Cuebiq Data for Good program allows researchers to visualise their data on mobility trends. More information can be found here https://www.cuebiq.com/about/data-for-good/  

Cuebiq Mobility Index (CMI) - A measure of the median distance traveled by all devices.

• The Cuebiq Mobility Index (CMI) quantifies how far users move each day. It is calculated using a derivative factor indicating the distance between opposite corners of a box drawn around the locations observed for users on each day. Using an index allows counties to be compared against one another even if users travel different distances on a normal day. The CMI for each county is the median of the aggregated movements of all users within a county.

• CMI values can be interpreted as:

5 - 100 km,  4 - 10 km,  3 - 1 km,  2 - 100m,  1 - 10m

A CMI of 2.5 for a county, would mean the median user in that county is traveling 250m.

All variables:

1. Reference date - Date

2. Week Name - Epoch week code String/Varchar

3. County Name -Name of geographical region String/Varchar

4. State Name State name of county String/Varchar

5. County FIPS Code - The Federal Information Processing Standard code that uniquely identifies a county String/Varchar

6. CMI - Cuebiq Mobility Index - Float/Double

7. Sheltered in Place - Percent of devices that sheltered in place, defined by a max distance of 100 meters - Float/Double

8. % less than 1 mile - Percent of devices that traveled 1 mile or less - Float/Double

9. % less than 10 miles - Percent of devices that traveled 10 miles or less - Float/Double

Mobility Index Data Structure

The table below outlines the structure of Mobility Index data that will be shared with clients. Files will be delivered in CSV format and organized by daily partitions. There will be one file that is updated each day with the entire history of available Mobility Index data.

Cuebiq reserves the right to add additional columns to the data schema at Cuebiq's discretion. Any additional columns will always be added to the end of data schema.

For analysis we are using several open python packages: osmnx, geopandas, plotlly. 

We studied data provided from Cuebiq https://www.cuebiq.com/resource-center/resources/cuebiq-spearheads-covid-19-data-collaborative/#utm_source=twitter&utm_medium=socialmedia&utm_campaign=blog-cuebiq-spearheads-covid-19-data-collaborative

We look at a set of measures that could be directly used for calibration of epidemic models on a nationwide scale. These measures are directly reflecting mobility and contact patterns in population.

• Cuebiq Mobility Index (CMI) - A measure of the median distance traveled by all devices. e.g. A CMI of 2.5 for a county, would mean the median user in that county is traveling 250m.

• The Shelter-In-Place Analysis represents the percentage of users staying at home in any given county

As it was mentioned in cuebiq, there is a clear threshold when plotting the distribution of #distance from home vs. #number of users: it is evident that the distribution is bimodal, the first peak (about 10 meters) is that of people staying at home while the second is related to people who move - to work on other third places.

For more mobility insights please look at github (send me request to access the code) https://github.com/Liyubov/cuebiq_mobility 

Acknowledgments 

 "Aggregated mobility data is provided by Cuebiq, a location intelligence and measurement platform. Through its Data for Good program, Cuebiq provides access to aggregated mobility data for academic research and humanitarian initiatives. This first-party data is collected from anonymized users who have opted-in to provide access to their location data anonymously, through a GDPR-compliant framework. It is then aggregated to the census-block group level to provide insights on changes in human mobility over time" .