Inside Washington's Headlines

by Ken Feltman

Every election, the best analytical minds that the parties can find scour the maps and voter records to determine how the voters will vote. Recently, the maps have been getting a workout.

In 2004 the question was whether John Kerry could change the color of Ohio from red to blue and, in so doing, win the election. He could not despite the fact that Democrats registered hundreds of thousands of new voters and brought them out to vote Democratic in November.

Quietly, away from the media centers in the urban (read Democratic) areas, the Republicans registered more voters and got them out. Despite the election night media predictions that Ohio was going from red to blue, Bush carried Ohio.

Similar things happened in other contested states. A key Kerry operative summed it up early on election night: 'There are several millions more people voting and we don’t know who they are.' But the Bush people knew. There were no major surprises at Bush headquarters.

Under the radar

The Republicans and the Democrats used the same maps and statistics but Republicans found votes where votes had not existed in 2000, 1996, perhaps not even in 1992 or 1988. The Republicans did not find newcomers who had just moved in; they found long-time residents who had dropped out, neglecting to vote for one, two, even several elections. Those disengaged voters were analyzed and when they were inclined to be Republican voters, if only they would vote, they were given good reasons to vote. They went to the polls. The Democrats did not see them coming.

In fact, they had been coming for a long time. They were heard from in 1993 in New Jersey when Christine Todd Whitman was running from far behind for governor. No way, the party bosses said, could a pro-choice Republican win in New Jersey. But she did. She crept back, precinct by precinct, picking up a few votes here, several there, till she put together enough to squeeze by. How did she do it? The same observers who missed Whitman’s comeback soon had various theories to explain it. Several theories made a little sense and all theories were at least partly valid. But few really understood where Whitman got the votes that put her over the top.

A small, independent political organization worked to figure out if there were people in New Jersey who were not motivated to vote but could be attracted to make the effort to vote for Whitman. They looked at the maps, the statistics, the voter rolls. They looked at civic and labor groups, at educational and business groups. Finally, they found their voters. The key voters had been there all the time, but hidden in various different subsets of voters and lots more non-voters, clustered according to religion, ethnicity, issues, sex - and marital status.

As the campaign chugged toward November, the independent organization targeted the chosen group. Messages went directly to those individuals, usually through an old-fashioned method: letters delivered by the U.S. Postal Service. They went to thousands and thousands of occasional voters and more non-voters. The recipients were largely Roman Catholic, all were women, most were from blue collar families, all were married or widowed with children.

The Democrats figured that these women were likely Democratic voters if they voted at all. They fit the normal definition of a Democrat in New Jersey. Their husbands, brothers and fathers were Democrats. They were reliably on the Democratic side of the issues, with one exception: They were good church members and, therefore, figured to be pro-life in a pro-choice party. Democrats easily carried the areas where most of these women lived so their participation in the election was not really all that important. Anyway, that’s what the Democratic leadership thought.

The Republicans figured it was a good thing that more of the women did not vote. Best ignore them, the Republicans decided. They were the typical definition of Democratic voters in New Jersey, except for that presumed pro-life position, which would not help a pro-choice Republican woman candidate. Anyway, that’s what the Republican leadership thought.

They were all wrong. The independent organization did some homework. Overwhelmingly, the women were pro-choice. A simple message went out:

All your life you have been dominated by strong men. Your father, your brothers, your husband, the priest. Isn’t it time you voted for yourself? Whitman is pro-choice, too. Vote for yourself. Vote for Whitman.

In thousands of New Jersey homes, women read the letter and decided to vote. On election night, the two parties were surprised by the increased turnout in many working class, Catholic neighborhoods. They were surprised again at the reduced margins recorded for Democratic candidates in those areas - reduced margins that tipped things just enough to Whitman.

Disaffected voters return to the polls

Two years ago, in that part of Ohio that is more like West Virginia or Kentucky than Ohio, thousand of disaffected voters were convinced to come back to the polls. Years ago, they had abandoned the Democratic Party over social issues. But they had not embraced the Republican Party as disaffected Democrats did in the deep South. Again, the common-denominator issue was found and the messages went out. Ohio stayed red despite one of the most remarkable voter-registration efforts that the Democrats ever mobilized.

