I noticed that you discussed a little about some legislators that your models could not predict as well, who all switched parties during their terms. Were there any other factors that affected the prediction power of your models that you did not discuss (maybe amount of donations, majority of donations coming from outside PACs, special elections)?
I noticed that you discussed a little about some legislators that your models could not predict as well, who all switched parties during their terms. Were there any other factors that affected the prediction power of your models that you did not discuss (maybe amount of donations, majority of donations coming from outside PACs, special elections)?