Margie Omero at pollster.com has interesting analysis on how the "gender gap" between Obama and Clinton is muddled by other demographic variables. Her analysis is based on a recent Pew poll that finds that Clinton leads among all voters over 50 (+26 among older women and +21 among older men) but the gender gap only really shows among younger voters where she leads by +17 among women and trails by -9 among younger men.
But the far more important predictors of support for Clinton are socioeconomic status and ideology. Men and women in households earning over $50,000 a year both favor Clinton to Obama by 41% to 36%. Clinton among all educated voters, by -3 among women and -9 among men.
Obama leads by +5 with liberal women but trails by -15 among liberal men. Clinton leads by +37 among non-liberal women but ties with Obama among non-liberal men.
So the picture is definitely too muddled to be simply summed up with "gender gap." But some of the sub-narratives do hold up. Hillary is strongest among women in households with less than $50,000 a year (+36) and is stronger among older voters. Meanwhile, youth, higher income, and higher education all play in Obama's favor. I can't say that I can make sense of the interaction between ideology and gender.
Subscribe to:
Post Comments (Atom)
2 comments:
I saw this earlier, it was interesting. But I have come to the conclusion that to really understand the gender gap / racial gap we are going to have to wait until the election is over and study this from the real voting records instead to relying on pollster.
I take it you mean exit polls? They do help but I'm not sure they'll completely resolve all the disputes. Why did Clinton win NH? The conventional wisdom has congealed around the crying and connecting with women but perhaps it was because people stayed home or crossed over for McCain because Obama was winning by so much. Have we had a Bradley effect (which Pew has not found evidence of in recent elections) followed by reverse Bradley effect?
Empirically, I don't see how we can ever disentangle all the theories with the limited data available.
Post a Comment