A Look at Religious Voters in the 2008 Election |
Site last published: 01/06/10
A Look at Religious Voters in the 2008 Election
February 10,2009
Pew Research Center Publications
Some of the nation's leading journalists gathered in Key West, Fla., in December 2008 for the Pew Research Center's Forum on Religion and Public Life's biannual Faith Angle Conference to look at the impact of religious voters in the 2008 election.
John Green, a senior fellow in religion and American politics at the Pew Forum, discussed how a small change overall in voting behavior among religious groups had a big impact at the ballot box. Green said that the Democrats and Barack Obama made their largest gains among minority religious groups but that Obama made only modest gains among white Christian groups. Although these shifts were enough to put Obama in the White House, they did not change the overall structure of the faith-based vote compared with 2004.
Anna Greenberg, senior vice president for Greenberg Quinlan Rosner Research, said that Obama's faith was the main religious narrative of the campaign. But that debate wasn't necessarily about religion itself, she said, but a stand-in for a conversation about Obama -- who he was, where he came from, what values he represented. Another ongoing theme was whether Obama could win votes among the more-conservative religious groups. Greenberg cited polling data showing that younger evangelicals were more likely to support Obama than older evangelicals and that the under-30 set was more progressive on the issues of climate change and gay marriage.
Speakers:
John Green, Senior Fellow in Religion and American Politics, Pew Forum on Religion & Public Life
Anna Greenberg, Senior Vice President, Greenberg Quinlan Rosner Research
Moderator:
Michael Cromartie, Vice President, Ethics and Public Policy Center; Senior Advisor, Pew Forum on Religion & Public Life
In the following edited excerpt, ellipses have been omitted to facilitate reading. Read the full transcript, including follow-up discussion, at pewforum.org.
John Green
JOHN GREEN: The Pew Research Center for the People & the Press typically does a survey the weekend before the election, and then right after the election. I combined those two surveys into a single data set.1 So we have people right before they voted and right after they voted, giving us 3,000 to 4,000 respondents to work with. Among other things, the combined surveys allow us to look at some small religious groups, which were quite interesting in 2008. They also allow us to look at some other questions not asked in the exit polls. Yet another advantage is that, because our colleagues at the Pew Research Center do these kinds of surveys regularly, we can compare 2008 to a similar combination of pre-election and post-election surveys conducted in 2004.
So, how did the faith factor operate in 2008? There are three points worth making. First is that the basic structure of faith-based politics in the United States was very similar in 2008 and 2004. There was remarkably little change. The change was quite important, of course, but the basic structure was very similar. Second, the Democrats and Barack Obama made their largest gains among various religious "minorities" -- groups that can be described as minorities either in ethnic, racial or religious terms -- and we'll look at some of those groups in just a moment. Third, the Democrats made only modest gains among white Christian groups, and we'll talk about that in a little bit of detail too.
Effect of religion on voting
Before we get into the numbers, though, it's worth just spending a moment talking about the different pieces of the structure of faith-based politics in the United States, a structure that's been in operation for more than 20 years. There was a lot of speculation this year that this structure would change. In fact, it didn't change very much, but it may change in the future. There are three basic pieces to this structure. Religious affiliation -- the religious communities to which people belong -- is a very important part of the structure of faith-based politics. As in the past, religious affiliation is very closely linked to ethnicity and race. It is worth spending a moment on this linkage in the form of "ethno-religious" groups.
Ethno-religious groups have been important in American politics from the beginning of the republic and some examples are well-known, such as Irish Catholics, Scottish Presbyterians and German Jews. Some of those groups are still important today, but new groups have come on the scene. Thus we have a new version of an old story, with new ethno-religious groups such as Mexican Catholics, Korean Presbyterians and Arab Muslims, where religious affiliation, ethnicity and race are very closely tied together.
But in recent times, religiosity -- typically measured in these surveys by frequency of worship attendance -- has created new groups within religious affiliations. So whereas once we could talk about "the white Catholic vote," which was really the European Catholic vote, we really can't do that anymore because there are huge differences between regular mass-attending and less-observant Roman Catholics. We'll see some examples of this pattern in the structure.
The effect of religiosity is the strongest in white Christian groups, but you can see it in almost every religious group. When we get the full 2008 exit poll data, I think it would be fun to look and see if these differences by religious attendance show up in all the different groups. In 2004, they did. There wasn't a single religious group where there weren't political differences by worship attendance. But the really big differences are among white Christian groups. Anyway, we can define the basic structure of faith-based politics with affiliation, ethnicity/race and attendance.
Let's look at what the structure looked like in 2008. This picture is a bar graph that lists the key religious groups in order of the Democratic vote -- the blue bars are the vote for Obama; the red bars are the vote for McCain. The most Democratic group is at the top of the graph and the least Democratic group, or the most Republican group, is at the bottom. One can see the structure of the faith-based vote at a glance. This picture provides a sense of how polarized American religion was, even in 2008. There were strong Democratic groups, and there were strong Republican groups.
Source: Analysis of aggregated election weekend and post-election callback surveys conducted by the Pew Research Center for the People & the Press.
The next bar graph is for 2004. Note how similar the 2008 and 2004 graphs are. There are some differences, of course, but the basic structure was largely intact. What this similarity says to me as a political scientist is that these differences based on religious affiliation and attendance are very deeply embedded in American politics.
Source: Analysis of aggregated election weekend and post-election callback surveys conducted by the Pew Research Center for the People & the Press.
Let's just look at some of the key groups in 2008. Toward the top of the chart, virtually all of the strong Obama groups have this character of being minorities in one form or another. Of course, the first group, black Protestants, has been a strong Democratic group for a long time. In these data they voted nearly 100% for Obama. Of course, part of the black Protestants' story in 2008 was not just the Democratic margin but the higher turnout, so this is a very important group for Obama. But if you go down the graph, you see Jews, and then a composite group of "Hispanic and other minority Catholics" -- this group includes other racial categories such as Asians and individuals of mixed race. It's interesting because this group was very Democratic in 2008. In the past they've tended to be Democratic, but they've moved in a more Democratic direction.
Next is the composite category of "Other (non-Christian) faith" -- that's the combination of Hindus, Buddhists, Muslims and other non-Christian groups. This category was also a very Democratic group in 2008.
Then we come to another very Democratic group: seculars. These are people without a religious affiliation and who gave us no indication that they had any kind of religious belief or behavior. If we skip down a little farther, we come to a related group: the unaffiliated believers. These are people who tell us they don't have a religious affiliation but show some indication of having religious beliefs and behaviors. They were also a strongly Democratic group, but notice how different they are from the pure seculars, who were even more Democratic.
Then finally we come to a white Christian category, the less-observant white Catholics. These people reported attending worship less than once a week. This group has tended to vote Democratic in the past, but in 2008 they were even more Democratic, though not like the religious minorities. It's worth looking down the graph and noting how different this group was from Catholics who reported attending mass once a week or more -- the less-observant were markedly more Democratic. This difference is a good example of the impact of worship attendance on the vote.
Then we come to a minority group that was more evenly divided. In these data, the composite group of "Hispanic and other minority Protestants" nearly broke evenly between McCain and Obama. But, as some of you may remember, in 2004 Hispanic Protestants voted Republican. So this was a very important shift into the Democratic column in 2008.
Next we get into the groups that voted, on balance, Republican -- near the bottom of the chart. Less-observant white mainline Protestants voted for McCain by a slight majority, while weekly attending white mainline Protestants backed McCain by a larger majority. Weekly attending white Catholics also voted Republican, by a still larger majority. And then at the very bottom of the chart are white evangelical Protestants, where both the less-observant and weekly attenders voted solidly Republican. But note that the weekly attenders were markedly more Republican. Indeed, weekly attending white evangelical Protestants were the strongest Republican group in 2008, as in 2004.
