George MacKerron, 31, is an environmental economist at the London School of Economics. After graduating in biological anthropology, he worked in web and software development for the BBC and the University of Cambridge before returning to academia. While studying for his PhD at the LSE he devised and developed Mappiness, an iPhone app-based research project designed to measure wellbeing against environmental factors in a new way, and on a much larger scale than previously attempted.
How did the idea come about?
I found myself increasingly interested in environmental issues. Then a friend sent me some of Richard Layard's [an LSE economist's] lectures in happiness economics, and I thought that it would be interesting to put these two areas together and explore the effect of people's natural environment on their wellbeing. Most of the little work that has been done on this looks at the negative side of wellbeing – the effects of environment on mental and physical ill-health. There is very little simply asking people how they feel.
Tell me about Mappiness
It is a free iPhone app. It is anonymous – Mappiness will beep you twice a day, and you respond telling it a few details: how happy you feel, how awake you feel, how relaxed you feel. Then it asks for a little bit of context: what types of people you are with; are you indoors or outdoors; what are you doing – there are about 40 options. The app will also use the microphone to measure noise levels. And while you are answering the questions, the iPhone's GPS tells us via satellite where you are. The whole process takes about 20 seconds. And you can get feedback – a record of your happiness levels – through the app.
What has the response been like?
Since we launched in August nearly 32,000 people have taken part. That is 10 times the number I optimistically hoped for. I imagined most people would stay involved for a week or two and some did, but others have given us hundreds of responses. We have about 7,000 people who are being beeped every day, and more than 1.5 million responses overall.
So it is groundbreaking in scale?
It is enormous. Studies of this sort, that monitor continuously people's feelings, usually involve a few dozen people, who are given handheld computers [PDAs] or diaries. My idea was to take that methodology but apply it to something that thousands of people could take part in.
Isn't it simplistic to look at one variable - environment - and attribute happiness levels to that?
A lot of the details we are asking for are control questions. So for example if we want to see whether people are happier when they are in the park, there could be lots of question marks. Is it the park that makes people happy or is it that happy people spend time in the park? Are they happier because they are having a picnic that happens to be in the park? Because they are with friends? Because they are not at work? The questions we ask help us control for those variables and focus on the effect of the environment.
What about weather?
Because we know where people are, and when, we work out what the weather is and factor that in. And yes, rain does make people less happy. If you think of the happiness scale as one to 100, when it is raining people are about half a point less happy if indoors, and almost one and a half points less happy if outdoors.
Not a huge effect…
To put it in context, our preliminary results suggest that the "happiest" activity is "intimacy, making love", which adds about 12 points. Being ill in bed is the least happy activity and takes off almost 20 points.
What else are you finding? Are people happier in cities or in the country?
It turns out that people are happier in every other environment than the urban environment, and the effect appears to be between about one and five points. Mountains and coniferous forests have come out as the happiest places so far – four or five points higher than a continuous urban setting. Being in the suburbs scores about one point happier than a continuous urban environment.
Is there an argument that cities attract more angsty people?
The great thing about Mappiness is that we don't have to worry about that because we get results from the same people again and again. It's called a "fixed effects" model – the person doesn't change, so we can factor out their typical response level and focus purely on the variation. If someone always answers 20 out of 100, then answers 30 when they are up a mountain, that comes out as a positive, even though other people are always answering 70.
Have there been any big surprises in your findings?
No, I have been pleasantly surprised at how intuitive the results have been. The big fear is that you get the data back and they are just a lot of random noise and you can't make anything of them. So it has been great that the things that you expect to make people happy, make people happy. People are happier at the weekend than in the week. People are happier when they are not at work, when the sun is shining and so on. The really interesting part will be finding out just how big these effects are.
What are the potential applications?
On the way we spend money, publicly and privately. Governments are not necessarily doing the best job at finding the balance that makes the population happiest. And on an individual level, people don't always make optimum choices. The major decision you make in terms of your environment is probably where you live. I don't yet have an answer – it is a trade-off between factors that affect wellbeing. It isn't necessarily that we should all go and live on mountain-tops…
Any advice for someone planning a similar project?
The biggest advantage that I had is that I had worked in software development so I was able to create the technical side of Mappiness myself. One of the biggest problems in developing software systems is communication between the people who understand the software and the people who don't exactly know what they want because they don't understand the software.
