One of the challenges of my daily job is to find new and engaging information to put on a map. However – and you may think this heresy – sometimes a map just isn’t necessary.
It really depends on the accessibility of the data, it’s volume and how it is formatted. This is why I suggest you ask yourself the following questions before you undergo the task of mapping data.
- Is a map the best way to express the message behind the data? (What’s the point you’re making?)
- Do you have enough data to forcefully convey that message?
- Is there another form of visual communication that could get your message across quicker and more effectively? (Especially where the data is not in map-freindly spreadsheet form.)
Putting data on a map
Let’s take an example of what I mean. This morning, I found two sets of data about levels of sleep around the world. Here’s how the first set of data worked out as a heat map:
It’s not a masterpiece – but it’s okay and does a decent job of showing the different levels of sleep in a number of developed countries, with Japan and France standing out for having the highest and lowest levels of sleep, respectively. (Just click on the map to go full-screen and interactive in a different tabbed window). With a little more time, and refinement, it could certainly have held its own as a visualization.
Like-for-like data map comparisons
The data I used above is from a survey carried out in 2009. I wondered was there data available that was more up-to-date? It turns out that there is. The National Sleep Foundation released a report earlier this year, which supplied data on sleep. However, the problem with this report is that it only covered 6 countries. Also, it only deals with work day patterns – so a like-for-like comparison with the first map was not possible.
Nonetheless, the resultant heat map therefore looked like this:
Now I’m not saying this is uninteresting, but the limited data it imparts (which incidentally was not in spreadsheet format) – about six countries – may have been just as well expressed in a simple table:
– though that certainly wouldn’t have been as colorful.
Personally, I would have favored the table over the map in this case. There just isn’t enough data to make an interesting visualization. Also, to convert the raw data into map data format (i.e., csv file) seems to be not the best use of precious time.
That said – I’m open to persuasion! I’d love to know what you think, so comment in the box below should you feel the need.
If you want to explore how your data looks on a map (as opposed to a table), sign up for a free eSpatial account