Where race was a factor
During the campaign, there was much talk about how race might matter at the polls. Many people worried that race might undermine the Obama campaign. While those worries didn't materialize, these graphs suggest that race had an impact at the ballot box in a profound way. Racial and ethnic minorities voted more Democratic, and whites voted more Republican in 2008. This pattern is not entirely new -- as one can see in the 2004 graph -- but these patterns were much sharper in 2008.
Now a lot of the minority groups that I've identified are small. For instance, the category of "Hispanic and other minority Protestants" made up about 4% of the 2008 electorate. That's about twice the size of the Jewish community. So 4% of the electorate's not a trivial group, given the enormous diversity of American religion. And if you add up all these small ethno-religious groups, they account for a significant bloc of voters.
One of the reasons that I wanted to highlight this pattern is that there is constant talk about the effect of immigration in the United States and the fact that the United States is becoming much more diverse in ethnic and racial terms. But oftentimes we don't see an example of the practical impact of this new diversity -- and these graphs provide a good example. I think these patterns had some special things to do with Barack Obama, but nonetheless, it does show how increased diversity can ultimately matter at the ballot box. It's interesting that all these minority groups lined up on the same side politically in 2008, whereas in 2004 they didn't. Maybe they will line up on the same side in future elections, but maybe they will divide up between the major political parties once again.
Changes between 2008 and 2004
Let's move on and talk a little bit about some of the changes between 2004 and 2008. Overall, the biggest changes in the Democratic vote came among the religious minorities, including Hispanic Catholics, Hispanic Protestants, and other minority Catholics and Protestants. There were some changes among black Protestants, but of course, that group was so Democratic to begin with, there wasn't a lot of movement that could occur. Obama also made gains among Jews, the "other faiths" category and the composite categories of "Hispanics and other minorities" among Catholics and Protestants.
A lot of attention was focused in the campaign on the large, white Christian groups -- evangelical and mainline Protestants and Catholics -- and these data show some interesting stories. To begin with, there was essentially no change in the vote of regular worship-attending white evangelical Protestants, the core of what sometimes is called "the religious right" -- one of the strongest Republican voting groups. I don't think the Obama campaign had any expectation that they would make inroads in this group, but there were quite a few commentators who thought that that might happen.
There was, however, some change in the evangelical community, and it occurred mostly among less-observant evangelical Protestants. Among mainline Protestants there was an interesting pattern. In the exit polls, there was essentially no change among white mainline Protestants. But the data presented here suggest that there were some changes within this large religious community. For instance, Obama may have made some gains among regular worship-attending mainline Protestants.. And it may very well be that a lot of the efforts to mobilize the religious vote paid off in that particular community. However, these data show essentially no change among the less-observant mainline Protestants, who were evenly divided. This group was where one might have expected bigger Democratic gains.
In the exit polls, white Catholics stayed on the Republican side of the ledger, and the most observant Catholics moved in a Republican direction by a few percentage points. These data suggest there's a bit of a polarization among white Catholics, with the regular mass-attenders moving more Republican than they were in 2004, but the less-regular mass-attenders moving more Democratic.
Looking at the Republican side, George W. Bush generally did better than John McCain in many religious groups. But where McCain had his biggest problem was among religious minorities of various kinds.
Why was the change so modest?
Given the focus in the campaign on religion and the efforts to mobilize religious groups, why was there so little change? There are a couple of ways to answer this question.
First is that it's important to remember that, in terms of the overall vote, the change between 2004 and 2008 actually wasn't that large. John Kerry got approximately 49% of the two-party vote in 2004, and Barack Obama got about 53% of the two-party vote in 2008. So there's a shift of roughly 4 percentage points. Barack Obama won a solid victory, particularly compared with the 2000 presidential election. But it shouldn't surprise us that changes were not that large within many religious groups because there just wasn't that much overall change.
Another thing to remember is that many religious groups are strongly partisan these days. They're deeply embedded into the party coalitions, with each party having strong religious constituencies, and some groups in the middle and up for grabs. We know that partisanship is a very strong predictor of the vote. And, as many of you know, between 2000 and 2004 there were some changes in partisanship. The Republican brand name took a big hit and that influenced the vote of religious groups as well.
The impact of partisanship can be seen in the following graph. Here some of the smaller religious groups were combined for ease of presentation, and they are arrayed from the group that voted least for Obama (weekly attending white evangelical Protestants) to the group that voted most for Obama (black Protestants). The solid line shows the Obama vote, and it rises steadily across the chart. But the dashed line is the interesting one: It's the percentage of each of these groups that identified as Democrats. Note that the solid and dashed lines are basically the same line. So part of what was going on in 2008 is that the campaigns were mobilizing their partisans, and even though this wasn't the kind of base-mobilization election that 2004 was, partisanship still mattered a great deal.
Source: Analysis of aggregated election weekend and post-election callback surveys conducted by the Pew Research Center for the People & the Press.
Partisanship develops very slowly over time. It took about 30 years to get the faith-based structure we're talking about, so maybe it's not surprising that it didn't go away in a single election. But of course, over the next 30 years or so, we may see some important changes, and we may look back to this election and see some of the modest alterations as the beginnings of important trends.
A final explanation for the modest change in 2008 may be the patterns of faith-based mobilization by the campaigns. In 2008 as in 2004, the People & the Press asked survey respondents about campaign contacts within their congregations. The levels of reported contacts were down in 2008 for most religious communities -- an interesting pattern given the focus of the campaigns. Interestingly, the least decline in reported contacts was for religious minorities that strongly backed Obama. And among white Christian groups, weekly worship-attenders still reported higher contact rates, and these were the groups that by and large stuck with the Republicans and voted for McCain.
[T]here's another question that the Pew Research Center asks regularly that is, in some ways, even more interesting: the likelihood of presidential success. The results are shown in a final bar chart, where the red bars represent the percentage of each group that thought Bush was going to be successful in office, back in 2004, and the blue bars the percentage that thought Obama will be successful in 2008.
Source: Analysis of aggregated election weekend and post-election callback surveys conducted by the Pew Research Center for the People & the Press.
Results for Bush based on interviews with 62 black Protestants, 71 minority Christians and 95 weekly attending white mainline Protestants. Results for Obama based on interviews with 95 members of non-Christian faiths. All other results based on interviews with at least 100 respondents.
Look at the blue bars. Every religious category has a majority that thinks that Obama will be successful in his first term. That's interesting because it suggests that President Obama has an opening, even with religious groups that didn't vote for him. Now, looking at the red bars, compare that with what happened in 2004. Notice that many of the Democratic groups didn't think that Bush was going to have a successful second term. Maybe they were onto something, given the decline of Bush's approval after the 2004 election.
This chart does suggest that there may have been a bit of a change between 2004 and 2008. So behind the modest shift in the voting behavior of religious groups, there may be a larger change lurking. And these possibilities may have to do with how successful President Obama and his Democratic colleagues actually turn out to be in office. Campaigns are important things to voters, but they're not nearly as important as the performance of the government itself.
Anna Greenberg
ANNA GREENBERG: I want to focus more on looking at individual groups, but also, what the religious narrative was and how it played out in this election. [T]he Pew Forum did an excellent analysis of the coverage of religion in the election. Their analysis showed what I think we all know instinctively in our gut, that the main religious narrative of this election was about Barack Obama's faith. That came out in a lot of different ways: the question of whether or not he was a Muslim or a Christian, his middle name, etc., but also the Jeremiah Wright conversation, which started during the primaries and extended throughout. Even though John McCain didn't advertise on Jeremiah Wright, other groups did, and certainly it was part of the discussion.
