willing to participate in our survey. Employed respondents did not receive a compensation
payment.4
The questionnaire consisted of three parts. In the first part, respondents were asked to
assess how their mental and physical well-being the previous day compared to a typical day in
their life, to list all activities they were engaged in during the course of that day, and to note
the start and end time of each activity. The diary had to be constructed without gaps before
starting with the second part.5 In the second part, respondents gave a detailed account of what
they did, with whom they interacted, and how they felt during each activity listed in their
diary. We specifically asked respondents to assess how strongly they experienced various
affect dimensions on a scale from 0 (“not at all”) to 10 (“completely”). Positive affect was
measured using the attributes “relaxed”, “happy”, “comfortable/at ease”, and “enjoying
myself”. Negative affects comprised “lethargic/dull”, “insecure/anxious”, “stressed”, and
“frustrated/annoyed”. Respondents also rated their general satisfaction during each episode on
a scale from 0 to 10. In the third part, respondents answered questions about themselves and
their life circumstances, e.g. their general life satisfaction, job satisfaction (where applicable),
health status, education, income, number of children, social contacts, employment, and
marital status. Instead of these questions being asked at the beginning, they were asked at the
end of the interview to avoid that drawing attention to these issues would influence the
responses to earlier questions.
3. Results
3.1 Descriptive Statistics
Table 1 summarizes some descriptive statistics, separated into subsamples of the employed
and the unemployed. The average age of employed respondents is 44.2 years, of unemployed
respondents 38.2 years. Male and female respondents are equally represented in our sample.
About 27 percent of respondents in both groups are single, while the other 73 percent are
either married or cohabiting with their partner. The employed have, on average, 1.2 children
4 Our pre-tests showed that the response rate was in fact lower when we offered a compensation of 10 euros to
the employed than when we did not offer any compensation. This surprising effect suggests that people have an
intrinsic motivation to participate in scientific studies which could be crowded out if a monetary compensation is
paid. The unemployed, on the other hand, were more willing to participate if given compensation.
5 We followed the recommendation by Kahneman et al. (2004c) that the diary be completed before respondents
become aware of the specific contents of later questions. Otherwise their construction of the diary might suffer
from selection bias.