In both New Jersey and Ohio, you can sort the voters out many different ways. The women who ended up voting for Whitman were very like their husbands on the issues, including abortion rights. They just voted, in that one election, for someone that they identified with.
 

From this beginning, a new theory developed. Actually, it is a recycled old theory: The most reliable predictor of voting behavior is not an issue. It is not income, not sex. It is not union membership, educational attainment, religion. It is martial status.

Unmarried New Jersey women who matched every other pro-Whitman characteristic did not turn out for Whitman. These single women found reasons other than Whitman’s pro-choice stance to vote reliably Democratic. Unmarried Ohio voters who seemed to be like their married counterparts did not vote Republican. After all the other demographic analyses were complete, marital status was an excellent indicator of election day behavior.

Across the country, red states and blue states fit the same pattern. Urban areas are blue and rural areas (and most suburban areas) are red. The blue urban areas are often overwhelmingly blue and the rural/suburban areas are usually less overwhelmingly red. Of course, many areas within blue states are red, and vice versa.

This leads many to conclude that population density is a good predictor of voting patterns. For example, if you look at counties across the U.S., you find that counties won by Kerry in 2004 had about seven times as many inhabitants, on average, as counties carried by Bush. There are simply a lot more rural and suburban counties.

Where you live and where you worship

Does this mean that the elections are a battle between the cities and the countryside? Some observers think so but a closer analysis leads some to a different conclusion. Another way to look at these same counties is to consider that urban areas have more unmarried people while rural and suburban areas are populated by more married people. In 2004, Bush carried 97 of the 100 counties with the highest percentage of married people heading households. Kerry won 90 of the 100 counties with the lowest percentage of married households. More urban households are headed by African-Americans, who tend, under similar circumstances, to be married less frequently than comparable whites.

Married householders in both red and blue areas tend to behave like other married voters and not like their unmarried neighbors. Unmarried householders - in red and in blue areas - behave more like other unmarried householders in other areas and less like their married neighbors. So place of residence may not be as important as marital status.
 

Other observers believe that religion is a key factor in voting behavior. To understand the role of religion, we must understand that churches are social centers in rural areas. People go to church to meet neighbors and gossip as well as to worship. In urban areas, the local coffee shop may serve the same social purpose. Beside, urban churches may be older, more run down, darker and less appealing as places to congregate than they used to be. Comparing church-goers in red and blue areas leads to the conclusion that church attendance is not as big a factor as marital status in determining how those individuals will vote.

The political parties, at some level, understand all this and adjust their platforms to the expectations of their constituencies. But the parties are organized around precincts, districts, cities, towns, counties and states. Geography is useful for dividing the territory into workable units but not as useful for dividing the voters into cohesive categories. That is where the new political organizations come in. We have all heard of them. The Swift Boat ads in 2004 were the product of one such group.

Caveat voter

Most of these organizations are not bound by the parties or the candidates. They operate independently. No wonder both parties want to restrict these organizations: They threaten the parties. They reach right out to the voters without the parties or candidates having a say. The parties have learned, of course, that these groups can have tremendous power and influence. That is why the parties are adopting their tactics of looking beyond the precinct voter lists to find their supporters. Whitman’s unexpected supporters were found that way. Many of the unanticipated Ohio voters had not voted in years but were found and motivated to vote for Bush.

The 2006 battleground is being decided now as smart technical people try to find this year’s best way to reach the formerly unreachable voters. So far, the Republicans seem to have the better effort underway. Will it be enough? Certainly, the independent groups - liberal and conservative -will be active and influential. Will they make the difference?

If you are married, expect to see messages designed to reassure you. If you are single, expect to see different messages - perhaps from the same political group that is reassuring your married neighbors with a very different message.

Caveat voter.

Please click here to review previous issues.