catallaxy files

catallaxy in technical exile

How happy were you at 3.45pm yesterday?

leave a comment »

Last week’s issue of Science magazine had another article on that much-discussed issue in happiness research, the relationship between money and happiness (there is a press release about the article here).

Most happiness research is based on questions like ‘All in all, how happy are you with you life these days?’, or equivalent questions on life satisfaction. If the answers are on (or converted to) a 0-100 scale, most people put themselves between 70 and 80, with on average poor people being closer to 70 and rich people closer to 80. The paradox the research has tried to solve is that while rich people are typically happier than poor people, the fact that we are all much richer than people were in the past isn’t showing in the trend data, which has fluctuated within a narrow band and without long-term direction for 50 years.

These global questions have their advantages. They ask respondents to consider their lives as a whole, and not to pay too much attention to transient ups and down such as whether they are having a good or a bad week (though in practice short-term influences have been shown to affect results). I particularly like the general life satisfaction question, which allows for the possibility that people gain satisfaction from activities that are meaningful even if they are not pleasant.

But these measures also have disadvantages. As Daniel Gilbert’s readable new book Stumbling on Happiness argues, we tend to misremember the past, making it surprisingly difficult to say accurately how we typically feel. Also, in making long-term comparisons of national happiness we cannot be sure that what people understand by happiness is constant over time; a similar problem afflicts international comparisons. I suspect that emotionally people expect more from life now than in the past; that we need a stronger positive emotional state to describe ourselves as ‘happy’ or ‘very happy’ than we once did.

One way around these problems is to take much shorter term measures of how people are feeling, and to be more specific about their emotional state. The Experience Sampling Method collects this information in real time, requiring people to answer questions about their immediate feelings at various times during the day. The Day Reconstruction Method (DRM) lets people fill out, on the following day, a time diary and record how they felt during each episode. The two methods have been found to get very similar results, and as the DRM is cheaper and easier to undertake it is likely to be the dominant method. In a recent article in the Journal of Economic Perspectives, which summarises these two methods, Daniel Kahneman and Alan Krueger suggest creating a ‘net affect’ index: the average of the positive feelings less the average of the negative feelings.

Using the DRM, the Science article, also by Kahneman and Krueger, reports than in a survey of women in Ohio the correlation between income and the amount of good mood in a day was lower than the correlation between income and a general life satisfaction question, .2 versus .32. We’d want much better samples and repeated similar results to be confident of this, but there are plausible theories as to why it could be right.

One of the arguments as to why happiness is not trending up with GDP per capita is the relative income is important. Rich people don’t just enjoy increased consumption; they enjoy a feeling of superiority over the poor as well. And the poor, even though they are materially better off than the rich of the past, feel less happy because they don’t have as much as others. But what the DRM suggests is that people don’t think about these things all the time (I have always had troubled believing that how much our neighbours earn can affect our actual well-being significantly). The global question makes relative position more salient; when asked how happy we are in general we are more likely to use other people as a reference point than when we are asked about our feelings at a particular point in time.

The Science article also suggests that well-off people aren’t doing better because their use of time is not sufficiently biased toward the things in the DRM that have the most net affect (there is a full list in the JEP article). For example, they spend more time than the poor working and commuting, which have the lowest net affect.

If it can be consistently replicated, this kind of survey work undermines, as the article points out, the belief that more money is necessarily the route to happiness (though there is a lot of straw man knocking over going on here – there is no evidence provided for the opening assertion of the Science article that ‘Most people believe that they would be happier if they were richer’). But it would also undermine the left-wing argument that reducing income inequality would necessarily increase happiness, since the DRM is suggesting that income relativities have less impact on day-to-day feelings than they do on the answers to general questions.


Written by Admin

July 1, 2006 at 10:06 pm

Posted in Uncategorized

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: