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

16:33:17 Hi! Everybody! Welcome back in this video. We'll discuss how you can make a histogram or a box plot in tableau so remember, these are 2 ways to visualize distribution.
16:33:29 So histograms will go ahead and make a series of bins that count the number of observations that fall within the range of that bin and box plots visualize the inter- quartile range, as well as certain observations that fall outside of like one and a half times that
16:33:49 Inter-cortile range. So we're first gonna learn how to make a histogram.
16:33:53 So again, instead of showing you all these slides, I'm gonna go ahead and switch over to the tableau app.
16:33:59 This is the desktop App. We're still working with the penguin data set that we worked with in the last video and we're going to show you how to do that.
16:34:07 So if you're doing this on the desktop app you should be able to access it directly from the repository.
16:34:11 If you're using the online version of the app, you'll need to drag and drop the penguins data set into there.
16:34:18 And just as a reminder at the end of the slides that are within the repository, you'll find a link to the Irish Institute tableau, public profile where you can download a completed version of the workbooks today okay, so the first one we're going to show you is how to build a
16:34:34 histogram. So in order to do that, I'm gonna again look at body mass, I'm gonna click on it and now I've clicked on it. I'm gonna go all the way over to show me and I'm gonna find the histogram which looks like sort of the normal distribution
16:34:47 here and then click on it so now that I've clicked on it, I have a histogram here, and we have some new things that we haven't seen yet, so there's something up here. Count.
16:34:59 So count just goes ahead and counts the number of observations that fall in each bin.
16:35:04 But the bins themselves are defined by this binning, variable or binning measure here, and you can notice the parenthesis bin.
16:35:13 This was created automatically by tableau when we wanted to make the histogram, and if we go over to our data pane over here, we can see that it has been created as a dimension, so we can go ahead and click this arrow, and if you notice we can click on edit and if we edit
16:35:33 it. It allows us to change the size of the bins.
16:35:37 And so here, the size of the bins are the width of the bin.
16:35:41 So currently bends are at the width of 225.
16:35:44 You can also change things like the range of values or suggest a bin size.
16:35:49 If you want. So let's go ahead and make these bins instead of 2 25, just make them 200 and click.
16:35:55 Okay. And we can see how that changes the visualization.
16:35:59 I also wanna point out that there is a null value here.
16:36:04 So anytime that you make a chart or a visualization, or a table in tableau, and it includes data that is missing you can go ahead and it will be notified.
16:36:15 Template, and notify you typically in the bottom, right hand corner.
16:36:19 I believe you can click on that and by clicking on that it gives you the option to filter out data where there's a missing value.
16:36:27 So if I click on filter data, now that missing value has been filtered out and won't be considered in any other observations before we continue on and show off box and whisker plots, I do want to show you what happens if you try to color the histogram by a dimension so here, i'm going to try and
16:36:46 put by the island.
16:36:49 And you can see that, unlike with Matt Plot, Lib, or Seaborne.
16:36:55 When I make a histogram and tableau, and add color to it.
16:36:58 The bins themselves are broken up according to the number of observations in each of the dimensions.
16:37:06 So here it's going to be each bar is broken up into the number of number of penguins from each island in that bin.
16:37:15 So here we can see it by hovering over and maybe this is a new thing we haven't talked about.
16:37:18 Tableau automatically provides tool tips. We'll talk about that in more in-depth.
16:37:25 In a later video. So here we can see from our tool tip that 9 penguins from the Torgerson Tgerson Island are in this bin, and then 15 from the Dream island are in this bin and 3 from the Biscu Island.
16:37:41 Are in this bin. So this is slightly different, and it I' say it does make it a little bit harder to see the distribution, because the bars are not all resting for the dream.
16:37:52 And Bisco Island. The bars are not all resting on their own separate plane.
16:37:57 So if we wanted to do this a little bit differently, we could also add to the columns the island sorry, not the columns, the island sorry, not the columns, but the rose, the island, and this would break out the histograms as their own plots, which makes it a little bit easier to see
16:38:11 the distribution for each island, I think. Okay, so that is gonna go ahead and be our talk about histograms.
