Don’t Stereotype Voters Based on Demographic Information

Many of us who write about politics have a sloppy habit of using catchall phrases when we should be more specific. We use terms such as “Hispanics” and “Latinos,” for example, to describe a group of 52 million Americans whose politics differ greatly, partly depending on their ancestral country. Cuban-Americans have traditionally voted Republican, though less so in recent years. Puerto Ricans are among the most ardent groups in the Democratic base. Mexican-Americans have traditionally voted more Democratic, but as their household incomes rose through the years, their willingness to vote Republican increased as well, at least until the immigration-reform issue came to prominence again. Remember that President Bush captured 44 percent of the Latino vote in 2004, one of several key factors in his narrow victory over John Kerry. As Mitt Romney proved, it’s hard to win the White House with just 27 percent of the Hispanic vote, even if you carry white voters overwhelmingly.

The same can be said about women voters: Generalizing too much is dangerous. Overall, President Obama beat Romney by 11 points among women, who made up 53 percent of the electorate last year, while Romney carried men, who were 47 percent of the electorate, by just 7 points. In the national vote for the House of Representatives, the numbers were almost identical: Republicans carried the men’s vote by 8 points, and Democrats won the women’s vote by 11 points.

But when you start slicing and dicing the women’s vote, big differences emerge that sometimes contradict the broad assumptions. For example, Romney’s 11-point deficit among all women masked a 14-point winning percentage among white women, 56 percent to 42 percent. Indeed, 45 percent of all Romney voters were white women; 44 percent were white men. (Women edged out men because there were more women in the electorate.)

Other big differences among women are tied to marital status: Single women vote more Democratic; married women vote more Republican. And women with children vote more Republican than those without kids.

The observation that the women’s vote is complicated is not new. Indeed, this column has focused a great deal on the cohort of Walmart moms, a group making up 12 to 17 percent of the electorate that has been explored in great detail by Alex Bratty and Neil Newhouse of the Republican polling firm Public Opinion Strategies (who coined the phrase) and Democratic pollster Margie Omero of Purple Strategies (formerly Momentum Analysis).

Another top Republican pollster, Linda DiVall of American Viewpoint, recently released a report (with an interesting interactive graphic at www.amview.com) segmenting women into 11 groups, or typologies. Based on a national survey, the graphic shows how these typologies differ on Obama’s job approval, the generic congressional ballot test, favorable/unfavorable ratings for the two major parties, the right direction/wrong track numbers, and top economic concerns.In DiVall’s analysis, the three groups most disposed to voting Democratic were “Millennials,” who preferred Democrats for Congress by 64 percent to 19 percent; “Married Moderates,” who voted Democratic 57 percent to 20 percent; and “Single Professionals,” who sided with Democrats 57 percent to 28 percent. Democrats captured 57 percent or more among each group on the generic ballot test. Not far behind were “Socal-Media Mavens,” who use Facebook more than once a day; they backed Democrats 53 percent to 26 percent on the generic ballot test.

Most supportive of Republicans were “The Disenchanted,” who are pessimistic about the direction of the country and disapproving of both Obama and their own members of Congress; and “Medicare Women,” ages 65 and older. Those groups favored Republicans 60 percent to 12 percent, and 51 percent to 27 percent, respectively. In between were “Married Homemakers,” “Baby Boomers,” “Walmart Women,” “Generation X,” and “Suburban Women.”

Pointing to the finding that 46 percent of women thought government was trying to do too much, versus just 39 percent who thought government should do more, DiVall argues that Republicans have opportunities to improve their standing with female voters if they use better messages for each group and run more female candidates, a point that Republican strategists have been talking up in recent weeks.

DiVall and other pollsters who have been waging a campaign within the party for better messaging since the election (and many before then) seem to have won over many GOP congressional leaders but are encountering strong resistance from the rank and file. If Republicans want to appeal only to older white men, their rhetoric would not have to change one iota. But with the makeup of the electorate changing dramatically, it’s hard to avoid the conclusion that the Republican Party has an unsustainable business model.