Repository relies upon the data maintained by Johns Hopkins at this repository. The scripts combine the separately reported confirmations, deaths and recoveries into one file. The data were converted from wide to long. NYT original story did not include key aspects--deaths and recoveries-- of the data on the interactive map.
Additional variables like population density and volume of international air traffic appear to have some role in covid infection rates, leading to disproportionate media coverage in those cities and states. More rural and isolated states, like Kentucky, must also have rigorous analysis to judge the potential health impacts to their populations.
A bash shell file was included and serves to update the data on the local build to keep the data fresh. The file was run just after the Johns Hopkins data was refreshed using crontab.
To run, source("./000_main_script.R").
## Scripts run in numerical order
.
├── 000_main_script.R
├── 010_download_confirmed_case_data.R
├── 011_download_death_data.R
├── 012_download_recovered_data.R
├── 020_filter_to_confirmed_us_cases.R
├── 021_filter_to_deaths_us_cases.R
├── 022_filter_to_recovered_us_cases.R
├── 030_combine_tidy_data_sets.R
├── 040_add_variables_for_final_data_frame.R
├── 050_convert_csv_to_json.R
├── 060_build_confirm_case_growth.R
The data was the foundation for two blog posts: The first one dealt with the distribution of cases by state in the United States. The map was built in jquery using Leaflet.
The second post showed the growth in Covid-19 confirmations in Kentucky and some contiguous states. The chart was generated in Plotly.
Plotly requires registration and an API key.
Distributed under the MIT License. See LICENSE for more information.
Rob Wiederstein @RobWiederstein - khuon68@gmail.com
Project Link: https://github.com/RobWiederstein/covid-19-ky
- New York Times for their free Covid-19 coverage and data-driven stories.
- Washington Post for their equally rigorous data reporting and data animations.