20
previous day compares to an “average” day of the respondent. The determinants of general
life satisfaction are in line with the literature. This shows that respondents in our sample
behave in a similar way to people observed in large-scale social surveys. Specifically, life
satisfaction is significantly reduced by unemployment. Income is positively correlated with
life satisfaction and highly significant. People with some professional education report to be
more satisfied with their life than people without vocational training, but we do not find a
significant difference between persons with vocational training and those who received a
university degree. Life satisfaction is U-shaped in age. People that report to be more satisfied
with their health also report higher satisfaction with their life in general. This could be
because health is an important determinant of quality of life in and of itself, but since our data
is cross-sectional and not a panel, we cannot preclude the possibility that this correlation
captures differences between general degrees of optimism between people that simultaneously
affect both satisfaction measures. Differences between the life satisfaction reported on
weekdays and weekends as well as the general “goodness” of the previous day compared to
an average day do not have significant effects on general life satisfaction.
The same explanatory variables have a quite different impact on a person’s duration-
weighted experienced utility than on life satisfaction. In particular, unemployment is not
associated with a drop in well-being. After controlling for the other factors reported in Table
5, the unemployed feel better than the employed on weekdays. In the case of the net affect,
this difference is even statistically significant, thus confirming our results in section 3.3. On
weekends, both employment groups report higher net affects, although the increase is only
significant for the employed. The difference between both groups on weekends is not
statistically significant for any of the three measures of experienced utility. Hence, while we
find that a person’s life satisfaction is harmed by unemployment, we do not find evidence of a
similar effect for experienced utility.