# How to draw a box and whispers plot

00:00:01,764 --> 00:00:22,285
We're going to create the excitingly named box and whiskers plot for our data. These charts have the advantage that they let us compare both the center of the data for a variable, in this case it's going to be the management score data for each country and how spread out that data is,so you can intuitively understand two useful things by looking at one picture.

00:00:24,794 --> 00:00:44,104
The first thing to do is select a new sheet and filter and add the data for at least two countries to it. If you can't remember how to do this or you haven't done it before, there are instructions for this in the walkthroughs in Project 1, but we've already done this to save time and here's the data for the US and Chile.

00:00:44,104 --> 00:00:51,853
You can use this type of chart to compare many countries and you should try to do that but we've just kept it at 2 to make the explanation clearer.

00:00:54,409 --> 00:01:06,000
Making the basic plot is simple: select the columns and click on 'Insert' - you can find the box and whiskers plot icon with the statistics charts.

00:01:06,000 --> 00:01:25,647
The line in the center of the box shows the median - that's the middle of the data. The box covers the middle 50% of the data and the edge of the whiskers are the top and bottom measurements, although maybe not quite - we'll explain why not in a moment. But first we need to make the chart easier to read.

00:01:25,978 --> 00:01:36,817
Select 'Design', 'Add chart elements' and 'Legend' will add the legend underneath the data because that's going to be the easiest to see without moving your eyes around too much.

00:01:36,817 --> 00:01:59,835
The legend has automatically recognized that the first entry in the column is the title. We can check this, so right click on the chart, click 'Select data' and you should have seen this dialogue box before, on the left side click on 'Series 1' and edit. As you can see it is using the top cell in the first column as the name.

00:01:59,835 --> 00:02:19,316
This is useful because if we want to update a column header it also updates the legend. We should always try and remove information that isn't useful from a chart. In this case the horizontal axis. Click on it and delete it and we should always add a title.

00:02:20,796 --> 00:02:38,457
Excel has a quirk for this kind of chart, but it's a sensible one: the whiskers may not show the actual maximum or minimum since any values larger than 1.5 times the width of the box are considered outliers and shown a separate points.

00:02:38,457 --> 00:03:04,374
This prevents a few stray data points from creating a giant set of whiskers. You can choose whether or not to show these dots. Now we've got one here. If you double-click on the dot and the box appears on the right, click on the icon of a chart in this box and deselect 'Show outlier points', and the dot disappears. 

00:03:04,374 --> 00:03:16,993
Here our box and whiskers plot tells us that on average US management scored slightly higher, but it also shows us that the best US managers scored much better than the best in Chile.