Just demonstrating reference lines.
15:40:40 Another thing we can do is add a trend line. And so what a trend line is is a regression line.
15:40:47 So here we've drawn on trend line, and now we get the regression line for I'm regressing height on father and sex.
15:40:57 And so you can see here, you're getting the different values so unlike with a traditional linear regression where you have a indicator variable, it's just giving you the estimates of the different regression lines for each option of sex.
15:41:12 I'd also hang, you know, if you're interested in that gives you the r squared and the p-value for the regression.
15:41:18 It's I don't know if this is the p-value for the estimate of the coefficient, or for the estimate of the intercept, because they are different.
15:41:26 Okay, so that's that, let's double check.
15:41:30 Did we check everything I wanted to show. So reference lines?
15:41:32 We did that we also did trend lines. I believe I'm not sure if I covered, but you're trend.
15:41:41 Lines can be a simple linear regression, or at least squares linear regression, and not simple, but at least squares regression.
15:41:48 This is the default, but you can also do log rhythmic regression, exponential representation, polynomial and power.
15:41:54 Okay, so now, we've seen how to make a chatter, a chatter, a scatter plot in tablo.
15:42:02 We saw a little bit about on enabling or disengaging the aggregate measures we've also seen how to add a reference line and a trend line, and also saw a little bit about how you can edit your axis.
15:42:17 So we're gonna keep building upon that by allowing more visualization types and how you can make or make them in tableau.
15:42:23 And the next video I hope you enjoyed learning about scanner plots.