But there should be a better way of looking at this than just correlating. Below find links are some attempted analyses of the data, and thankfully none show that primary turnout correlates with who wins the general. So many factors play into general election results such as third term curse, economy, incumbency, etc.. Therefore, I think that a better way to look at it would be if some of the geeky 538 type folks would look at “predicted vote” to see if candidates emerging from lower turnout primaries underperformed their expected performance (based on prediction models that take into account incumbency/economy etc) in the general or not. Probably even less data to go on that way, but would be more relevant than just simple correlation.
These links are for future reference for me or if anyone wants to take a better stab at it:
Politifact summary
Sean Trende at RCP look at it in 2012
Philip Bump at the Fix
Monkey Cage
If anyone sees any other looks at this, especially ones who look at relative performance, not just who won or not, let me know in comments. Not staying up any later tonight though :)
This is mostly just a diary with some data for if I ever get around to trying to look harder at this...