covid-nyc-2022-group-1

Preview

Based on the article, Differential COVID-19 case positivity in New York City neighborhoods: Socioeconomic factors and mobility by Matthew Lamb, Sasikiran Kandula, and Jeffrey Shaman (2020).

In the Figure 1 section, we created six maps to replicate the Figure 1 maps in the article. We used the ACS data to plot the proportion of 18-64 year olds who are uninsured (14%), the median income ($59,355), the proportion of the population self-identifying as white (48%), the proportion of the population in households of 4 or more (25%), the proportion of commuters who traveled by bus (10%), and the proportion of the population 65 and older (13%).

In the Figure 2 section, we plotted the change in mobility in March and April, using February as a baseline.

We then performed the individual regressions that were mentioned in the articles and we replicated the muli-variate regression for April 1, 2020 cases from table 1 and 2. Included in our report is a comparison on how our R^2 values compare to the values from the original article.

Getting the ACS Data

• Data for each zipcode was taken from the American Community Survey (ACS) API. You will need to set up an API Census key (signup here).
• Use 01_download_nyc_zip_codes.sh to download the NYC zipcodes csv to a Data subfolder.
• Use 02_download_april_1_cases.sh to download the data for the April 1, 2020 COVID cases by zipcode.
• Use 03_get_acs_data.R to get the ACS census data from the ACS API. You will need to insert your own API Census key in the commented out api_census_key. You may need to install.packages("sf") if you get an error for geometry.
• 03_get_acs_data.R will generate a .RData file for easy loading of the data for further analysis.

Generating the Final Report

• Use 04_Group_1_Final_Report.Rmd to generate the final report with the maps and regressions.
• Link to Knitted HTML