pikapixels

HackDTech 2020 submission

Comment

pikapixels

HackDTech 2020 Submission

Kyra Chan, Jin Cho, Evelyn Cupil-Garcia, Jessica Yang

Track: Health/Inequality

Data Exploration and Visualization of COVID-19 Racial Disparity with a Focus on Testing Sites and Case Statistic Comparisons

The Issue

The United States has a history of racial inequality in a multitude of aspects of ones life, and the COVID-19 pandemic is no exception to this. A significant part of the pandemic response has relied greatly on data analysis, as there is always new information being discovered about the virus. However, one issue that the Center for Disease Control has indicated is that there are a lot of holes in data on COVID cases and deaths by race and ethnicity, whether there be many unreported statistics or complete datasets missing related to race. In a situation where data is pertinent for better understanding the impacts of the outbreak and bettering the resources available for response, this missing race-related data has limited the racial and ethnical perspective on the impacts of the disease, therefore leaving greater potential for racial disparities in what resources are made available.

Our goal with this web application is to begin to bridge the holes created by missing racial data, specifically in North Carolina counties, expanding the resources that could be used to catalyze more equal decisions of COVID response resource allocation in terms of racial communities. In order to emphasize the legitimacy severity of the issue at hand, we analyzed the racially-split data on COVID cases and deaths. We then took another approach through the analysis of racial population distribtution in NC by county alongside the distribution of COVID-testing sites. And finally, in order to allow for this web application to act more than just a compilation of data visualizations, we also embedded an interactive map for users to be able to search for NC COVID testing sites.

Methodology in Data Analysis

Datasets and sources:

Racial Data on Cases

Due to our focus on the resources available for COVID response, we chose to calculate and use the percent of positive COVID cases that resulted in death. This percentage serves as an indication of how effective the response after being tested positive was meaning that, the higher the percentage of deaths, the less effective the response. With the focus on Black and White communities in NC, the trends of this percentage changing over time for each show a steeper decrease in percentage over time amongst White communities. Despite having had a higher percentage of deaths in the earlier months of the outbreak, there was a steep decrease as time went on while the trand for Black communities has a much slower decrease with current percentages higher than in White communities. This is in support of the findings by the Center for Control Disease in that there is a clear difference in racial impacts of COVID in multiple states, making it important to distribute response resources in a way that is easily accessible to all.

Racial Data on Testing Site Distribution

Access to COVID testing sites is very important and the first step to response. Therefore, we tried to see if there potential explanation for disparity in racial data on deaths vs. positive cases in potential uneven distribution of test sites. We made observations both by race and by ethnicity, with a focus on recially Black and ethnically Hispanic communities.

  • Multiple observations made in process (more details on webapp):
    • Number of Hispanics vs. Number of Non-Hispanics per site in each county
    • Percentage of Hispanics vs. Percentage of Non-Hispanics vs. Number of sites in each county categorized by different total population sizes
    • Percentage of Hispanics vs. Percentage of total sites in NC in each county
    • Percentage of Black vs. Percentage of White vs. Number of sites in each county

Although we weren't able to see clear relationships and conclusions in each of these approaches to the data, we chose to focus on the Hispanic population and how there is a disparity in access to sites in some counties that with large Hispanic populations. Although there were few clear trends over all counties, there were anomoly counties in which, through our analysis and web application, we hope to emphasize the possibilities of inequal resource allocation for COVID response.

There are also confounding variables that should be taken into consideration, as our research focuses on one aspect that could act as a foundation for further research. Such variables for further research include county budget, insurance availability, etc.

Web Application

Built With

Share this project:

Updates