Daily Mobility Statistics

The Daily Mobility Statistics were derived from a data panel constructed from several mobile data providers, a step taken to address the reduce the risks of geographic and temporal sample bias that would result from using a single data source. In turn, the merged data panel only included data from those mobile devices whose anonymized location data met a set of data quality standards, e.g., temporal frequency and spatial accuracy of anonymized location point observations, device-level temporal coverage and representativeness, spatial distribution of data at the sample and county levels. After this filtering, final mobility estimate statistics were computed using a multi-level weighting method that employed both device- and trip-level weights, thus expanding the sample represented by the devices in the data panel to the at-large populations of each state and county in the US. Data analysis was conducted at the aggregate national, state, and county levels. To assure confidentiality and support data quality, no data were reported for a county if it had fewer than 50 devices in the sample on any given day. Trips were defined as movements that included a stay of longer than 10 minutes at an anonymized location away from home. A movement with multiple stays of longer than 10 minutes--before returning home--was counted as multiple trips. The Daily Mobility Statistics data on this page, which cover the COVID and Post-COVID periods, are experimental. Experimental data products are created using novel or exploratory data sources or methodologies that benefit data users in the absence of other statistically rigorous products, and they not meet all BTS data quality standards.