Many articles have been written decrying one political party or the other to have a poor or invalid response to the COVID-19 pandemic. But does the political party indicate success or failure when it comes to the pandemic? We explore what the data shows.
Pundits have declared one party or the other responsible for poor responses around the nation with regards to the pandemic. But is this the case? Being genuinely curious (and having our doubts about the claim) we’ll take the study one step further and explore some other highly likely factors using the Pearson product-moment correlation coefficient. The coefficient judges how closely two variables are related, either directly (a positive result) or inversely (a negative result).
To help identify and rank states we used the COVID-19 Project dataset to identify the total deaths in each state. We then weighted these deaths by population to generate deaths per million statistics.
Politically we classified each state by both it’s Governor and the combination of governor and legislature. Governors were classified by party, either Democratic or Republican. The combinations of the legislature were categorized relative to the party of the governor. If both houses of the legislature and the governor were of the same party this was classified as a Democratic or Republican trifecta. If it was not it was considered divided.
Practically we thought there may be other, better factors to consider. In our analysis, we considered the population density, the health care measures of access, quality and overall public health ranking. Finally, we considered the average miles of public transit taken per capita of state residents.
Political affiliation or legislative makeup had almost no relationship to the deaths per million. It seems that these factors have no bearing on the outcome as it related to COVID-19.
But what does? The pandemic is a healthcare centric pandemic, so we considered Healthcare access, quality, and overall public health. Surprisingly these factors do not relate to the number of COVID-19 deaths. This could be explained in several ways. The virus may be relatively easy to treat, making healthcare quality irrelevant, so long as healthcare services are not overrun (the goal of social distancing).
Population density also stood out as a highly likely candidate. It had a moderate relationship with COVID-19 deaths. On a state level, however, this is not the case. There is more of a relationship here than the other factors above, but it is not statistically significant. We do feel that this could warrant more investigation when more granular data becomes available. We feel that the averaging population density across a state may be diluting this factor quite a bit.
Finally, we evaluated public transportation. The average miles of public transportation utilized per capita had the highest relationship with COVID-19 deaths. The statistical significance is considered strong and was the strongest of the factors we evaluated.