My last three posts have showed how I looked for a few simple stories in an excel spreadsheet on NHS waiting times, using filters and pivot tables.
I produced three results:
Here, I’m going to talk through my attempts to visualize all of these using some basic free tools on the internet. There are dozens available. Some of the most popular are:
- datawrapper.de – Pros: quick and easy to use, you just copy and paste in your data (making sure to take out symbols like £ signs or any commas first), and pick your chart. Cons: It doesn’t work well with labels of more than a couple of words or with visulizing quite large datasets – it tends to run the categories/points together.
- Tableau – Pros: Beautiful visualizations and a wide range to choose from. Cons: You have to install it on your computer first (although this doesn’t take long), and I’m told it can be temperamental.
- Google Charts –Pros: More control than datawrapper offers over the appearance of your charts, and much more able to handle big data. Cons: Limited in your choice of charts – other services like ManyEyes or Tableau may offer a bigger range.
I’m going to try to show you how to visualize data using a few of the above.
Let’s start with my data on which treatment function was responsible for the most treatments taking over 18 weeks:
Datawrapper.de talks you through creating the visulization fairly explicitly. Go to the website and hit ‘create a chart’. Then, just highlight all of the data that you want to visualize (including row/column labels or headers, but leaving out any ‘total’ tows), and paste it into the box. Mine looks like this:
Hit upload and continue and check that it’s formatted your data into its proper columns. Make any necessary tweaks and then hit ‘visualize’.
Here you can choose between a few different layouts – I stick with the basic bar chart. If your data doesn’t look right, try transposing it. If the labels are running together, your better bet is to try google charts or tableau.
And the final product should look something like this. Unfortunately you can’t easily embed the charts into a wordpress.com blog, but the link above shows my final result.
Let’s move on to my data on which provider was responsible for the longest waiting times over 26 weeks. I’m going to give tableau a try for this.
Tableau‘s not difficult to get to grips with – after installing it on my computer I hit ‘excel’ under the ‘connect to data’ option. I upload my file, choose the right sheet, and we’re away.
The layout works a bit like a pivot table. Drag and drop the measures you want to use from the sidebar on the left, into the boxes of the table in the centre.
I drag ‘treatment function’ into the rows box and ‘sum over 6 months’ into the values box. This gives me an identical table to the one in my spreadsheet, albeit not in order of largest to smallest.
The toolbar hovering on the right, titled ‘show me’, allows you to experiment with different charts for the data that you’ve entered. You can play with this for a while, trying out different combinations. For now, I’m going with a simple bar chart again.
Hovering either on the bar at the top of the chart, or located on the ribbon at the top, are two buttons that looks like a series of bars getting larger or smaller respectively, with an arrow – these will allow you to sort the data.
In order to share my final chart, I go to ‘file’ and then ‘share to web’. You have to create an account to do this. I name my sheet and it presents me with both an embed code and a link to my chart:
For my final set of data that I want to visualize, showing which providers had the most treatments taking over 52 weeks (or a year), I’m going to use Many Eyes.
Many Eyes has a process similar to datawrapper.de – after logging in, clicking ‘create a visualization’ gives you the option to upload your data by pasting it into the box provided.
Where Many Eyes has the edge over datawrapper is in giving you a much wider range of visualizations to choose from – from treemaps to stack charts.
Treemaps generally work better for data that is nested, or hierarchical, but I gave one a go:
I’ll move on to google charts in another post that will also cover fusion maps.