I am carrying out a first differences analysis. One of my covariates is “mobility” as it kind of represents the effect of social distancing restrictions.
Since it is a first differences analysis, I need to obtain the differences in mobility from 2019 to 2020 on a week-by-week basis. It is, I need to obtain the difference in mobility between week1-2019 and week1-2020, and week2-2019 and week2-2020 and so on (at the county level).
I have thought about the number of visitors, distance from home, and Home Dwell, but I am not sure if they would be comparable since I guess SafeGraph was following fewer people in 2019 than in 2020.
Therefore, I have some questions,
Which data would it be convenient to use in this case?
Since it is a county analysis, which variable would be convenient to use?
Where could I find Home Dwell information at the county level?
Hey @Paola_Jaimes - I’ll answer these slightly out of order. To filter by county, I’d recommend using the variable poi_cbg in our Patterns product. The first five digits of the CBG code will be the county FIPS code (first two digits indicating state, following three indicating the specific county). That will allow you to group the POIs in your dataset by county.
You’re absolutely right - our panel has grown significantly since its inception. It may be important to normalize the data when doing time series analysis across long periods of time or multiple releases. We have seen success by normalizing visits by the total number of visits in the SafeGraph Panel, month by month. It is also worth exploring normalizing based on state or census block group. With each delivery, we provide you with the Panel Overview Data files to enable you to do these calculations.
The mobility metric you’re trying to create sounds very similar to the Social Distancing Metrics which we deprecated back in April of 2021. However, you should check out this notebook on how to use the Weekly Patterns dataset as a proxy for the Social Distancing Metrics. Think you’ll find it very helpful for your work.