Main effects: Online petition and donation
| . | Intention to Sign . | Self-report: Sign . | Actual Sign-up . | Index: Petition . | Donation . |
|---|---|---|---|---|---|
| . | (1) . | (2) . | (3) . | (4) . | (5) . |
| A. Experiment 1 | |||||
| Treatment | –0.031 | 0.021 | 0.002 | –0.005 | 0.067 |
| (0.053) | (0.055) | (0.019) | (0.050) | (0.056) | |
| [1] | [1] | [1] | |||
| Number of observations | 1,193 | 1,193 | 1,193 | 1,193 | 1,193 |
| Scaled effect | –.04 | .03 | –– | –.01 | .171 |
| Control mean | 0 | 0 | 0.112 | 0 | 0 |
| B. Experiment 2: Main | |||||
| Treatment | 0.061 | –0.069 | –0.036† | –0.004 | 0.222* |
| (0.063) | (0.054) | (0.019) | (0.054) | (0.082) | |
| [.271] | [.271] | [.212] | |||
| Number of observations | 800 | 800 | 800 | 800 | 800 |
| Scaled effect | .09 | –.15 | –– | 0 | .363 |
| Control mean | 0 | 0 | 0.106 | 0 | 0 |
| . | Intention to Sign . | Self-report: Sign . | Actual Sign-up . | Index: Petition . | Donation . |
|---|---|---|---|---|---|
| . | (1) . | (2) . | (3) . | (4) . | (5) . |
| A. Experiment 1 | |||||
| Treatment | –0.031 | 0.021 | 0.002 | –0.005 | 0.067 |
| (0.053) | (0.055) | (0.019) | (0.050) | (0.056) | |
| [1] | [1] | [1] | |||
| Number of observations | 1,193 | 1,193 | 1,193 | 1,193 | 1,193 |
| Scaled effect | –.04 | .03 | –– | –.01 | .171 |
| Control mean | 0 | 0 | 0.112 | 0 | 0 |
| B. Experiment 2: Main | |||||
| Treatment | 0.061 | –0.069 | –0.036† | –0.004 | 0.222* |
| (0.063) | (0.054) | (0.019) | (0.054) | (0.082) | |
| [.271] | [.271] | [.212] | |||
| Number of observations | 800 | 800 | 800 | 800 | 800 |
| Scaled effect | .09 | –.15 | –– | 0 | .363 |
| Control mean | 0 | 0 | 0.106 | 0 | 0 |
Notes: Outcome variables in columns 1 and 2 are normalized by the mean and the standard deviation of the variable for the control group (Kling et al. 2007). In other words, the coefficients represent the effect size in terms of standard deviations away from the mean. The scaled effect is the treatment effect divided by the average difference in the answers given by Democrats and Republicans in the control group. In panel A, we display the results from Experiment 1. In panel B, we display the results from Experiment 2. We include the same list of controls as in Table 2. Robust standard errors are displayed in parentheses; p values adjusted for a false discovery rate of 5% are presented in brackets.
†p < .10; *p < .05