December 16th: attitudes to work and leisure are shaped by economic reality

First off, I know the text on the chart is small, so if you want to zoom in to investigate where various countries are, there’s a pdf (and hence zoomable) version here. I’m also going to explain the various axes and what they mean, step by step, in the commentary below.

I’ve been having a wonderful time exploring the World Values Survey dataset, which — amazingly — you can download on their website at the granularity of every respondent (around 75,000 respondents across 50 countries). The dataset is not user-friendly: for example, there are more than 250 questions in the questionnaire, but the dataset only gives you the question number, and you have to look up the question — and the various options that were given for answers — in the codebook.

Nevertheless, I’m slowly working through some of the more interesting questions. As you’d expect from the name of the survey, a lot of what it delves into is about people’s beliefs, attitudes and values. A key set of questions is around how important the respondent considers various aspects of life. The exact question is: “For each of the following aspects, indicate how important it is in your life. Would you say it is very important (4), rather important (3), not very important (2), or not important at all (1)?”* The aspects queried include family, friends, work, leisure time, religion, and politics.

[* Actually, in the original questionnaire, the numbers are the other way round, with 4 indicating “not important at all” and 1 indicating “very important”. However, for the purposes of the chart I’ve reversed this, as I thought it would be more intuititive to give a higher score to something that people said was important.]

There is of course already a ton of research on these surveys and how they help us understand the world around us. Today’s chart is just one snippet which I found interesting, comparing people’s attitudes to work and leisure, and to famiy and friends. While the findings make sense, I’m not sure this is what I would have hypothesised before looking at the data. (So, yes, I’m back to my vice that is also my delight: P-hacking.)

What the chart indicates is that the richer the country, the less people consider work to be important (top left panel) and the more they consider leisure time to be important (middle left panel). If you calculate the difference for each person between how important work is vs. leisure time is (bottom left panel), then the pattern is even clearer. Only in a few economies, including the US, Germany, Japan and Russia, do people on average rate leisure time as being more important than work.

Such stated preferences would indeed make sense based on economic theory: in poorer countries, work — and the income it conveys — is significantly more important for people’s livelihoods than in more developed countries where there is at least some element a social safety net (and in many cases, work is also less scarce, i.e. unemployment is lower). Similarly, one could hypothesise that in developed countries, people can “afford” to worry about leisure time (even though they probably already have more of it than people in poorer countries), as their incomes are already sufficient to pay for basic amenities.

The next 2 columns of the chart are also fascinating. In countries where people say “family” is particularly important, they also say, on average, that “work” is important. [Note: “Family” is very important everywhere — you can see that I’ve really had to stretch the X-axis in the middle column to create a spread of dots in the scatter graph.] One could interpret this in at least two ways. First, family is important because it provides a safety net. Second, work is important because it helps support one’s family. The last column shows a reflection of these attitudes in relation to friends. There appears to be less space for them in countries where work (and families) are important; whereas (quite logically) friends are more important where leisure is more important.

Finally, what is also interesting to me about this (because I did also peek at the data on importance of “religion”) is the lack of evidence for the so called “protestant work ethic”, at least when compared to the poorer countries. Of course, we don’t have that many traditionally protestant countries in the sample here, but take Germany. The weighted average importance rating given to work was 3.4, in contrast to, say, Indonesia where the number was 3.9. Looking at the data in another way, just over 40% of Germans said work was “very important”, compared to Indonesia (and many other developing or emerging countries) where this figure is more than 90%.