For LBS/activity type data, we perform a population-level normalization for each month of data in order to determine a "pop-factor". For each census block, StreetLight InSight measures the number of devices in that sample that appear to live there, and makes a ratio to the total population that are reported to live there per the 2010 US Census. A device from a census block that has 1,000 residents and 2 StreetLight devices will be scaled differently everywhere in comparison to a device from a census block that has 1,000 residents and 200 StreetLight devices. From there we calculate a single "pop-factor" for each device by adding up the assigned population from each of its census tracts, weighted appropriately.
Since residential blocks are highly affiliated with income, race, and other demographic features, this approach also normalizes for bias for those features.