This is the next post on my series on County Health Rankings and Trump’s PA county vote regression model. After the number of unhealthy days and flu vaccination, I will focus on the percentage with access to exercise and Trump’s vote. Access to exercise is “percentage of population with adequate access to locations for physical activity.” The model states that for every 1% increase in access to exercise opportunities, there is a predicted 0.183% decrease in Trump’s county vote in PA.
The graph above shows the univariate relationship between the variables. Eyeballing the graph, it is apparent that the relationship between the two variables is not linear. A best fit straight line accounts for 53.4% of the variability. More precisely, a quadratic line accounts for 66.5% of the variability. Even more precisely, a cubic trend line accounts for 71.2% of the variability. The graph with the cubic line is below.
The cubic line above shows an in increase in Trump vote for counties with less than 45% with exercise access. The curve then flattens out between 45 and 65%. Finally, it decreases precipitously for counties with more than 65% access. A possible outlier is Fulton County with 48% access and 86% for Trump.
One should always be careful about concluding a cause and effect relationship from correlational data. There is the directionality: does greater access cause greater vote for Trump or greater support for Trump cause there to be less access? Also there is always a potential 3rd variable that can explain the relationship. With the multiple regression model we can rule out unhealthy days and flu vaccination as an alternative explanation. Interestingly, an infinite number of 3rd variables have not been considered.
Allegheny Independent Media is affiliated with Nonprofit Partners. It is a subsidiary of the Community Foundation for the Alleghenies. Through the site you can donate to the website. It is essential to allow us to provide coverage to the critical issues facing the Alleghenies. You may donate to our site here.