When posting the previous posts on poverty on the sub-reddit for Pennsylvania I was asked repeatedly how many of the poor counties voted for Trump. This post is an attempt answer their question. I looked at the correlation between median income, poverty rates and Trump’s % of the vote for 2024. What I found was that median income predicts Trump’s 2024 Pennsylvania vote but poverty is at best a weak predictor.
Median Income
The graph above shows the relationship between median household income and Trump’s % of the vote in 2024 for PA counties. Median income means that if all of the incomes in a county are ranked from highest to lowest, the median appears right in the middle. There is a significant negative correlation between median income and Trump’s vote for PA’s 67 counties. This means that as county income levels increase, Trump’s % of the vote is likely to decrease. This relation accounts for 31.7% of the variability in Trump’s vote in PA counties. If 100% of the variability were accounted for, all 67 counties would form a straight line sloping downward.
As usual, Philadelphia County is an outlier in this relationship. It had the 10th lowest median income ($60,500) and the lowest % of the vote for Trump (20%). In 2020, Trump had 17.9%. If Philadelphia county is excluded from the above analysis, 45.9% of the variability is explained.
Poverty Rates
When we look at poverty rates and Trump’s vote for 2024 a different pattern emerges. There is a slight positive correlation but it only accounts for 1% of the variability in Trump’s vote. Philadelphia County is an outlier again with the second highest poverty rate in PA at 19.7%. If Philadelphia is excluded, on the 7.9% of the variability is explained. At best this is a very weak correlation. The poverty rates are based on the federal definition of poverty.
Conclusion
Median income predicts Trump’s % of the vote in 2024. This means that as those in the middle level of the economic spectrum lose buying power, they are more likely to vote for Trump in that county. Sociologists and political scientists will debate the reasons why this is so.
Bucks, Chester, and Montgomery counties have the highest income levels and some of the lowest % of the vote for Trump. Their income levels have been rising faster than the state and national levels.
As the poverty rates in a county increase, there is no clear pattern regarding Trump’s vote in that county even with Philadelphia considered. Having larger portions of their population at the lowest end of the economic scale does not necessarily translate into higher votes for Trump.