Schools may be undercounted for visits because Safegraph do not collect data from <13 year olds individuals. Is this the case?

Hi @Ryan_Fox_Squire_SafeGraph! We have some questions we hope you will give us some insights:

  1. Schools may be undercounted for visits because Safegraph do not collect data from <13 year olds individuals. Is this the case? Can we split POI by primary/middle and high schools? And within high schools how confident you are in your measure (since we saw in another post that students that have a regular pattern might be counted as workers). Thanks for your insights.

Yes, there is no data for < 13 y/o in accordance with COPPA. This certainly has an impact on how visits to schools are measured (but should apply to both pre and post pandemic in the same way).

We use naics codes and unfortunately does not separate these different types of schools. You could possibly try to separate this based on the names (i.e. contains “Elementary” contains" Middle" contains “High”)

Historically visits to schools has not been a core focus of our business since these patterns are well established and known and not of high interest to our pre-pandemic customers.

One challenge with schools is that, unlike a lot of other retail locations, they tend to be more isolated in residential areas. The challenge with this is that its easier for us to mistake a visit to an adjacent residence as a visit to the school.

It’s also possible that students could get classified as workers.

Schools aren’t exactly consumer businesses points of interest, so they are not our core specialty.

That’s all to say that there are reasons why I am less confident in the school visits data. Sorry I can’t provide even more concrete advice.

Ok thanks for this information. To be clear a student would be counted as a worker only if >18 correct?

@Seth_Benzell_MIT_IDE we do not know the age (or anything else) about the individual devices.

All we know is that NONE of the data is from < 13 year old.
Note 13, not 18.