One way to understand what happened on Election Day is to look at the vote and simple demographic characteristics, including an examination of the gender, marriage, God, and race gaps in the 2008 election.

Senior Fellow Karlyn Bowman

In a 2006 essay in the academic journal PS: Political Science & Politics, Laura Olson and John Green introduced the idea of "gapology," or the study of differences in voting behavior.

The authors argued that studying voting gaps was a "potent" way to understand elections. "Like batting averages," Olson and Green wrote, "such simple statistics offer the power of language in describing the political world." They described some of the biggest determinants of voting disparities in the 2004 presidential election: gender, marital status, religiosity and race. Here's a 2008 update on those influences.

The gender gap refers to the differences in the voting patterns of men and women. This gap emerged in the 1980 election, when Ronald Reagan did better among men than he did among women, and it has been with us ever since. In this election, Barack Obama won the women's vote decisively, 56 percent to 43 percent, and the male vote narrowly, 49 percent to 48 percent, for an overall gender gap of 12 points. Women are a larger share of the population and of the electorate than men (they were 53 percent of voters in 2008), so their tilt toward Democrats carries electoral weight. Obama improved on recent Democratic presidential candidates' performance among men, and this helped propel him to victory.

But the marriage gap (at 36 points in this election) dwarfs the gender gap. Two-thirds of voters told the exit pollsters they were married, and they voted for McCain 51 percent to 47 percent. Obama, though, did better among this large group of voters than recent Democratic presidential candidates. Unmarried voters (those who are single, divorced or widowed) were 34 percent of the electorate, and they voted for Obama over McCain by 65 percent to 33 percent. These unmarried voters are a growing share of the electorate.

It's important to note that the gender and marriage gaps also appear across groups. That is, younger women and younger men vote differently, married men and married women vote differently, and so on.

Political scientists used to study religion and voting through the prism of denomination. But today, religiosity, measured by worship attendance or lack thereof, is the more potent variable. A Catholic who goes to mass every week is closer politically to a Protestant or Jew who attends services weekly than that Catholic is to another Catholic who doesn't go to mass at all. Those who never attend church are a smaller slice of the electorate than those who attend weekly (16 percent vs. 40 percent). In 2008, voters who said they never attend church voted overwhelmingly for Obama, 67 percent to 30 percent, while those who attend weekly voted for McCain, by 55 percent to 43 percent. That makes the so-called God gap 49 points.

Another big gap this year, naturally, involved race. Whites were 74 percent of the electorate in 2008, down from 90 percent in 1976. They voted for McCain by a margin of 55 percent to 43 percent. Obama did better among them, however, than his recent predecessors. African-Americans increased their representation from about 10 percent to 13 percent of the electorate. Given the slow growth of African-American population, this seemingly small gain is actually quite impressive. This group's decisive vote for Obama (95 percent to 4 percent) was its strongest showing for a Democratic presidential candidate, and it helped him win. The black-white race gap in this election was a whopping 103 points.

There are other gaps in our politics--like generation, income and place of residence--and the size of all gaps changes over time. Some, like the gender and God gaps, are relatively new. As Olson and Green noted, gapology will continue to fascinate us because it connects "something of compelling importance (... who was elected president) with some key facts of everyday life (people's most obvious characteristics)." These correlations always excited pollsters and pundits, but the analysis of gaps in this year's election will prove to have long-lasting repercussions for all future presidential candidates.

Karlyn Bowman is a senior fellow at AEI.