I think that the whole question of Barack Obama's faith and his affiliation raised lots of questions for lots of voters. It wasn't necessarily a debate about religion itself, but it was a stand-in for a conversation about Barack Obama. Who is he? Where does he come from, as John McCain said. What values does he represent? Is he patriotic? There's a whole set of questions around this question of his affiliation -- of his faith -- and that's on the more negative side. On the more positive side, would he have the ability to reach out and would he be able to make inroads among white evangelicals? Would he be able to be a unifier? Could he bring people together? Could he reach across religious lines and break down some of the trends that John was showing us?
He actually had a very significant religious outreach program -- not as extensive as the kind of religious outreach that Rove and Bush did in 2002 and 2004, but certainly more extensive than what Democrats had done in the past.
So what's interesting about all of this is that, first of all, it's very hard for Democratic candidates to escape stereotypes about their religiosity. In a survey I did with Religion & Ethics NewsWeekly and the United Nations Foundation, we asked the question: Which candidate do you think is more religious? I think we all know, objectively, that Barack Obama was the more religious candidate. And actually most of the Democrats -- if you look at Hillary Clinton -- were more religious than John McCain. What's interesting is that we asked this question in September, so this was well after conversations about belonging to the Trinity Church for 20 years. It's after Saddleback2, so it's after a whole lot of conversation about Barack Obama and his religious affiliation, his religious participation.
It's not by a huge margin, but it's actually beyond the margin of error -- 32% think that McCain was the most religious candidate, 28% Barack Obama, another third didn't know. So it's very hard for Democrats ever to be perceived as the more religious. [W]hat's really interesting is that only 46% of Democrats said that Barack Obama was the more religious candidate. So even among Democrats, the stereotypes about Democrats and religion are pretty strong.
The Pew Research Center asked: Do you happen to know what religion Barack Obama is? And 57% said he was Christian, 12% thought he was Muslim, and the rest -- I want to know who that 1% who think he's Jewish is -- I don't understand what that's about. (Laughter.) Among the people who thought that Barack Obama was a Muslim, there were an equal number of Democrats and Republicans; this was not necessarily a partisan issue.
We did a lot of focus groups with people who had voted for Hillary Clinton in the primary but were not voting for Barack Obama in the general and with people who called themselves Democrats but weren't sure what they were going to do or were voting for McCain. They all ended up being sort of the same people.
In these focus groups, this whole conversation about whether or not Barack Obama was a Muslim was a very big piece of it. Now people aren't necessarily rational when they have these conversations, so in the same sentence they would say: "I think he's a Muslim" and "I can't believe he belonged to that church for 20 years." You're trying to figure out how they can reconcile those two pieces of information -- that he's both a Muslim and a member of this radical Christian church. It was my belief that this whole question of whether or not he was a Muslim, on its face, was an issue for some people but also was a stand-in for race.
One of the things we also spent a lot of time trying to figure out was how this whole so-called Bradley effect -- I'm so glad that conversation is over; I never want to have it again -- but we took it seriously. When it was clear that Obama was going to be the nominee, as early as February in my view, we started doing some real research on this whole question of, how do you measure the impact of race?
It turns out if you just ask directly -- this isn't surprising -- does Barack Obama's race matter to you? -- that's not a very good way to measure it. In fact, what you found was that for people who said that it did, they were more likely to vote for Obama. They were younger; it was a positive. But there were other kinds of questions that could be a stand-in for race, and one of them was this idea that Barack Obama was too close to extremists. Actually, that ended up being a pretty powerful predictor. Lots of other things predict how you're going to vote, including partisanship and church attendance, but that kind of question in terms of the race issue -- and you can word it lots of different ways -- was actually a predictor of how people were going to vote.
On that question it was about a third Democrats, a third Republicans and a third independents; it wasn't correlated with partisanship. There were clearly a bunch of white Democratic voters for whom this whole question of race/Muslim/extremist -- however you want to ask it -- Who is he? Where does he come from? What are his values? -- was important. You had some people in these focus groups who thought he was literally like a Manchurian candidate, that he was going to get elected and then get into office and reveal himself as a Muslim and take over the country.
Now just to make the point, keeping in mind that half of these folks are Democrats, among people who thought that Obama was a Muslim, 51% would vote for McCain, 37% would vote for Obama, and among people who thought that Obama was a Christian, the numbers are flipped -- 52% would vote for Obama, 39% would vote for McCain.
One of the groups for whom this whole question of Barack Obama as a Muslim -- whether or not he's a Muslim but also extending to other questions around Israel and Palestinians and things he might have said about the Palestinians, etc. -- was very important to Jewish voters. I've done a fair amount of work on Jewish voters, though it's not easy to do because Jews are such a small percentage of the population and of the electorate. But I was on a panel that the National Jewish Democratic Council put on at the Democratic convention, and there were three different sources of data. There was the J Street survey, which was an Internet survey of Jewish voters; there was Gallup, which wasn't actually a separate survey of Jewish voters -- they do so much polling that they basically aggregate their surveys and pull out the Jewish interviews; and then we had our own internal databases, where I had about 1,500 Jews.
Winning the Jewish vote
Looking at all three sources of data going into the convention, it was pretty clear that Barack Obama was pretty seriously underperforming among Jewish voters. This trend line from Gallup shows you that in the summer, he's getting anywhere between 60 and 65% of the Jewish vote, which seems very high obviously. But relative to -- certainly since Clinton -- Democrats get around 80% -- 78, 80% -- of the Jewish vote. So that kind of number is actually pretty low. Now it doesn't really matter nationally, that difference between what previous candidates have gotten and what Obama was getting, but it actually mattered a lot in Florida.
I did not believe that Obama was going to win Florida, certainly not at the beginning of the year. But as more and more polls were coming out showing the race was competitive in Florida, we did our own poll -- a couple polls -- in Florida, and Obama was doing very well in places like Miami, and he was very competitive in the I-4 corridor, which includes Tampa and Orlando. That's always considered kind of the swing area of Florida, and he was actually tied with McCain, which was a very good result. But he was really underperforming in Broward County, Dade County, West Palm -- the Jewish parts of Florida.
It might have been a 20-point gap between the percentage who were calling themselves Democrats and the percentage who were voting for Obama. Now my grandmother and grandfather retired to Ft. Lauderdale. I've spent a lot of time in those condos in Broward County, and I know what those folks are like. And of course, there was "The Great Schlep" and that sort of thing. Did everyone see the Sarah Silverman video, "The Great Schlep?" I mean, nothing really happened. I don't think lots of Jews went down to Florida to convince their grandparents to vote for Obama, but it was funny. It was very Jewish.
But, anyway, so what was really fascinating about this was at this meeting at the Democratic Convention, it was like a mob of angry, Jewish state legislators from Florida, who were really angry at the Obama campaign because they felt like, we're going into these condos and we got nothing. We have to convince these old Jewish voters that Obama's okay, that he's not a Muslim, that he's not going to favor the Palestinians in some way and not support Israel. If you looked at the data that I had, it was really clear that the issue was with older Jews, which was not surprising; all of Obama's issues were with older voters, older white voters. Jews aren't different -- radically different -- in some ways than other voters; they're the same groups.
But this obviously mattered a great deal in Florida. I don't know exactly what the Obama campaign did. I think some of it was Sarah Palin; I think some of it was the debates. There were a variety of things that happened independent of the Obama campaign, though my understanding is that they put significant resources into outreach in this part of Florida. But what you can see nationally with the Gallup data is that over time Obama went from 62, 61% of the Jewish vote to 74% by the end of the election. And if you look at the exit polls, Obama got 78% of Jewish voters -- that's actually slightly better than Kerry did and it's about where Gore and Clinton were.
