Extrapolation
Only a fraction of the population is observed, so the counts are scaled up to represent the full population. How the scaling factor is derived depends on the signal: it can come from an observed market share (devices relative to population) or be applied to a model's output and calibrated against a population baseline. Either way the goal is the same, output that reflects real-world visit volume rather than the size of the underlying sample.
Where it applies
In United States this applies to GPS. Each dataset's page shows the model that instantiates it.
Related
Home & work inference
Estimating the likely home and work areas of visitor groups, always at area level and never an address, to enable resident / worker / visitor breakdowns, trade areas, and visitor-only scoping for Visit Length.
Did you find what you were looking for?