If you are a regular reader of this blog you might have noticed how easily I fall into the trap of going into long rants against this and that. Not too big a problem, rants seem to work well in the blogging arena, but I realize that while I have some good criticism here and there, I almost never ever spend time praising someone else’s work. This is partly due to my not being too excited by the average visualizations I encounter on the web, but rest assured there is good stuff to talk about nonetheless. And here we are. I am very pleased to break this habit today and spend some time praising a recent project you might have noticed on the web: the OECD Better Life Index. Let me first explain what it is and then I’ll tell you why I think this great work.
The Better Life Index
The Better Life Index is a visualization developed by the OECD (Organisation for Economic Co-operation and Development ) under the supervision of Jerome Cukier and his colleagues and the help of two external designers, Moritz Stefaner and RauReif. You can find a very interesting description of the the process and the role of the designers in Jerome’s post mortem article.
“The idea was to come up with some kind of Progress Index, which we could communicate once a year or something. Problem – this was exactly against the recommendations of the commission, which warned against an absolute, top-down ranking of countries.
Eventually, we came up with an idea. A ranking, yes, but not one definitive list established by experts. Rather, it would be a user’s index, where said user would get their own index, tailored to their preferences.”
So, the OECD had a bunch of indicators summarizing aspects that concur to quality of life and happiness and wanted to find a way to represent this data in a way that makes people compare one country to another, but in a flexible and not totally pre-ordered way.
The final result, after a number of iterations is a interactive visualization based on icons (they call it flowers). The user can give a weight to the various factors building the index and see how the countries change their ranking. If you want to better understand how it works give it a try, it’s much easier than trying to figure out how it works from my description.
Here is the default result taking into account the whole set of factors.
Here is my own Better Life Index.
What I like and why
Why do I like this visualization so much? I have a number of reasons that I will list in a minute but in summary I like it because it’s a an example of how function and aesthetics can serve each other’s purpose.
There have been endless (and useless IMHO) discussions about form and function in visualization, where people are always a bit right and a bit wrong at the same time. My take at it is that when something works you can see it, no need to spend a thousand words on it. The Better Life Index, is beautiful and functional at the same time and this is the way to go.
The first and most important reason why I like the BLI is that it uses a technique I technically call “multi-dimensional icons”, probably the most neglected tool in the history of data visualization. The thing they call “flowers” it’s actually an icon (some core vis people would rather call it glyph) representing multi-dimensional data.
Every country is an icon, and each icon is made out of a number of “petals”. Each petal represents a data dimension (a better life factor in this case) and its length is proportional to the value the given country has on the corresponding factor. Plus, each factor is colored with a unique hue.
The same technique can be used every time you want to represent multiple dimensions of a series of objects at the same time. If you think about it carefully you will see that there are not many alternatives around. The only problem of this technique is that it doesn’t scale to a large number of objects, but this is not a problem here.
I also like the fact that color was used redundantly to represent the data dimensions. In principle each petal has a unique angle and position so it shouldn’t be a problem to identify which petal is which, but with color it obviously works much better. Plus, it works nicely as a cross-reference to the legend.
Using height as a way to convey the total score is a good choice. You can also change the ordering of the countries from alphabetical to by rank, which I actually prefer (see the gap in the middle now?)
Position is the most powerful visual primitive we have and it makes sense to use it to convey the most important information: how the countries score according to the selected factors.
Speaking of interaction, it is quite simple but effective. It’s easy to understand the function of the menu on the right hand side (see it on the web site) and as soon as you use it you get an understanding of how it works. The animation makes the whole thing smooth and calm, giving the feeling that it’s always possible to return on your owns steps and make several experiments.
To some extent it is possible to get the feeling of which country changes the most when a new set of factors are used, but it doesn’t really work too well, as following all the movements at the same time it’s not easy.
This is the only limitation I have found so far. I would like to have a way to better understand how the rank changes when a new set of parameters is chosen. But I cannot really blame anyone for it. This is the kind of design where the more you try to add the more complex and less appealing it gets. Adding such a functionality would require a lot of additional complexity and there’s no simple and elegant solution that comes into my mind right now.
Money doesn’t buy happiness
I have been playing with it for a while and if you didn’t do it yet I encourage you to try it out. It’s a lot of fun! Try to ask yourself some questions or see whether there is anything strange you notice. I have a number of interesting things I’ve noticed or learned but this one is the one that really stands out for me: the best countries score quite badly in the income domain.
Regardless the composition of factors you choose, there are always a bunch of countries that are leading the BLI, e.g., Norway, Sweden, New Zealand, Canada. Give a look to their flowers: even if they are almost always at the top, they score very badly in one domain: income.
Where should I live?
I don’t know how you interpret the data depicted in the BLI. Personally, the first reaction I had was: “ok then … so where should I live next?” And driven by this idea I started exploring several alternatives. But after a while I felt like two main things were missing.
The first one is information about weather. While in fact I give a great deal of importance to things like work-life balance, I often find myself thinking about weather when I think about a potential new country to live in. Maybe this is the heritage of my Italian origins, but it’s crazy to see how good weather is poorly correlated with the average well-being of a country. Isn’t it? I think some information about weather would be a great addition to the BLI. I would expect to see a lot of shift in the ranking and … well I am sure Italy would score much much better!
Another subtle but more important factor is that it’s really hard to judge quality of life in terms of whole countries. Take the USA, can you really consider these parameters descriptive for any of the main cities in the U.S.? I am sure there are cities that score a lot better and a lot worse than the average. And probably the same is true for any other country. I think it would be really fantastic to see the same data on a city level. Do you think this is even possible? Is the data out there? I would be much more interested to know how the BLI is in Rome, Berlin, Toronto, New York, etc … because in the end when you move to a new place you move more into a city than into a country.
In summary as I said I think the BLI is a very nice example of simple and effective visualization. It takes into account all the constrains posed by the project and comes up with a little nice tool that works just right for its purpose. Is it the best for everything? No, it is not of course. But it’s very well done for the goal people at OECD had. Everyone can learn from it.
My sincere compliments to Moritz and Jerome with whom I often interact on Twitter, and to all the other guys who worked on it. Very well done guys!
Did you like this review? Please do not forget to tweet about it and/or comment on it below. Thanks!