TABLE 3.
Relationship between COVID‐19 cases and in‐person voting per polling location
| Panel A | Panel B | Panel C | ||||
|---|---|---|---|---|---|---|
| Total cases per hundred | New cases per hundred | Ln growth rate | ||||
| (1) | (2) | (3) | (4) | (5) | (6) | |
| IPV/Loc × Week −3 | −0.043 | −0.025 | ||||
| (0.055) | (0.023) | |||||
| IPV/Loc × Week −2 | −0.030 | −0.018 | −0.479 | |||
| (0.039) | (0.020) | (5.297) | ||||
| IPV/Loc × Week −1 | −0.038* | −0.019 | 0.002 | 0.005 | −1.290 | −1.329 |
| (0.022) | (0.022) | (0.015) | (0.015) | (5.052) | (5.077) | |
| IPV/Loc × Week 1 | 0.065** | 0.046* | 0.031 | 0.029 | 3.126 | 3.264 |
| (0.032) | (0.027) | (0.024) | (0.024) | (3.304) | (3.327) | |
| IPV/Loc × Week 2 | 0.235** | 0.205** | 0.138 | 0.135 | 4.470 | 4.637 |
| (0.117) | (0.096) | (0.090) | (0.088) | (2.930) | (2.944) | |
| IPV/Loc × Week 3 | 0.439** | 0.392** | 0.194** | 0.189** | 7.154** | 7.147** |
| (0.208) | (0.158) | (0.092) | (0.087) | (3.383) | (3.364) | |
| IPV/Loc × Week 4 | 0.641** | 0.515** | 0.162* | 0.146* | 2.097 | 2.032 |
| (0.306) | (0.205) | (0.091) | (0.081) | (3.169) | (3.145) | |
| IPV/Loc × Week 5 | 0.784** | 0.607*** | 0.130** | 0.106** | 3.654 | 3.310 |
| (0.351) | (0.225) | (0.054) | (0.049) | (3.062) | (3.064) | |
| N | 648 | 504 | 648 | 504 | 576 | 504 |
| Time fixed effects | Y | Y | Y | Y | Y | Y |
| Social distancing controls | Y | Y | Y | Y | Y | Y |
| Controls for tests | Y | Y | Y | |||
| County fixed effects | Y | Y | Y | Y | Y | Y |
Notes: The data sources are identical to Table 2. Each regression is estimated using OLS using the same controls as the last column in Table 2. In‐person voters per polling location are measured in 1000s of voters. Total cases and new cases are measured per 100 county residents. This table shows coefficients and SEs clustered at the county level in parentheses. Stars denote statistical significance levels: *10%; **5%; and ***1%.