I think if you look at the polls going into Election Day and then you look at the exit polls, the one group where Obama really did worse than Kerry was with older white voters. If you look toward the end, particularly post-debates, which I think were enormously reassuring to voters who had concerns about him and who he was and all the questions that McCain asked, you started to see a slight movement. But when you look at the exit polls, he actually didn't do very well with older white voters, and what I don't know is, was that movement real? Was there a pull back on Election Day? If there was a Bradley effect, maybe it was just with that group. But the point is, Jewish voters are the one group where the campaign, in any event, was successfulin overcoming some of these doubts about him.
Obama's gains with younger evangelicals
So the other narrative in this election was, can Obama make inroads with more-conservative religious groups? He poured resources into that in his campaign. People had anecdotal evidence that young evangelicals might be more favorable to Obama than older. So when I did this poll for Religion & Ethics NewsWeekly, we did an over-sample of younger evangelicals.
Most of the national tracking polls had Obama at about 60% among under-30s, and he actually got 66%. So what we did with this poll was [a] multi-mode study. A certain number of interviews came from random digit dial, a certain percentage from the internet -- we didn't do the cell phone for a variety of reasons. We had, I think, the first real sample of young evangelicals and what we found was that, in fact, it was true that younger white evangelicals were more likely to support Obama than older white evangelicals.
Data from Religion & Ethics NewsWeekly/United Nations Foundation survey of 1000 adults and 400 young evangelicals, conducted September 2008 by GQR.
I was on a panel with GOP political strategist David Winston last week and he said, well, you know, when they get older, they'll vote more Republican because people when they get older, they tend to vote more Republican. And I said maybe, but the reason why people do that is they tend to get married and start going to church and these folks already go to church, so I don't know what's going to happen. The kinds of changes -- you get married, you have children, you pay taxes -- for younger evangelicals, they've already done a lot of the things that make you more conservative when you get older.
Even though this is from September, the 30% number is true across all three data sets that I looked at, so I feel pretty confident that Obama got 30% of younger -- this is under age 30 -- white evangelicals. So about 8 points better. [S]till, it's a very Republican group. Let's not get carried away here. I mean, 62% voted for McCain, and maybe even higher. But 30% for Obama, so 8 points higher [than among older evangelicals].
Data from Religion & Ethics NewsWeekly/United Nations Foundation survey of 1000 adults and 400 young evangelicals, conducted September 2008 by GQR.
Much more dramatic is the Republican brand among younger evangelicals. To measure favorability, we do something called a feeling thermometer, where you have to rate your feelings toward candidates, organizations or groups on a 0-to-100 scale. Anything above 50 is warm and anything below 50 is cool.
If you look at the white evangelicals over age 30 and at [feelings toward] George Bush, [you find] 57% positive, 29% negative. And you know what Bush's numbers are overall, so it's radically different. But look at where the younger ones are; he's got net negative ratings, 39% warm, 48% cool. You see a similar sort of pattern, though it's not as bad, on the Republican Party -- plus-45% positive rating, 64% warm, among older white evangelicals and only a net plus-5 positive rating for the Republican Party among younger evangelicals. So Obama gets 30%, but the Republican brand itself is really challenged among this group. I think this is going to be one of the more interesting things to track as this group gets older and becomes a bigger and bigger part of the white evangelical group in the electorate.
Then you can see, looking at a variety of issues, the differences between older and younger evangelicals. Again, we don't know if, when younger evangelicals age they'll get more conservative, but I was really struck by this slide, which is a question on gay marriage.
Data from Religion & Ethics NewsWeekly/United Nations Foundation survey of 1000 adults and 400 young evangelicals, conducted September 2008 by GQR.
What I would point out is that a majority of younger white evangelicals favor some kind of legal recognition of same-sex couples. So 26% say the right to marry, which I think is a very high number, and 32% say that they should be offered the same protections and benefits of marriage. Overall, 58% of younger white evangelicals think there should be some kind of legal recognition of marriage between same-sex couples. If you look at the older evangelicals, you can see that it's pretty dramatically different -- 9% for actual marriage, 37% for some kind of civil union.
If you look at the question of Iraq, again, the younger evangelicals are still different from other young people, but there is a slight majority for reducing the troops as opposed to staying the course, and that's different from older evangelicals. If you look at the question of global warming, the differences aren't huge, but still, the younger evangelicals are more likely to believe that global warming is kind of imminent and we need to take action now as opposed to a long-term threat that we can do something about later. But then, on this question of abortion, there's literally no difference between older and younger evangelicals. We asked a lot of different kinds of abortion questions in this survey. Every single way you asked the abortion question, you had the younger and the older -- and in fact, in some cases the younger being more conservative than older evangelicals on abortion.
Data from Religion & Ethics NewsWeekly/United Nations Foundation survey of 1000 adults and 400 young evangelicals, conducted September 2008 by GQR.
I wanted to end on one final discussion of where I think religion played an interesting role in this election, and that's ballot initiatives. In general, gay marriage and abortion -- not very big issues in this election nationally, and that often happens when you have elections dominated by the economy and the war in Iraq in 2006. But there were some pretty important ballot questions, particularly Prop 8 in California, which I've actually spent some time doing research on and which we can talk about during Q & A.
Evangelicals' impact on ballot initiatives
I'm throwing this out here as a kind of interesting, potentially provocative idea, but I don't have an explanation. It looks like, if you look at the different states where there were abortion and gay marriage initiatives, that evangelicals were key to the passage of gay marriage bans but not sufficient to pass the abortion restrictions. I don't know why, exactly, but I'm just going to show you because I think it's interesting. If you look at Prop 8, which is the gay marriage ban in California, [among] white evangelicals-- only 17% of the electorate in California -- 81% voted "yes." Everybody else was a slight majority for the "no" vote. Obviously that's gotten into a lot of discussion of African-Americans and other things we can talk about -- Hispanics -- and that's all part of the story. But the evangelical piece is key here, and the Mormon church and others did a huge amount of organizing work in the evangelical community around Prop 8.
If you look at Florida and the gay marriage ban, 81% [of evangelicals], and they are 25% of the electorate in Florida, voted for the gay marriage ban. Arkansas, it wasn't gay marriage -- it was an adoption measure -- and it's a lower number, 65%. But then when you compare the abortion ballot questions, especially in California, to the gay marriage ones, it's kind of interesting -- all the gay marriage ones passed and all the abortion ones failed. So, again, I'm not entirely sure what that's about, but it's interesting to think about.
But if you look at the abortion limits-- and the abortion limit in California was parental notification, which is actually something that gets pretty high levels of agreement with everybody-- it's interesting that you have about a 10-point drop in the number of evangelicals voting "yes" on parental notification compared with the gay marriage ban -- and obviously a much larger majority of non-evangelicals voting "no" on abortion limits.
South Dakota [the abortion limit on the ballot] was a redo of the one that failed two years ago. If you remember, the main reason [the earlier initiative] failed was there was no exception for health of the mother, and so they changed the language, but it still failed in South Dakota. Again, a smaller percentage of evangelicals, compared with the gay marriage bans -- 64% -- and a very big majority of the non-evangelicals voted against it.
I think it's something to investigate, both on the question of the differences in how evangelicals organized around these gay marriage bans in these different states, especially California, relative to the abortion bans, but also why there was a lower level of support among evangelicals on the abortion limitation measures than there was on the gay marriage bans. Again, I don't really have an answer for it, but I think it would be interesting to dig into that and think about it.
