Today, we’ll build an interactive, data-rich map with Tableau. As is the case with its data visualization capabilities, Tableau is a powerful tool, but its interface can be a bit confusing. In particular, it’s important to place measures and dimensions on the canvas in a particular order — trust me, I discovered this the hard way after clicking around fruitlessly for hours.
We’ll be building a map that shows the home counties of people accused of witchcraft in Scotland in the late sixteenth through mid-eighteenth centuries. The source of the data is the wonderful Survey of Scottish Witchcraft Database from the University of Edinburgh. Please download the CSV here.
As you did before, you’ll click on Connect to text file (not Connect to Excel file), because Tableau considers CSVs text files. Select the dataset you downloaded earlier.
Once your data has loaded in, click on the orange Sheet 1 button at the bottom of the Tableau window.
As you’ll recall, Tableau divides your columns into dimensions and measures, and then further subdivides them into data types (like “text,” or “number”). Tableau has done a pretty good job, but there’s one field it didn’t recognize: in the Dimensions pane, Date should be classified as a date field, not a text field.
To change it, click on the tiny Abc icon to the right of the word Date. From the drop-down menu, select (of course!) Date. The Abc icon will transform into a tiny calendar.
For reasons that I confess aren’t entirely clear to me, it really matters in which order you add dimensions and measures to the canvas during this process. So follow along as closely as you can!
Double-click first Latitude and then Longitude. (Do not use Latitude (generated) or Longitude (generated). We’ll talk about what these are in the next tutorial.)
These field names will appear in the Rows and Columns panes, respectively. A map will appear on the canvas, containing a single data point. (Apparently the average of the longitude and latitude!)
Let’s create a map that shows a point for the home county of each person accused of witchcraft. To do this, drag the Accused ID dimension onto the Detail button within the Marks pane. (Got that? Drop the Accused ID onto the Detailbutton.)
(Why the Accused ID dimension? It’s because each record has a unique Accused ID number, so Tableau will be sure to map each record.)
Now, each record appears on your map! Not bad!
This is probably a good time to mention that you can “grab” your map to move it around by clicking your mouse and holding down the Shift key.
Perhaps you’d like the points to grow in size if multiple people originated from that location. For that you’ll need a measure. From the Measures pane, grab the Number of Records measure and drop it on the Size button within the Marks pane.
(In reality, no location has that many records associated with it, so this isn’t hugely illuminating.)
As you’ve navigated your map, you may have noticed that tooltips — little information boxes — appear as you hover over points. However, at the moment, the tooltips aren’t as informative as they might be, since they just include latitude, longitude, and an ID number.
There are two steps to creating tooltips. First, for each field you’d like to appear in the tooltip, drag the name and drop it on the Tooltip button within the Marks pane.
For example, if I want each person’s First Name to appear in the tooltip, I’ll drag that dimension over to the Tooltip button.
Do the same for each field you’d like to see in the tooltip.
Now each of those fields appears in your tooltip when you hover over a point. However, you may want to neaten it up a bit. To do that, click on the Tooltip button.
You can delete the fields you don’t want to appear, like latitude and longitude. You can also rearrange things as you like. If you accidentally delete something, click on the Insert button within the tooltip pop-up window to re-insert it. When you’re done, click OK.
If you have several categories in your data, you might use color to distinguish among them. Our records include data for Sex, so we can use color to easily see points for men versus women.
To do that, drag the Sex dimension to the Color button within the **Marks **pane.
If you then click on the Colors button, you’ll find that you can alter the colors Tableau uses by default.
Sometimes it’s helpful to look at only some data, rather than all of it at once. This can be especially useful when you’re trying to explore a dataset. Let’s filter by date. There are several parts to this, mostly because of the way that Tableau treats dates.
First, drag the Date dimension to the Filter pane.
A popup window will ask you how you want to filter on dates. We want to use years, so select Years.
In the next popup window, select All (to indicate that we want to use all the years in our filter) and press OK.
In the Filters pane, you now have a YEAR(Date) dimension. Click on it to produce a drop-down menu. From that menu, select Continuous.
Why continuous? As you’ll see in the next step, we want Tableau to treat dates not individually but as a flowing range of values.
After you select Continuous, a popup box will allow you to use scrubbers to define a date range.
Now you can save and publish your map by clicking on File and then Save to Tableau Public.