Friday, November 13, 2009

Day of the Week Again

A new IZA working paper. Weekends are worse. The findings in this paper are very counter-intuitive. The findings are not the result of something funny going on in the specifications. Figure 1 on page 39 is a straight descriptive graph showing well-being by day of the week. Wellbeing in Germany is lowest on Friday, Saturday and in particular Sunday, which is different to the USA and different to the Irish findings Gerard produced. The German sample is certainly large and well sampled. The authors point out that their data allow them to put forward a range of controls but the findings are robust to these. I think the most important issue is that the simple descriptive stats are so out of line with what one would expect. Initially, I wondered whether there was a simple miscoding of the satisfaction variable but its average value is what you would expect, and it reacts to unemployment and so on in the manner one would expect (and the authors are both experienced published analysts).

I prefer weekdays to weekends myself for the most part but I had assumed (and many people agree) that was due to me being weird. Perhaps Germanic is a better description.

link here


Kevin Denny said...

Naw, we'll stick with "wierd". When I worked in Germany decades ago, weekends could be pretty dull as lots of things shut from Saturday lunchtime - I think of it as a Protestant thing. But that may well have changed. It was similar in Dublin & worse up north where in some communities Sunday was for reading your Bible.
What are the results for Austria?

Michael Daly said...

It's an interesting one.. we expect affect levels to fluctuate each day due to contemporaneous factors like sleep deprivation etc. But questions relating to thinking about life shouldn't vary so much.

Looking at the question used in this survey- How satisfied are you at present with your life, all things considered?’ 0 (completely dissatisfied) and 10 (completely satisfied), it's clear that trends in Sunday life reflections may bear no relation to trends in affect patterns.

It would be worth seeing are affect and satisfaction patterns similar or do they diverge and if so is this true everywhere. It could just be due to an unspecified 'Germanic' cultural disposition that produces less positive reflective life-satisfaction judgements on sundays!