Background
Replication of Figure 1, Figure 2, Table 1 (April 1st) and Table 2 (April 1st) from the article Differential COVID-19 case positivity in New York City neighborhoods: Socieconomic factors and mobility.
In their article, Matthew R. Lamb, Sasikiran Kandula and Jeffrey Shaman, discuss the effect of certain socieconomic factors, specifically the proportion of 18-to 64-year-old population that is uninsured, median household income, proportion of population that self-identified their race as white, proportion of population living in households with more than three inhabitants, proportion of population using public transportation to commute to work that includes bus travel and proportion of population that is eldery on New York City's Covid positivity rate. They also considered the daily change in mobility as a factor for NYC's Covid positivity rate. They used data from the US Census (ACS) and Safegraph to make these findings.
Our Project
In this project, our goal was to replicate their findings and reproduce the figures and tables from the article.
- File containing all the replications here.
- File containing the code that produced those replications here.
- To use the ACS data provided by tidycensus, users may need to include their own API key using the commented line 19.
All CSV file raw links are provided in the Rmd file and do not need to be predownloaded in order to run the file, however, we attached the links below as well:
- NYC Zip Codes
- April 1 (cumulative) NYC Covid-19 information
- May 1 (cumulative) NYC Covid-19 information
Extended Question
We decided to choose two counties, Manhattan and Queens, to analyze and compare using the same socioeconomic and mobility factors. We were interested in seeing what factors influenced Covid positivity rate the most in these two different counties. We included the distributions of each socioeconomic factor through visual heat maps and developed different linear models for Manhattan and Queens.