6

I have read that wealthier states tend to vote Democratic while poorer states tend to vote Republican. There is data that backs this up. What is the correlation between statewide median incomes and 2 party Democratic vote percentages?

Improve this question
4
  • en.m.wikipedia.org/wiki/… -- here is a list. This is a good faith question I want to know how strong a particular link is.
    – Number File
    Aug 9, 2021 at 14:12
  • Performing the correlation on a statewide basis seems meaningless since states have so much variation within them. For there to be any point to this, you need to look at smaller units
    – divibisan
    Aug 9, 2021 at 14:46
  • 1
    Oh. State level correlation should be interesting. Counties will be in a future question
    – Number File
    Aug 9, 2021 at 14:50
  • 6
    If you've got the data already, why don't you just do this yourself? A simple correlation is easy to do, and based on many of the questions you've been asking here, I think you'd really enjoy picking up some basic skills in R and statistics.
    – divibisan
    Aug 9, 2021 at 17:46

1 Answer 1

Reset to default
9

As indicated in the comments, a simple correlation is not hard to do. You don't need R, Google sheets will do fine:

enter image description here

The percentage voting Democrat in 2020 can be predicted from the average wealth by the formula Democrat-party-percentage = (wealth÷1265)% and R²=0.48 so roughly half the variation between states can be explained by this factor.

But repeat after me... correlation isn't causation

This is somewhat surprising as until 2012, richer people tend to vote Republican, see Columbia U report. As Relaxed notes, it's not really that surprising once you realise that the correlations tell you something different, and they both can be substantial in opposite direction at the same time. Jumping from group correlations to individual correlations is known as the “ecological fallacy”.

Improve this answer
2
  • 3
    (+1) It's not really that suprising once you realise both correlations tell you something different and can perfectly be high at the same time. Jumping from group correlations to individual corrrelations is known as the “ecological fallacy”.
    – Relaxed
    Aug 9, 2021 at 19:59
  • 1
    I agree, and I've adopted your comment.
    – James K
    Aug 9, 2021 at 21:08

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .