Recruitment location influences bias and uncertainty in SARS-CoV-2 seroprevalence estimates