Brian Beutler has an excellent post up at Talking Points Memo, debunking an alarming Washington Post report that “The number of black and Hispanic registered voters has fallen sharply since 2008, posing a serious challenge to the Obama campaign…” Here’s the crux of Beutler’s take-down:
In recent years, according to Michael McDonald, a government and politics professor at George Mason University, the Census’ Current Population Survey statistic the Post relied on has varied in a troubling way with the ultimate turnout figures. Whether you compare presidential years (2004 and 2008) or midterm-years (2006 and 2010) the CPS measure has found turnout decreasing. The opposite has been the case.
That, McDonald argues, is because of a peculiarity in the way Census compiles its registration and turnout figures. It asks one adult to answer for all members of a household, and counts those who fail to respond “yes” or “no” to the voting survey question as having not voted.
When you revisit the numbers after throwing out all the non-respondents, the results track the official figures much more closely.
What does all this have to do with registration (as opposed to turnout)? When you perform the same correction to the registration results — the ones the Post used — the problem goes away.
You can almost hear a collective ‘ouch’ coming from WaPo’s HQ. Beutler quotes McDonald’s conclusion:
“The Obama campaign appears better situated in terms of registering of Blacks and Hispanics in the wake of the 2010 election than in the wake of the 2006 election. That these minority populations are also growing in size relative to the non-Hispanic White population should give more worry to the Romney campaign than to the Obama campaign.”
Apparently the sky isn’t falling on the Obama campaign after all. Nice catch for Beutler and TPM.