Some of the nation's leading journalists gathered in Key West, Fla., in December 2008 for the Pew Research Center's Forum on Religion and Public Life's biannual Faith Angle Conference to look at the impact of religious voters in the 2008 election.
John Green, a senior fellow in religion and American politics at the Pew Forum, discussed how a small change overall in voting behavior among religious groups had a big impact at the ballot box. Green said that the Democrats and Barack Obama made their largest gains among minority religious groups but that Obama made only modest gains among white Christian groups. Although these shifts were enough to put Obama in the White House, they did not change the overall structure of the faith-based vote compared with 2004.
Anna Greenberg, senior vice president for Greenberg Quinlan Rosner Research, said that Obama's faith was the main religious narrative of the campaign. But that debate wasn't necessarily about religion itself, she said, but a stand-in for a conversation about Obama -- who he was, where he came from, what values he represented. Another ongoing theme was whether Obama could win votes among the more-conservative religious groups. Greenberg cited polling data showing that younger evangelicals were more likely to support Obama than older evangelicals and that the under-30 set was more progressive on the issues of climate change and gay marriage.
Speakers:
John Green, Senior Fellow in Religion and American Politics, Pew Forum on Religion & Public Life
Anna Greenberg, Senior Vice President, Greenberg Quinlan Rosner Research
Moderator:
Michael Cromartie, Vice President, Ethics and Public Policy Center; Senior Advisor, Pew Forum on Religion & Public Life
In the following edited excerpt, ellipses have been omitted to facilitate reading. Read the full transcript, including follow-up discussion, at pewforum.org.
John Green
JOHN GREEN: The Pew Research Center for the People & the Press typically does a survey the weekend before the election, and then right after the election. I combined those two surveys into a single data set.1 So we have people right before they voted and right after they voted, giving us 3,000 to 4,000 respondents to work with. Among other things, the combined surveys allow us to look at some small religious groups, which were quite interesting in 2008. They also allow us to look at some other questions not asked in the exit polls. Yet another advantage is that, because our colleagues at the Pew Research Center do these kinds of surveys regularly, we can compare 2008 to a similar combination of pre-election and post-election surveys conducted in 2004.
So, how did the faith factor operate in 2008? There are three points worth making. First is that the basic structure of faith-based politics in the United States was very similar in 2008 and 2004. There was remarkably little change. The change was quite important, of course, but the basic structure was very similar. Second, the Democrats and Barack Obama made their largest gains among various religious "minorities" -- groups that can be described as minorities either in ethnic, racial or religious terms -- and we'll look at some of those groups in just a moment. Third, the Democrats made only modest gains among white Christian groups, and we'll talk about that in a little bit of detail too.
Effect of religion on voting
Before we get into the numbers, though, it's worth just spending a moment talking about the different pieces of the structure of faith-based politics in the United States, a structure that's been in operation for more than 20 years. There was a lot of speculation this year that this structure would change. In fact, it didn't change very much, but it may change in the future. There are three basic pieces to this structure. Religious affiliation -- the religious communities to which people belong -- is a very important part of the structure of faith-based politics. As in the past, religious affiliation is very closely linked to ethnicity and race. It is worth spending a moment on this linkage in the form of "ethno-religious" groups.
Ethno-religious groups have been important in American politics from the beginning of the republic and some examples are well-known, such as Irish Catholics, Scottish Presbyterians and German Jews. Some of those groups are still important today, but new groups have come on the scene. Thus we have a new version of an old story, with new ethno-religious groups such as Mexican Catholics, Korean Presbyterians and Arab Muslims, where religious affiliation, ethnicity and race are very closely tied together.
But in recent times, religiosity -- typically measured in these surveys by frequency of worship attendance -- has created new groups within religious affiliations. So whereas once we could talk about "the white Catholic vote," which was really the European Catholic vote, we really can't do that anymore because there are huge differences between regular mass-attending and less-observant Roman Catholics. We'll see some examples of this pattern in the structure.
The effect of religiosity is the strongest in white Christian groups, but you can see it in almost every religious group. When we get the full 2008 exit poll data, I think it would be fun to look and see if these differences by religious attendance show up in all the different groups. In 2004, they did. There wasn't a single religious group where there weren't political differences by worship attendance. But the really big differences are among white Christian groups. Anyway, we can define the basic structure of faith-based politics with affiliation, ethnicity/race and attendance.
Let's look at what the structure looked like in 2008. This picture is a bar graph that lists the key religious groups in order of the Democratic vote -- the blue bars are the vote for Obama; the red bars are the vote for McCain. The most Democratic group is at the top of the graph and the least Democratic group, or the most Republican group, is at the bottom. One can see the structure of the faith-based vote at a glance. This picture provides a sense of how polarized American religion was, even in 2008. There were strong Democratic groups, and there were strong Republican groups.
Source: Analysis of aggregated election weekend and post-election callback surveys conducted by the Pew Research Center for the People & the Press.
The next bar graph is for 2004. Note how similar the 2008 and 2004 graphs are. There are some differences, of course, but the basic structure was largely intact. What this similarity says to me as a political scientist is that these differences based on religious affiliation and attendance are very deeply embedded in American politics.
Source: Analysis of aggregated election weekend and post-election callback surveys conducted by the Pew Research Center for the People & the Press.
Let's just look at some of the key groups in 2008. Toward the top of the chart, virtually all of the strong Obama groups have this character of being minorities in one form or another. Of course, the first group, black Protestants, has been a strong Democratic group for a long time. In these data they voted nearly 100% for Obama. Of course, part of the black Protestants' story in 2008 was not just the Democratic margin but the higher turnout, so this is a very important group for Obama. But if you go down the graph, you see Jews, and then a composite group of "Hispanic and other minority Catholics" -- this group includes other racial categories such as Asians and individuals of mixed race. It's interesting because this group was very Democratic in 2008. In the past they've tended to be Democratic, but they've moved in a more Democratic direction.
Next is the composite category of "Other (non-Christian) faith" -- that's the combination of Hindus, Buddhists, Muslims and other non-Christian groups. This category was also a very Democratic group in 2008.
Then we come to another very Democratic group: seculars. These are people without a religious affiliation and who gave us no indication that they had any kind of religious belief or behavior. If we skip down a little farther, we come to a related group: the unaffiliated believers. These are people who tell us they don't have a religious affiliation but show some indication of having religious beliefs and behaviors. They were also a strongly Democratic group, but notice how different they are from the pure seculars, who were even more Democratic.
Then finally we come to a white Christian category, the less-observant white Catholics. These people reported attending worship less than once a week. This group has tended to vote Democratic in the past, but in 2008 they were even more Democratic, though not like the religious minorities. It's worth looking down the graph and noting how different this group was from Catholics who reported attending mass once a week or more -- the less-observant were markedly more Democratic. This difference is a good example of the impact of worship attendance on the vote.
Then we come to a minority group that was more evenly divided. In these data, the composite group of "Hispanic and other minority Protestants" nearly broke evenly between McCain and Obama. But, as some of you may remember, in 2004 Hispanic Protestants voted Republican. So this was a very important shift into the Democratic column in 2008.
Next we get into the groups that voted, on balance, Republican -- near the bottom of the chart. Less-observant white mainline Protestants voted for McCain by a slight majority, while weekly attending white mainline Protestants backed McCain by a larger majority. Weekly attending white Catholics also voted Republican, by a still larger majority. And then at the very bottom of the chart are white evangelical Protestants, where both the less-observant and weekly attenders voted solidly Republican. But note that the weekly attenders were markedly more Republican. Indeed, weekly attending white evangelical Protestants were the strongest Republican group in 2008, as in 2004.
