Table A2.
COVID-19 Case Growth and Prior Proximity to Cases, by Two-Week Period.
| log(Change in Cases per 10k Residents + 1) |
|||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| March 31, - April 13 | April 14, - April 27 | April 28, - May 11 | May 12, - May 25 | May 26, - June 8 | June 9, - June 22 | June 23, - July 6 | July 7, - July 20 | July 21, - Aug. 10 | Aug. 11 - Aug. 24 | Aug. 25 - Sep. 7 | Sep. 8 - Sep. 21 | Sep. 22 - Oct. 5 | Oct. 6 - Oct. 19 | Oct. 20 - Nov. 2 | |
| 2 Week Lag: | 0.735*** | 0.411*** | 0.150** | 0.204*** | 0.580*** | 0.178** | 0.287*** | 0.225*** | 0.243*** | 0.305*** | 0.314*** | 0.152** | 0.371*** | 0.149** | 0.313*** |
| log(Change in Social Proximity to Cases + 1) | (0.093) | (0.088) | (0.060) | (0.061) | (0.062) | (0.074) | (0.057) | (0.067) | (0.072) | (0.077) | (0.080) | (0.068) | (0.073) | (0.069) | (0.069) |
| 4 Week Lag: | 0.339 | -0.190 | 0.157* | 0.053 | -0.116* | 0.196*** | 0.057 | 0.097 | 0.084 | 0.024 | -0.046 | 0.041 | 0.128* | 0.049 | -0.141** |
| log(Change in Social Proximity to Cases + 1) | (0.434) | (0.127) | (0.082) | (0.060) | (0.062) | (0.074) | (0.058) | (0.062) | (0.069) | (0.075) | (0.083) | (0.072) | (0.070) | (0.068) | (0.065) |
| Share of Friends within 50 Miles | 0.250 | -0.167 | 0.069 | -0.180 | -0.218 | 0.039 | 0.484* | -0.424* | 0.774*** | 0.203 | 0.060 | -0.547** | -0.455* | 0.398* | 0.776*** |
| (0.247) | (0.292) | (0.278) | (0.272) | (0.261) | (0.281) | (0.262) | (0.251) | (0.232) | (0.253) | (0.267) | (0.259) | (0.250) | (0.229) | (0.214) | |
| Share of Friends within 100 Miles | 0.066 | 0.657* | 0.259 | 0.913*** | 0.191 | -0.043 | -0.514* | 0.824*** | -0.213 | 0.300 | -0.018 | 0.629** | 0.845*** | -0.475* | -0.655*** |
| (0.284) | (0.336) | (0.320) | (0.311) | (0.298) | (0.322) | (0.300) | (0.286) | (0.264) | (0.285) | (0.301) | (0.292) | (0.282) | (0.258) | (0.243) | |
| 2 Week Lag: | 1.125*** | 0.486 | 2.089*** | 1.207*** | -0.112 | 2.281*** | 1.401*** | 1.821*** | 2.129*** | 2.098*** | 1.977*** | 0.985* | 2.723*** | 4.118*** | 3.876*** |
| log(Change in Physical Proximity to Cases + 1) | (0.189) | (0.388) | (0.284) | (0.256) | (0.316) | (0.436) | (0.355) | (0.428) | (0.607) | (0.761) | (0.752) | (0.521) | (0.637) | (0.456) | (0.480) |
| 4 Week Lag: | -2.193*** | -0.156 | -1.686*** | -1.072*** | 0.429 | -2.705*** | -1.551*** | -1.802*** | -2.127*** | -1.929** | -1.824** | -0.949* | -2.452*** | -3.748*** | -3.664*** |
| log(Change in Physical Proximity to Cases + 1) | (0.724) | (0.430) | (0.289) | (0.274) | (0.287) | (0.443) | (0.332) | (0.405) | (0.618) | (0.806) | (0.768) | (0.544) | (0.633) | (0.448) | (0.496) |
| 2 Week Lag: | 0.172*** | 0.381*** | 0.554*** | 0.463*** | 0.276*** | 0.361*** | 0.328*** | 0.346*** | 0.371*** | 0.315*** | 0.283*** | 0.423*** | 0.255*** | 0.367*** | 0.327*** |
| log(Change in Cases per 10k Residents + 1) | (0.059) | (0.050) | (0.037) | (0.036) | (0.036) | (0.041) | (0.033) | (0.036) | (0.037) | (0.041) | (0.042) | (0.035) | (0.037) | (0.035) | (0.035) |
| 4 Week Lag: | -0.083 | 0.124* | -0.026 | 0.046 | 0.128*** | -0.005 | 0.003 | 0.016 | 0.004 | 0.033 | 0.131*** | 0.056 | 0.014 | 0.076** | 0.155*** |
| log(Change in Cases per 10k Residents + 1) | (0.247) | (0.075) | (0.047) | (0.037) | (0.035) | (0.040) | (0.033) | (0.034) | (0.036) | (0.040) | (0.043) | (0.039) | (0.036) | (0.033) | (0.033) |
| Pop. Density FEs | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y |
| Median Household Income FEs | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y |
| State FEs | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y |
| Sample Mean | 1.239 | 1.257 | 1.334 | 1.372 | 1.429 | 1.586 | 2.038 | 2.530 | 2.707 | 2.675 | 2.627 | 2.629 | 2.840 | 3.040 | 3.356 |
| R-Squared | 0.608 | 0.574 | 0.644 | 0.648 | 0.666 | 0.615 | 0.673 | 0.705 | 0.732 | 0.657 | 0.606 | 0.617 | 0.637 | 0.680 | 0.713 |
| N | 3,135 | 3,135 | 3,135 | 3,135 | 3,135 | 3,135 | 3,135 | 3,135 | 3,135 | 3,135 | 3,135 | 3,135 | 3,135 | 3,135 | 3,135 |
Note: Table shows time-specific results from regression 5. Each observation is a county. The dependent variable is log of one plus the number of new COVID-19 cases per 10,000 residents in one two-week period between March 30, and November 2, 2020. All columns include log of growth in social and physical proximity to cases, as well as log of growth in actual cases, lagged by two and four weeks (one and two time periods). All columns include time-specific fixed effects for percentiles of population density and median household income, time-specific fixed effects for state, and estimations of the share of a county’s Facebook connections that are within 50 and 150 miles Significance levels: *(p0.10), **(p0.05), ***(p0.01).