Figure 1. Sample allocation and geographic heterogeneity in proxy measures of epidemic intensity.

(A) Survey participant catchment distribution. Wards are shaded by $P_{\text{j}}^{\text{direct}}$, the proportion of all directly recruited participants from each of Somerville Wards 1–7; (B) Cumulative incidence of prior PCR-confirmed SARS-CoV-2 infections by Ward as of June 8th, 2020 $\left({\theta }_j^{PCR}\right)$; (C) Correlation between $P_{\text{j}}^{\text{direct}}$ and ${\theta }_j^{PCR}$. Significance of the correlation is calculated via permutation testing, as described in the Supplementary Material.