Where race was a factor
During the campaign, there was much talk about how race might matter at the polls. Many people worried that race might undermine the Obama campaign. While those worries didn't materialize, these graphs suggest that race had an impact at the ballot box in a profound way. Racial and ethnic minorities voted more Democratic, and whites voted more Republican in 2008. This pattern is not entirely new -- as one can see in the 2004 graph -- but these patterns were much sharper in 2008.
Now a lot of the minority groups that I've identified are small. For instance, the category of "Hispanic and other minority Protestants" made up about 4% of the 2008 electorate. That's about twice the size of the Jewish community. So 4% of the electorate's not a trivial group, given the enormous diversity of American religion. And if you add up all these small ethno-religious groups, they account for a significant bloc of voters.
One of the reasons that I wanted to highlight this pattern is that there is constant talk about the effect of immigration in the United States and the fact that the United States is becoming much more diverse in ethnic and racial terms. But oftentimes we don't see an example of the practical impact of this new diversity -- and these graphs provide a good example. I think these patterns had some special things to do with Barack Obama, but nonetheless, it does show how increased diversity can ultimately matter at the ballot box. It's interesting that all these minority groups lined up on the same side politically in 2008, whereas in 2004 they didn't. Maybe they will line up on the same side in future elections, but maybe they will divide up between the major political parties once again.
Changes between 2008 and 2004
Let's move on and talk a little bit about some of the changes between 2004 and 2008. Overall, the biggest changes in the Democratic vote came among the religious minorities, including Hispanic Catholics, Hispanic Protestants, and other minority Catholics and Protestants. There were some changes among black Protestants, but of course, that group was so Democratic to begin with, there wasn't a lot of movement that could occur. Obama also made gains among Jews, the "other faiths" category and the composite categories of "Hispanics and other minorities" among Catholics and Protestants.
A lot of attention was focused in the campaign on the large, white Christian groups -- evangelical and mainline Protestants and Catholics -- and these data show some interesting stories. To begin with, there was essentially no change in the vote of regular worship-attending white evangelical Protestants, the core of what sometimes is called "the religious right" -- one of the strongest Republican voting groups. I don't think the Obama campaign had any expectation that they would make inroads in this group, but there were quite a few commentators who thought that that might happen.
There was, however, some change in the evangelical community, and it occurred mostly among less-observant evangelical Protestants. Among mainline Protestants there was an interesting pattern. In the exit polls, there was essentially no change among white mainline Protestants. But the data presented here suggest that there were some changes within this large religious community. For instance, Obama may have made some gains among regular worship-attending mainline Protestants.. And it may very well be that a lot of the efforts to mobilize the religious vote paid off in that particular community. However, these data show essentially no change among the less-observant mainline Protestants, who were evenly divided. This group was where one might have expected bigger Democratic gains.
In the exit polls, white Catholics stayed on the Republican side of the ledger, and the most observant Catholics moved in a Republican direction by a few percentage points. These data suggest there's a bit of a polarization among white Catholics, with the regular mass-attenders moving more Republican than they were in 2004, but the less-regular mass-attenders moving more Democratic.
Looking at the Republican side, George W. Bush generally did better than John McCain in many religious groups. But where McCain had his biggest problem was among religious minorities of various kinds.
Why was the change so modest?
Given the focus in the campaign on religion and the efforts to mobilize religious groups, why was there so little change? There are a couple of ways to answer this question.
First is that it's important to remember that, in terms of the overall vote, the change between 2004 and 2008 actually wasn't that large. John Kerry got approximately 49% of the two-party vote in 2004, and Barack Obama got about 53% of the two-party vote in 2008. So there's a shift of roughly 4 percentage points. Barack Obama won a solid victory, particularly compared with the 2000 presidential election. But it shouldn't surprise us that changes were not that large within many religious groups because there just wasn't that much overall change.
Another thing to remember is that many religious groups are strongly partisan these days. They're deeply embedded into the party coalitions, with each party having strong religious constituencies, and some groups in the middle and up for grabs. We know that partisanship is a very strong predictor of the vote. And, as many of you know, between 2000 and 2004 there were some changes in partisanship. The Republican brand name took a big hit and that influenced the vote of religious groups as well.
The impact of partisanship can be seen in the following graph. Here some of the smaller religious groups were combined for ease of presentation, and they are arrayed from the group that voted least for Obama (weekly attending white evangelical Protestants) to the group that voted most for Obama (black Protestants). The solid line shows the Obama vote, and it rises steadily across the chart. But the dashed line is the interesting one: It's the percentage of each of these groups that identified as Democrats. Note that the solid and dashed lines are basically the same line. So part of what was going on in 2008 is that the campaigns were mobilizing their partisans, and even though this wasn't the kind of base-mobilization election that 2004 was, partisanship still mattered a great deal.
Source: Analysis of aggregated election weekend and post-election callback surveys conducted by the Pew Research Center for the People & the Press.
Partisanship develops very slowly over time. It took about 30 years to get the faith-based structure we're talking about, so maybe it's not surprising that it didn't go away in a single election. But of course, over the next 30 years or so, we may see some important changes, and we may look back to this election and see some of the modest alterations as the beginnings of important trends.
A final explanation for the modest change in 2008 may be the patterns of faith-based mobilization by the campaigns. In 2008 as in 2004, the People & the Press asked survey respondents about campaign contacts within their congregations. The levels of reported contacts were down in 2008 for most religious communities -- an interesting pattern given the focus of the campaigns. Interestingly, the least decline in reported contacts was for religious minorities that strongly backed Obama. And among white Christian groups, weekly worship-attenders still reported higher contact rates, and these were the groups that by and large stuck with the Republicans and voted for McCain.
[T]here's another question that the Pew Research Center asks regularly that is, in some ways, even more interesting: the likelihood of presidential success. The results are shown in a final bar chart, where the red bars represent the percentage of each group that thought Bush was going to be successful in office, back in 2004, and the blue bars the percentage that thought Obama will be successful in 2008.
Source: Analysis of aggregated election weekend and post-election callback surveys conducted by the Pew Research Center for the People & the Press.
Results for Bush based on interviews with 62 black Protestants, 71 minority Christians and 95 weekly attending white mainline Protestants. Results for Obama based on interviews with 95 members of non-Christian faiths. All other results based on interviews with at least 100 respondents.
Look at the blue bars. Every religious category has a majority that thinks that Obama will be successful in his first term. That's interesting because it suggests that President Obama has an opening, even with religious groups that didn't vote for him. Now, looking at the red bars, compare that with what happened in 2004. Notice that many of the Democratic groups didn't think that Bush was going to have a successful second term. Maybe they were onto something, given the decline of Bush's approval after the 2004 election.
This chart does suggest that there may have been a bit of a change between 2004 and 2008. So behind the modest shift in the voting behavior of religious groups, there may be a larger change lurking. And these possibilities may have to do with how successful President Obama and his Democratic colleagues actually turn out to be in office. Campaigns are important things to voters, but they're not nearly as important as the performance of the government itself.
Anna Greenberg
ANNA GREENBERG: I want to focus more on looking at individual groups, but also, what the religious narrative was and how it played out in this election. [T]he Pew Forum did an excellent analysis of the coverage of religion in the election. Their analysis showed what I think we all know instinctively in our gut, that the main religious narrative of this election was about Barack Obama's faith. That came out in a lot of different ways: the question of whether or not he was a Muslim or a Christian, his middle name, etc., but also the Jeremiah Wright conversation, which started during the primaries and extended throughout. Even though John McCain didn't advertise on Jeremiah Wright, other groups did, and certainly it was part of the discussion.
