Covid and Correlations
When looking at the data sets, there were a handful of things that I was expecting in regards to covid before doing any analysis. For example, I expected there to be a positive correlation between covid deaths and percent of people over the age of 65. I thought that there wouldn’t be a correlation between percent of people under 18 and covid deaths since from my knowledge, younger people tended to stay healthier when introduced to the virus. However, to my surprise, there was no correlation whatsoever between being older than 65 and dying from the coronavirus. Why was this the case? I had assumed that older people were at greater risk and more likely to have fatal effects. But apparently there was no correlation. Was there something I wasn’t considering when looking at the data? This tripped me up. Then I realized that this statistic kind of made sense because it was showing the relationship between the percentage of people over 65 years of age and covid deaths. This meant that the percentage of people in a given county who were older than 65 were not more or less affected by covid. Although when it came to people under the age of 18, it turns out that as expected, there wasn’t a correlation between the percentage of people at that age and percentage of those who died to covid. Nothing seemed too unusual here.
Next I was curious to see how the political aspect of the data related to each other. First, I looked at what the correlation was between the percentage of Trump supporters in 2020 and the number of covid cases per county. And shockingly, there was a negative correlation. At -0.3259, the correlation wasn’t very strong, but it was still there. This was the opposite of what I predicted. During 2020 it seemed as though a lot of Trump supporters were against wearing masks and getting vaccines. So, if this were the case, wouldn’t they be more likely to get the virus? Well, apparently not according to this data. According to this correlation, as the percentage of Trump supporters in 2020 rose, the number of cases of covid decreased across counties. This was very unusual to me. From my knowledge of politics around the year 2020, the more Trump supporters there were, the more covid cases there should have been, but this clearly wasn’t reality. Now the next question was whether or not this correlation was causal. Did an increase in the percentage of Trump supporters in a county cause a decrease in covid? Well, it’s hard to tell. It could be a factor, but it’s not the one thing that decreases covid cases. For example, when we see the correlation between single females and covid cases, it seems to be very weak at 0.095. As for percent of veterans and covid cases, there was a negative correlation, but it was also pretty weak at -0.1445. Now although there are probably more factors at play, after seeing these two relationships and other similar ones that have a very weak correlation, it could be said that according to this data set, as the percentage of 2020 Trump supporters increases within counties, the number of covid cases decreases.
One case could be made against this is the fact that the correlation between the percent of college graduates and the number of covid cases sits at -0.274. This is a pretty strong negative correlation that shows that when the percentage of college graduates increases in a county, the number of covid cases decreases. Implications that come from this include that being a 2020 Trump supporter isn’t the only attribute to a decrease in covid cases within counties and it can’t be guaranteed that increasing the population of these supporters brings down the covid cases.
To bring things to a close, the correlations I expected to see within the data didn’t actually show up. Instead, I found something that proved the opposite of what I thought would be the case. Although there are probably some factors that went unseen during this analysis, it seems that when comparing the correlations of different variables, having a decrease in covid cases within counties may be causal of an increase in 2020 Trump supporters.
