| Does prosocial behavior increase happiness? |
1a. Prosocial acts will increase happiness relative to affectively neutral acts. |
Sample: Canadian and American respondents from Amazon’s Mechanical Turk. |
Analysis: four random intercept models of the form: |
Hypothesis 1a, 2a, 3a, and 4a confirmed if: |
| 1b. Prosocial acts will increase happiness relative to self-focused acts intended to gratify personal emotional needs. |
Design: between subjects; respondents randomly assigned to complete prosocial acts, self-focused acts, or track activities 3 days/week for 3 weeks |
yit = μt+γztzi+αi+εit
|
for: happiness, valued life
|
for: depression, anxiety
|
| where yit is the outcome of interest (happiness, valued life, depression, or anxiety), and zi includes both p (prosocial acts condition) and s (self-focused acts condition). The effects of interest are: |
γp1 > 0 |
γp1 < 0 |
|
γp2 > 0 |
γp2 < 0 |
|
γp3 > 0 |
γp3 < 0 |
|
Measurement intervals: Respondents report activities on the three study days each week, and report happiness, valued life (an aspect of sense of meaning), depression, and anxiety at baseline, at the end of weeks 1, 2, 3, and at follow-up (week 5). |
|
γp5 > 0 |
γp5 < 0 |
| Hypothesis 1b, 2b, 3b, and 4b confirmed if: |
| Does prosocial behavior increase the sense of meaning in life? |
2a. Prosocial acts will increase a sense of meaning relative to affectively neutral acts. |
|
γpt: the effects of p at each time point t
|
| 2b. Prosocial acts will increase a sense of meaning relative to self-focused acts intended to gratify personal emotional needs. |
|
Target Sample size: 360 per condition x 3 conditions = 1080, based on a power analysis targeting 95% power (α = 0.05, 2-tailed tests) |
for: happiness, valued life
|
for: depression, anxiety
|
|
γst: the effects of s at each time point t
|
|
γp1 > γs1
|
γp1 < γs1
|
| We will test all effects using Wald tests. |
γp2 > γs2
|
γp2 < γs2
|
|
γp3 > γs3
|
γp3 < γs3
|
|
γp5 > γs5
|
γp5 < γs5
|
|
Sampling strategy: |
|
| • sample in batches so sampling rate can be adjusted to hit target sample size–adjust based on observed attrition |
| We will count a hypothesis as confirmed if the associated parameter is statistically significant at p < 0.05. Note that hypotheses might be confirmed at some time points but not others. |
| Does prosocial behavior reduce depression? |
3a. Prosocial acts will reduce depression relative to affectively neutral acts. |
| • prevent responses from suspicious IP addresses |
| • replace responses that meet our data exclusion criteria |
| 3b. Prosocial acts will reduce depression relative to self-focused acts intended to gratify personal emotional needs. |
| Does prosocial behavior reduce anxiety? |
4a. Prosocial acts will reduce anxiety relative to affectively neutral acts. |
| 4b. Prosocial acts will reduce anxiety relative to self-focused acts intended to gratify personal emotional needs. |