I think that the whole question of Barack Obama's faith and his affiliation raised lots of questions for lots of voters. It wasn't necessarily a debate about religion itself, but it was a stand-in for a conversation about Barack Obama. Who is he? Where does he come from, as John McCain said. What values does he represent? Is he patriotic? There's a whole set of questions around this question of his affiliation -- of his faith -- and that's on the more negative side. On the more positive side, would he have the ability to reach out and would he be able to make inroads among white evangelicals? Would he be able to be a unifier? Could he bring people together? Could he reach across religious lines and break down some of the trends that John was showing us?
He actually had a very significant religious outreach program -- not as extensive as the kind of religious outreach that Rove and Bush did in 2002 and 2004, but certainly more extensive than what Democrats had done in the past.
So what's interesting about all of this is that, first of all, it's very hard for Democratic candidates to escape stereotypes about their religiosity. In a survey I did with Religion & Ethics NewsWeekly and the United Nations Foundation, we asked the question: Which candidate do you think is more religious? I think we all know, objectively, that Barack Obama was the more religious candidate. And actually most of the Democrats -- if you look at Hillary Clinton -- were more religious than John McCain. What's interesting is that we asked this question in September, so this was well after conversations about belonging to the Trinity Church for 20 years. It's after Saddleback2, so it's after a whole lot of conversation about Barack Obama and his religious affiliation, his religious participation.
It's not by a huge margin, but it's actually beyond the margin of error -- 32% think that McCain was the most religious candidate, 28% Barack Obama, another third didn't know. So it's very hard for Democrats ever to be perceived as the more religious. [W]hat's really interesting is that only 46% of Democrats said that Barack Obama was the more religious candidate. So even among Democrats, the stereotypes about Democrats and religion are pretty strong.
The Pew Research Center asked: Do you happen to know what religion Barack Obama is? And 57% said he was Christian, 12% thought he was Muslim, and the rest -- I want to know who that 1% who think he's Jewish is -- I don't understand what that's about. (Laughter.) Among the people who thought that Barack Obama was a Muslim, there were an equal number of Democrats and Republicans; this was not necessarily a partisan issue.
We did a lot of focus groups with people who had voted for Hillary Clinton in the primary but were not voting for Barack Obama in the general and with people who called themselves Democrats but weren't sure what they were going to do or were voting for McCain. They all ended up being sort of the same people.
In these focus groups, this whole conversation about whether or not Barack Obama was a Muslim was a very big piece of it. Now people aren't necessarily rational when they have these conversations, so in the same sentence they would say: "I think he's a Muslim" and "I can't believe he belonged to that church for 20 years." You're trying to figure out how they can reconcile those two pieces of information -- that he's both a Muslim and a member of this radical Christian church. It was my belief that this whole question of whether or not he was a Muslim, on its face, was an issue for some people but also was a stand-in for race.
One of the things we also spent a lot of time trying to figure out was how this whole so-called Bradley effect -- I'm so glad that conversation is over; I never want to have it again -- but we took it seriously. When it was clear that Obama was going to be the nominee, as early as February in my view, we started doing some real research on this whole question of, how do you measure the impact of race?
It turns out if you just ask directly -- this isn't surprising -- does Barack Obama's race matter to you? -- that's not a very good way to measure it. In fact, what you found was that for people who said that it did, they were more likely to vote for Obama. They were younger; it was a positive. But there were other kinds of questions that could be a stand-in for race, and one of them was this idea that Barack Obama was too close to extremists. Actually, that ended up being a pretty powerful predictor. Lots of other things predict how you're going to vote, including partisanship and church attendance, but that kind of question in terms of the race issue -- and you can word it lots of different ways -- was actually a predictor of how people were going to vote.
On that question it was about a third Democrats, a third Republicans and a third independents; it wasn't correlated with partisanship. There were clearly a bunch of white Democratic voters for whom this whole question of race/Muslim/extremist -- however you want to ask it -- Who is he? Where does he come from? What are his values? -- was important. You had some people in these focus groups who thought he was literally like a Manchurian candidate, that he was going to get elected and then get into office and reveal himself as a Muslim and take over the country.
Now just to make the point, keeping in mind that half of these folks are Democrats, among people who thought that Obama was a Muslim, 51% would vote for McCain, 37% would vote for Obama, and among people who thought that Obama was a Christian, the numbers are flipped -- 52% would vote for Obama, 39% would vote for McCain.
One of the groups for whom this whole question of Barack Obama as a Muslim -- whether or not he's a Muslim but also extending to other questions around Israel and Palestinians and things he might have said about the Palestinians, etc. -- was very important to Jewish voters. I've done a fair amount of work on Jewish voters, though it's not easy to do because Jews are such a small percentage of the population and of the electorate. But I was on a panel that the National Jewish Democratic Council put on at the Democratic convention, and there were three different sources of data. There was the J Street survey, which was an Internet survey of Jewish voters; there was Gallup, which wasn't actually a separate survey of Jewish voters -- they do so much polling that they basically aggregate their surveys and pull out the Jewish interviews; and then we had our own internal databases, where I had about 1,500 Jews.
Winning the Jewish vote
Looking at all three sources of data going into the convention, it was pretty clear that Barack Obama was pretty seriously underperforming among Jewish voters. This trend line from Gallup shows you that in the summer, he's getting anywhere between 60 and 65% of the Jewish vote, which seems very high obviously. But relative to -- certainly since Clinton -- Democrats get around 80% -- 78, 80% -- of the Jewish vote. So that kind of number is actually pretty low. Now it doesn't really matter nationally, that difference between what previous candidates have gotten and what Obama was getting, but it actually mattered a lot in Florida.
I did not believe that Obama was going to win Florida, certainly not at the beginning of the year. But as more and more polls were coming out showing the race was competitive in Florida, we did our own poll -- a couple polls -- in Florida, and Obama was doing very well in places like Miami, and he was very competitive in the I-4 corridor, which includes Tampa and Orlando. That's always considered kind of the swing area of Florida, and he was actually tied with McCain, which was a very good result. But he was really underperforming in Broward County, Dade County, West Palm -- the Jewish parts of Florida.
It might have been a 20-point gap between the percentage who were calling themselves Democrats and the percentage who were voting for Obama. Now my grandmother and grandfather retired to Ft. Lauderdale. I've spent a lot of time in those condos in Broward County, and I know what those folks are like. And of course, there was "The Great Schlep" and that sort of thing. Did everyone see the Sarah Silverman video, "The Great Schlep?" I mean, nothing really happened. I don't think lots of Jews went down to Florida to convince their grandparents to vote for Obama, but it was funny. It was very Jewish.
But, anyway, so what was really fascinating about this was at this meeting at the Democratic Convention, it was like a mob of angry, Jewish state legislators from Florida, who were really angry at the Obama campaign because they felt like, we're going into these condos and we got nothing. We have to convince these old Jewish voters that Obama's okay, that he's not a Muslim, that he's not going to favor the Palestinians in some way and not support Israel. If you looked at the data that I had, it was really clear that the issue was with older Jews, which was not surprising; all of Obama's issues were with older voters, older white voters. Jews aren't different -- radically different -- in some ways than other voters; they're the same groups.
But this obviously mattered a great deal in Florida. I don't know exactly what the Obama campaign did. I think some of it was Sarah Palin; I think some of it was the debates. There were a variety of things that happened independent of the Obama campaign, though my understanding is that they put significant resources into outreach in this part of Florida. But what you can see nationally with the Gallup data is that over time Obama went from 62, 61% of the Jewish vote to 74% by the end of the election. And if you look at the exit polls, Obama got 78% of Jewish voters -- that's actually slightly better than Kerry did and it's about where Gore and Clinton were.