16:38:19 I'm gonna go ahead and create a new sheet.
16:38:22 And this worksheet is gonna go ahead and be where we talk about boxing whisker plots.
16:38:27 So I'm gonna go and create the body mass click on that.
16:38:31 And once again go to show me, and I should be able to go to a box and whisker plot.
16:38:38 So first, I mean, for some reason I had to make the histogram first. But now that I have a history, I can create a box and whisker plot here, and importantly, I believe I need to turn off aggregate measures, yes, so now that i've turned off aggregate measures and.
16:38:58 I don't know why it's still doing this.
16:39:00 I've turned off at aggregate measures.
16:39:02 So for some reason it's still showing it like this.
16:39:04 So I'm going to turn it into a dimension.
16:39:06 So now you can see that it's showing this. And so, and tableau, there's a couple of things that are going on here.
16:39:14 So in a to our traditional box and whisker plot.
16:39:18 Here, where we have our inner quarter tile range as a box, and then our whisker.
16:39:21 We also see plotted each of the observations it's a little bit difficult I would say, to see overlapping ones, but we see that plotted.
16:39:32 We can also include, for instance, columns here. So, for instance, we could get a column for the all the different types and so let's go ahead.
16:39:42 And we can get rid of those nulls once again from the body mass, and we can also get rid of the nulls for the sex.
16:39:50 So we're only gonna look at the data for which we have information.
16:39:52 So now we've got our intor indoor quartile ranges.
16:39:57 I also wanna point out something that, with our inter quartile ranges, those are not colored.
16:40:03 So, for instance, maybe I'm gonna color it by sex, and the boxes themselves are not colored.
16:40:12 Just the points are colored. So I can actually edit the appearance of the box by right-clicking or in Mac control-clicking and click.
16:40:19 Edit and once they hit edit it allows me to change the style of the inter-core tile range itself, so I can change it to be very dark gray, or I could change it to be purple or blue or brown.
16:40:37 So let's go back to the original light.
16:40:39 Gray. I can also change how far the whiskers extend to, so either the maximum extent of the data or the traditional 1.5.
16:40:50 You can hit this button here that will say, hide underlying marks except outliers.
16:40:56 So if I do that, that gets rid of all of the observations that fall within the inter-cartile 1.5 so within the whiskers, so it will get rid of those cause. Maybe you don't like looking at those I'm gonna bring them back just to demonstrate something
16:41:11 else. Okay, so we can also go into color and maybe you'll have some points that are gonna be existing outside that whisker.
16:41:24 Those are outliers, so those won't be removed when you hide them.
16:41:30 And maybe you'll have some points that are gonna be existing outside that whisker. Those are outliers, so those won't be removed when you hide them.
16:41:37 So if you turn the opacity all the way down to 0, they'll still be there.
16:41:40 But you won't be able to see them, so let me put the opacity back at 100.
16:41:44 But when I go ahead and edit the plot.
16:41:49 And then click the hide underlying marks. I can go back.
16:41:53 And now they are not there, you can tell, because the tool tip is not changing as I hover over.
16:42:01 I'm just seeing the tool tip for the box plot.
16:42:04 Okay, so let's go ahead and dial-check with our lecture to make sure, I covered everything.
16:42:11 So we made a histogram. We talked about bins. We talked about bins.
16:42:14 Talked about editing the bins. We talked about missing values, coloring the historyograms, making a box and whisker plot, coloring box plots and editing the box plot appearance okay?
16:42:29 Alright. So we've covered everything in histograms and box plots.
16:42:34 You're now ready to admit, I believe. Make those on your own.
16:42:36 If you would need to remember. I will be saving this workbook and adding it to the error.
16:42:41 Schrodinger's Institute public Tableau, public Profile.
16:42:44 So if you'd like to look at the completed version, you can find it there, and your version of the slides that are uploaded to the repository will have an easy-to-click link for you to go to okay, I hope you enjoyed this video I enjoyed having you watch