In this post, we're looking at visualising time series data, but displaying this in a timeline, with the values for each date being represented by the size of the mark on the date data points.
This is a relatively simple viz to put together but one that can communicate a lot of information at once
To build this one, start by connecting to the 'Timline.csv' data source attached below:
Drag the ‘Year’ Dimension onto the ‘Columns’ shelf
Next, change the chart type to ‘Circle’, drag the ‘Live births’ measure onto the ‘Size’ mark and change the view to ‘Entire View’
Next, we will want to create a fixed axis point for the circles so that we can effectively ‘cut them in half’. Create a calculated field called ‘Zero’ with a value of ‘0’
Drag this onto the ‘Rows’ shelf
Edit the Y axis so that it starts at 0 and ends at 1
Alright, now we’re getting somewhere
We want to only look at one area at a time, so let’s drag the ‘Area’ dimension onto the Filter card and select an area
Now, the sizes of the marks are all fairly equal, because there isn’t much variation in the overall values, but we can change that by adjusting the size settings
Click on the drop-down arrow in the size toolbar and select ‘Edit sizes’
Change the ‘Sizes vary’ drop down menu option to ‘By range’ and click ‘OK’
The view should now look something like this
Time for some cosmetics – remove all the lines and borders, increase the size of the circles, hide the headers, and reduce the opacity of the circles slightly
Create a new dashboard (I changed the size of mine to 800x600) and drag the sheet you have created on, add a title, and add a filter
Now all we need to do is annotate the viz. Select all of the marks, then right click any one of them and click ‘Annotate -> Mark’
Select what data you want to display (in my case this was ‘Year’ and ‘SUM(Live births)’ then format as desired
Now comes the boring bit – adjusting the sizes and positions of the boxes until they look right, but once you’ve done that you’re ready to go. Quick & simple timelines effectively showing comparative values over time