According to Julie

This article was published in Argument 3-2008. It is based on a midterm exam in the subject "American Presidential Elections". The exam question was: "Who will win the Democratic primary elections in Ohio and Texas - and why?" I predicted the results - that's how predictable American politics can be.

I took a break from studying for my midterm in "American Presidential Elections" to read the Norwegian newspaper Aftenposten. Their website about the American elections was depressing: rumors about Obama using cocaine, and Obama's wife described as the sexiest woman in American politics. Fortunately Aftenposten writes good articles about the elections too, but they don't write much about the candidates' politics. It's not Aftenposten's fault if Norwegians get the idea that American presidential elections are all about rumors, polls, support from the right people, dramatic media coverage, scandal and only the vaguest of political statements. That's the way it is.

The study of American elections involves as much statistics, media studies and sociology as it does political science. Scholars predict election results based on average age and average income in states. This is called "electoral sociology". Norwegian media publish humorous articles on the typical SV (Socialist Left) or Høyre (Right) voter, but this kind of knowledge is essential for the study of American politics. Some demographic "laws" have turned out to be myths. (For example, it's not true that Latin Americans always vote for Hillary Clinton.) Politicians still pay specialists to tell them which groups of people support them and where these people live. Add changes in constituencies and varying election rules from state to state and party to party, and commenting politics in the US turns out to be all about numbers.

Every vote counts

According to Steven Ekovich, professor in Political Science and History at The American University of Paris, Americans choose a President according to these criteria: The individual candidate's personality is the most important, party identification comes second, and political views and issues are third. For Ekovich, who describes himself as a "poll junkie", no day is complete without the newest polls, election results and political commentary. But every vote counts, whether it comes from a political expert or someone who votes by gut feeling and tradition. And most voters in the US belong to the second category.

The importance of personality is not surprising, given the President's political and symbolic power. It explains how important it is for Americans what Obama's pastor thinks, and how common it is for candidates' families to become public figures. American presidents are not just elected representatives; they are symbols of the American people.

Voting like their parents

Isn't party identification an expression of political views? Not necessarily. The ideological differences are not as clear in American politics as they are in European politics. Both the major parties are on the right, and the differences between representatives within each party can be just as important as the differences between the parties. Both the Republican and the Democratic Party are coalitions of local parties and state parties with varying views. Presidential candidates should not provoke the different factions of their own party, and this explains how vague the candidates often are about where they really stand. Because of this, identification with one or the other party is often a result of tradition and demographics.

Over a third of Americans are independents. In other words, they are not party members. Even so, almost all identify with either the Democrats or the Republicans, long before they are old enough to vote. Most vote like their parents. American election scholars talk about "red states" and "blue states", where one party always wins. With only two effective parties, switching to another party is far more dramatic than it is if there are more parties - it means you're going over to "the other side".

Demographic laws

Identification with individual candidates is obviously not as stable as party identification. You can be born into a Democratic family, but you're not born knowing who Barack Obama is. Even so, there's a lot of electoral sociology within parties.

In this year's primary elections there are only minor political differences between Hillary Clinton and Barack Obama. But political views do not decide the election. Style, personality and image do. Here the candidates differ, and this determines who will vote for them.

In general, women, those over 65, those with low income, and people without college degrees, vote for Hillory Clinton rather than Barack Obama. This group has been called "Clinton's coalition". Obama also has a coalition: black, college-educated, rich, but liberal (as in not conservative), and independent. With one exception - Nevada - Obama has won all caucuses.

Between February 5th (Super Tuesday) and March 4th (primary elections in Ohio and Texas), Obama won all primary elections. The media used the word "momentum". Momentum happens when it looks like you're about to win. The more Obama won, the more likely it was that he would win even more. In a political system where there is only one winner - the opposite of our own proportional representation system - cheering for the winner is a good option. Some commentators wrote that talking about momentum was exaggeration - until the primary election in Wisconsin. After Wisconsin, articles where written with headlines like "It's Over" - many people believed Clinton was finished unless she won Ohio and Texas.

Predictable in Ohio and Texas

Wisconsin broke the demographic rules. Women, Americans without college educations, and members of the Democratic party voted for Obama. Among the groups where Obama usually finds support, he was even further ahead of Clinton than before.

The results in Ohio and Texas were no surprise, however. Clinton won Ohio by more than 10 percentage points. In Texas, where primaries are held both by primary and caucus, the results are complicated. Clinton got 50.9% of the votes, against Obama's 47.7%. He still won 99 delegates, while she only won 94.

Demographically Ohio is a Clinton state. Compared to the US as a whole, Ohio is older and poorer, with a lower percentage of women and people with college educations. Obama's coalition is not strong in Ohio. However, there are a lot of students in Ohio, and young voters tend to support Obama. On the other hand, young voters tend to vote less. The rules for the primary in Ohio meant that many Ohio citizens had already voted when Obama won Wisconsin, weakening the impact of momentum. Ohio voted by primary, not caucus, which is also good for Clinton.

Texas was less certain, and the result itself turned out to be uncertain. Compared to the US as a whole, Texas has a lower percentage of white people and black people, and far more Latin Americans. One demographic "law" which has turned out to be a myth, is that Latin Americans vote for Clinton. Texas did well under Bill Clinton, and Clinton has strong ties to Texas, particularly among Latin Americans. On the other hand, even though there are fewer black people in Texas than in the US in general, the percentage is twice as high as in Wisconsin. Demographically Texas is similar to California (which Clinton won by 8 percentage points), but with a larger black population.

The voting rules in Texas are complicated, but they worked to Obama's advantage. The delegates were selected both through primary and caucus. Extra delegates were given to areas with a previous record of high voter turnout. These are the areas where Obama's traditional supporters live. Clinton's Latin American supporters generally live in areas without extra delegates. So Clinton could get more votes, even though Obama "won".

Texas is a Republican state. In Texas the wealthy and educated whites - who would have supported Obama - probably don't vote in the Democratic primary at all. And it will be almost impossible for the Democrats to win Texas in the actual presidential election. Ohio is one of the few states where the choice between a Democratic or a Republican candidate could go either way.

New rules?

Obama has mobilized new voters. He is popular among the young, but he has also gotten older Americans who have never voted before, to register as voters for the first time. That means he has changed the demographics of American voters. The old rules don't necessarily apply. When Americans choose their President, many of these new voters might not vote at all if voting for Obama is not an option.

"To draw a new political map, you need to believe that the demographics have changed," says Ekovich. "Obama does." This is one of the most important differences between him and Clinton. In other words: even when you describe the candidates' political views, you're talking about demographics.