Notions of rest and self-care – be it meditation, napping, or a walk in nature – have become increasingly popular, and recent economic research suggests these activities can improve not only mental well-being but also performance overall. Nevertheless, uptake and engagement of these kinds of activities remain low – especially in stressful labor market situations. In the context of online labor markets, we explore workers’ beliefs on and decisions to invest in rest. We document the productivity effects of mandated rest periods during a strenuous work task. Furthermore, we look at who chooses to invest in rest when given the option and ask whether “break-skippers” and “break-takers” are maximizing their earnings.
Intervention
Individuals will randomly be assigned to a break during the work period. During the break, participants listen to an audio script that is meant to help refocus.
Experimental design
Workers are hired to perform an incentivized attention-demanding task, and we allow them to voice a preference for resting or working. The choice will be randomly implemented for a subsample. Others will be assigned at random. First, we study the effects of breaks on performance and well-being. Second, we aim to understand what individuals prefer and how these preferences related to earnings – in both levels and effect sizes.
Furthermore, a separate sample of workers will be hired for a separate task that is designed to isolate their preferences over working and resting, holding earnings fixed.
Design details
Randomization method
Randomization done by a computer
Randomization unit
The treatment is randomized at the subject level
Clusters
N=1,000 individuals
Observations
N=1,000 individuals are included in the experiment. However, we will focus the analysis on 95% of the sample that were randomly assigned to the treatment. (For every individual, we observe 4 to 5 segments; and for every segment, we observe 300 image evaluations.)
N=400 individuals are included in the additional preference-only experiment.
We focus the analysis on lower-income individuals that we pre-define during a screener before data is collected.
Arms
For the main work experiment: 5% of participants will receive their stated preference over work
95% of participants will be allocated randomly to a break or no-break treatment (with ½ to each). We expect this to be around N=475 subjects for each of the break and no-break arms.
Primary outcomes
Work Performance:
- Overall experiment earnings
- Per-segment earnings (i.e. productivity)
Uptake of Break:
- Preference for break in middle of work period (for self)
- Preference for break vs work for randomly selected teammate whose earnings pay out to respondent
Financial optimality of break and heterogeneity:
- Effect of break on overall experiment earnings for those who select break vs those who do not
- Effect of break on per-segment earnings for those who select break vs those who do not
Secondary outcomes
Subjective well-being:
- [From main work exp] chronometric happiness, task satisfaction, and fatigue data (according to Liu & Netzer 2023)
- [From main work exp] Willingness to accept future work
- [From main work exp and short exp] Preferences over rest or work
Performance:
- Correct/incorrectly evaluated images per segment (per type of stimuli)
- Fraction of correctly evaluated per image within segment
Attrition:
- Segments completed
- Number of images evaluated
Beliefs:
- Workers beliefs about the effectiveness of breaks for self and others
- Beliefs about own performance (absolute and relative to others)
Cited by
No citation data available for this registration yet.