Data Snapshot: Are Hawaii Social Distancing Measures Changing Behavior?
by Nick Redding, Hawaii Data Collaborative
Social distancing measures have been implemented across the United States, with each State taking different approaches to minimizing person-to-person contact. A recent analysis, conducted by the Santa Fe-based geospatial modeling consultancy Descartes Labs and reported by the New York Times, used anonymized cell phone data to measure changes in average daily travel distance over the last three weeks. Below is what they found for Hawaii. Each column represents a cluster of weekdays, starting with the week ending March 6. The column on the far right shows the week that distancing measures were enacted.
Source: New York Times; Descartes Labs
You can use this analysis (albeit crudely) to explore the correspondence of the first official announcements in Hawaii to counter COVID-19 spread, and travel behavior of residents:
3/18/20: City and County of Honolulu orders the closing of establishments and public gatherings
3/20/20: County of Hawaii announces closing of County parks and beaches
As noted in the Times article, relying on cell phone data can certainly be problematic in terms of precision and inconsistency of measurement. The analysis also only looks at travel distance, which biases against rural counties where travel is more critical for work or meeting basic needs. However, for the question of whether or not social mitigation strategies are working, this analysis certainly provides some evidence of behavior change.
We are currently working to conduct a Hawaii version of this analysis, identifying changes in travel patterns at the community (sub-county) level as a proxy for social distancing effects. Such an analysis will allow us to see, in near real time, areas where mitigation strategies are working well, as well as “hot-spot” areas where movement and vulnerability to rapid spread of the virus remain high. Stay tuned.
