[WS:] Sorry, I'm ADHD ;) I started with quoting an opinion that giving and volunteering is associated with conservatism which seems to have some support. But then I tried to offer an alternative explanation arguing that the correlation, while apparent, may be related to a third factor - social class fraction, and the claimed correlation between political views and giving may thus be spurious.
This is a standard mode of analysis in sociology - you see A and B correlated, so you want to test if there is a causal connection. One way of disproving a hypothesized causal connection between A and B is to show that both are correlated with a third factor C. A classical example would be correlation between test scores and school type - you try to disprove causal connection between the two by showing that school choice and test scores are related to a third factor, e.g. socio-economic background. So if you successfully show that, you effectively falsify the causal hypothesis, if you try and fail, you provisionally accept that hypothesis.
Wojtek