There is an ongoing debate about opening the country’s economy. The issue centers around the notion of whether it is safe to get back to work. In the discussion there is also the issue about the rising unemployment numbers. Although the actual unemployment numbers are yet to be collected, anecdotally they seem to be as close, if not worse than the unemployment numbers during the Great Depression.
Part of the problem with the debate is that correlating incomplete data does not provide for a fact-based analysis of the Covid-19 problem. For example, the number of deaths that are reported by each state do not consider whether the reported deaths are counted equally by all jurisdictions.
Likewise, the number of reported infections of Covid-19 does not show an actual count of the infection rate because the testing for Covid-19 is incomplete and haphazard at best. Reported cases are a factor for testing. The less individuals tested the less reported cases exist.
Unemployment numbers are likewise difficult to ascertain. There are certain factors that cannot be quantified now. For example, are unemployment numbers temporary until the states begin to reopen or is the unemployment rate more permanent as businesses reassess their viability.
Regardless it is important to understand if there is a correlation between Covid-19 infections and the unemployment rate. Is the rush to reopen a factor of alleviating the pressure on the tax base? Or, is it because the pandemic problem has been resolved?
The best data for unemployment we have access to is from the Department of Labor. Their most current data is from April 18. The data they track is the number of unemployment filings for workers who were insured at the time they sought unemployment benefits. Not all employees are insured under unemployment. Thus, seeking unemployment does not mean that they will receive the benefits.
Our first data metric is the number unemployed benefits extended by each state. These are benefits that have been deemed as qualified for benefits. Our second and third metrics are the number of Covid-19 cases and deaths.
I am visual person.
To understand complex data, I need to visualize it. In this first analysis, I created cartogram hex map for each data point. A cartograph map is designed to visualize map data based on the density represented by the data rather than the geographic limitations. A traditional map distorts the data by the limitations of geography. For example, New York has a much larger population than Wyoming, but the map of the United States does not represent that accurately.
These are my first data visualizations. They are raw data and are inherently distorted because a state with a large population and an average number of cases is not equal to a state with a smaller population but a higher number of reported cases.
They are not apple-to-apple comparisons.
Nonetheless, the visualizations gives us the level of the problem across the nation.
But there is more work to do. We need to create an apple-to-apple comparison.
In tomorrow’s post I will delve deeper into the metrics and present a better comparison based on the available data. One of the things that will be interesting to look at is to see whether there is a correlation between states that are open for business and those that remain closed.