I think if you look at the polls going into Election Day and then you look at the exit polls, the one group where Obama really did worse than Kerry was with older white voters. If you look toward the end, particularly post-debates, which I think were enormously reassuring to voters who had concerns about him and who he was and all the questions that McCain asked, you started to see a slight movement. But when you look at the exit polls, he actually didn't do very well with older white voters, and what I don't know is, was that movement real? Was there a pull back on Election Day? If there was a Bradley effect, maybe it was just with that group. But the point is, Jewish voters are the one group where the campaign, in any event, was successfulin overcoming some of these doubts about him.
Obama's gains with younger evangelicals
So the other narrative in this election was, can Obama make inroads with more-conservative religious groups? He poured resources into that in his campaign. People had anecdotal evidence that young evangelicals might be more favorable to Obama than older. So when I did this poll for Religion & Ethics NewsWeekly, we did an over-sample of younger evangelicals.
Most of the national tracking polls had Obama at about 60% among under-30s, and he actually got 66%. So what we did with this poll was [a] multi-mode study. A certain number of interviews came from random digit dial, a certain percentage from the internet -- we didn't do the cell phone for a variety of reasons. We had, I think, the first real sample of young evangelicals and what we found was that, in fact, it was true that younger white evangelicals were more likely to support Obama than older white evangelicals.
Data from Religion & Ethics NewsWeekly/United Nations Foundation survey of 1000 adults and 400 young evangelicals, conducted September 2008 by GQR.
I was on a panel with GOP political strategist David Winston last week and he said, well, you know, when they get older, they'll vote more Republican because people when they get older, they tend to vote more Republican. And I said maybe, but the reason why people do that is they tend to get married and start going to church and these folks already go to church, so I don't know what's going to happen. The kinds of changes -- you get married, you have children, you pay taxes -- for younger evangelicals, they've already done a lot of the things that make you more conservative when you get older.
Even though this is from September, the 30% number is true across all three data sets that I looked at, so I feel pretty confident that Obama got 30% of younger -- this is under age 30 -- white evangelicals. So about 8 points better. [S]till, it's a very Republican group. Let's not get carried away here. I mean, 62% voted for McCain, and maybe even higher. But 30% for Obama, so 8 points higher [than among older evangelicals].
Data from Religion & Ethics NewsWeekly/United Nations Foundation survey of 1000 adults and 400 young evangelicals, conducted September 2008 by GQR.
Much more dramatic is the Republican brand among younger evangelicals. To measure favorability, we do something called a feeling thermometer, where you have to rate your feelings toward candidates, organizations or groups on a 0-to-100 scale. Anything above 50 is warm and anything below 50 is cool.
If you look at the white evangelicals over age 30 and at [feelings toward] George Bush, [you find] 57% positive, 29% negative. And you know what Bush's numbers are overall, so it's radically different. But look at where the younger ones are; he's got net negative ratings, 39% warm, 48% cool. You see a similar sort of pattern, though it's not as bad, on the Republican Party -- plus-45% positive rating, 64% warm, among older white evangelicals and only a net plus-5 positive rating for the Republican Party among younger evangelicals. So Obama gets 30%, but the Republican brand itself is really challenged among this group. I think this is going to be one of the more interesting things to track as this group gets older and becomes a bigger and bigger part of the white evangelical group in the electorate.
Then you can see, looking at a variety of issues, the differences between older and younger evangelicals. Again, we don't know if, when younger evangelicals age they'll get more conservative, but I was really struck by this slide, which is a question on gay marriage.
Data from Religion & Ethics NewsWeekly/United Nations Foundation survey of 1000 adults and 400 young evangelicals, conducted September 2008 by GQR.
What I would point out is that a majority of younger white evangelicals favor some kind of legal recognition of same-sex couples. So 26% say the right to marry, which I think is a very high number, and 32% say that they should be offered the same protections and benefits of marriage. Overall, 58% of younger white evangelicals think there should be some kind of legal recognition of marriage between same-sex couples. If you look at the older evangelicals, you can see that it's pretty dramatically different -- 9% for actual marriage, 37% for some kind of civil union.
If you look at the question of Iraq, again, the younger evangelicals are still different from other young people, but there is a slight majority for reducing the troops as opposed to staying the course, and that's different from older evangelicals. If you look at the question of global warming, the differences aren't huge, but still, the younger evangelicals are more likely to believe that global warming is kind of imminent and we need to take action now as opposed to a long-term threat that we can do something about later. But then, on this question of abortion, there's literally no difference between older and younger evangelicals. We asked a lot of different kinds of abortion questions in this survey. Every single way you asked the abortion question, you had the younger and the older -- and in fact, in some cases the younger being more conservative than older evangelicals on abortion.
Data from Religion & Ethics NewsWeekly/United Nations Foundation survey of 1000 adults and 400 young evangelicals, conducted September 2008 by GQR.
I wanted to end on one final discussion of where I think religion played an interesting role in this election, and that's ballot initiatives. In general, gay marriage and abortion -- not very big issues in this election nationally, and that often happens when you have elections dominated by the economy and the war in Iraq in 2006. But there were some pretty important ballot questions, particularly Prop 8 in California, which I've actually spent some time doing research on and which we can talk about during Q & A.
Evangelicals' impact on ballot initiatives
I'm throwing this out here as a kind of interesting, potentially provocative idea, but I don't have an explanation. It looks like, if you look at the different states where there were abortion and gay marriage initiatives, that evangelicals were key to the passage of gay marriage bans but not sufficient to pass the abortion restrictions. I don't know why, exactly, but I'm just going to show you because I think it's interesting. If you look at Prop 8, which is the gay marriage ban in California, [among] white evangelicals-- only 17% of the electorate in California -- 81% voted "yes." Everybody else was a slight majority for the "no" vote. Obviously that's gotten into a lot of discussion of African-Americans and other things we can talk about -- Hispanics -- and that's all part of the story. But the evangelical piece is key here, and the Mormon church and others did a huge amount of organizing work in the evangelical community around Prop 8.
If you look at Florida and the gay marriage ban, 81% [of evangelicals], and they are 25% of the electorate in Florida, voted for the gay marriage ban. Arkansas, it wasn't gay marriage -- it was an adoption measure -- and it's a lower number, 65%. But then when you compare the abortion ballot questions, especially in California, to the gay marriage ones, it's kind of interesting -- all the gay marriage ones passed and all the abortion ones failed. So, again, I'm not entirely sure what that's about, but it's interesting to think about.
But if you look at the abortion limits-- and the abortion limit in California was parental notification, which is actually something that gets pretty high levels of agreement with everybody-- it's interesting that you have about a 10-point drop in the number of evangelicals voting "yes" on parental notification compared with the gay marriage ban -- and obviously a much larger majority of non-evangelicals voting "no" on abortion limits.
South Dakota [the abortion limit on the ballot] was a redo of the one that failed two years ago. If you remember, the main reason [the earlier initiative] failed was there was no exception for health of the mother, and so they changed the language, but it still failed in South Dakota. Again, a smaller percentage of evangelicals, compared with the gay marriage bans -- 64% -- and a very big majority of the non-evangelicals voted against it.
I think it's something to investigate, both on the question of the differences in how evangelicals organized around these gay marriage bans in these different states, especially California, relative to the abortion bans, but also why there was a lower level of support among evangelicals on the abortion limitation measures than there was on the gay marriage bans. Again, I don't really have an answer for it, but I think it would be interesting to dig into that and think about it.