Thanks for the link. One question I'd pose to Nate is how multicollinear were the paper factor and the additional predictors? https://en.m.wikipedia.org/wiki/Mu

Thanks for the link. One question I'd pose to Nate is how multicollinear were the paper factor and the additional predictors? https://en.m.wikipedia.org/wiki/Multicollinearity https://en.m.wikipedia.org/wiki/Multicollinearity Estimates can be unstable if multicollinearity is too high, say if only richer more educated districts use a certain type if voting machine.