I thought I would take a break from politics to look at how demographics and local broadband access predict COVID vaccination rates and case mortality. The above graph shows a strong positive correlation between the percentage of the county that is non-Hispanic white and the case mortality. A possible outlier might be Huntingdon county with the second lowest percentage but the second highest case mortality. Even with this outlier, this relationship accounts for 64.8% of the variability in case mortality. For every 1% increase in the non-hispanic white population there is a predicted 0.1% increase in the case mortality rate. The county variables were assembled by County Health Rankings.
A similar negative relationship was found between the percentage non-Hispanic white and full vaccination rates in the above graph. Westmoreland county is a potential outlier with the 4th lowest non-Hispanic white rate but the second highest vaccination rate. This relationship accounts for 52.8% of the variability in the vaccination rate. The regression equation states that every 1% increase in the non-Hispanic white percentage predicts a 1.7% increase in the vaccination rate.
Broadband Access
The above graph shows that counties with higher broadband access rates have higher vaccination rates. Somerset and Indiana Counties have the lowest broadband rates around 75% and vaccination rates below 50%. Centre and Westmoreland Counties have the highest rates at 83%. A potential outlier is Bedford County with the lowest vaccination rate at 38% but not the lowest broadband rate at 77%.
The regression equation estimates that every 1% increase in the broadband rate predicts a 1.6% increase in the vaccination rate. This relationship accounts for 47.5% of the variability in the vaccination rate. This does not mean that access to broadband causes an increase in the vaccination rate. It is possible that access to media causes people to be more informed and motivates them to receive the vaccination. It is also possible that those who receive the vaccine are more likely to want greater broadband access.
The regression model for % broadband access and COVID Case mortality was negative but not statistically significant. The graph above shows that there is a more spread around the regression line.
Demographics and broadband access predict COVID vaccination and case mortality rates. A lack of broadband access could be a reflection of the higher poverty rates. Next I will look at other demographic variables.