 Oh, I'm sorry. OK. Well, welcome, everybody, to a very special edition of the Berkman Tuesday lunch series. This is, of course, election day. So I hope you've voted whether or not you're from here. In the tradition we have in Chicago, which is to say early and often, hopefully also in Boston. So I want to introduce our very special guest, Sunshine Hiligus. But before that, to offer a few thoughts, one, a word of warning to let you all know that you are being recorded and webcast and podcast and so forth so that we can get other folks who don't have the great good pleasure of having this lovely lunch and being here with us in this room involved in our discussions that Sunshine's going to talk for on the order of? Probably too long. Probably too long. But you can cut me up. But she's going to want you to interrupt her and engage with her. And at some point, we will devolve into spirited discussion and try to wrap up by about a quarter to two, at which point people will probably linger afterwards, and you can have all those side conversations that are so fruitful. And really, that's all I wanted to say. And perhaps we could begin by just making a loop around the room and everyone giving a very brief introduction of themselves so we know with whom we're sharing this lunch. Start us off. I'm Joe from Hobok. And for you guys in the back, you have to speak up, because these kind of sound like jet engines on this side of that. I'm Dan Jones, Brooklyn staff. I'm David Argy, a Brooklyn fellow. Hi, I'm Alvaro Santana. I'm from the Social Energy Department. I'm Gettavasque, Brooklyn staff. Lexi Koss, Brooklyn staff. John, I'm the gymnast consultant here personally. John Pahls, VHC with folks of the current NYC research fellow, just a number of folks. From the night program? Yes. Welcome, Mr. Evier. I'm the director. No signal. I'm your Brooklyn, and I too will have to live in the room. I'm the non-nature of Brooklyn staff. Ruben Rodriguez, I'm a 1L at the Law School. Have a great day, Brooklyn staff. Hi, this is Dan Steeke, a Brooklyn fellow. A Roxanne O'Vyrum research assistant with the Cooperation Group. Ethan Zerkerman, Brooklyn fellow. I voted in Massachusetts, Connecticut, and Rhode Island on my way in today, just to be heading out. It's like a hit through Hampshire. This is being recorded, which is illegal, right? I'm Chairman Goldberg. I'm at 2-up. Hi, David Larshall, Brooklyn staff. I'm putting on your genital work on time. And I'm going to not take almost any more of your time, but just ever so briefly to tell you what a great pleasure it is to have sunshine with us here today as being kind of an old friend of the Berkman Center. One of the only, probably the only person who's interested in the same issues that we are within both the GOV department and IQSS, the Institute for Quantitative and Social Science. And so we've had this long and great friendship in which she approaches elections, electoral behavior, voting behavior with a particular eye to its quantitative studies, and integrating really interesting aspects of the internet and the things that we care about so much around here. So rather than mingle or misdirect your perceptions, I will just give a plug for this relatively new book out this year, The Pursuatable Voter, Wedge Issues in Presidential Campaigns, which is not only interesting but also very handsome. I encourage you to find a copy. We have one up here for you to look at. And I will just ask Sunshine to take us away. Great. Terrific. Well, thank you so much for the invitation to be here. And I appreciate being so patient in whether or not I was going to be able to be here. I was trying to juggle, as I was saying, because I do surveys and elections. About the only time I ever get any attention from anybody but some boring political science colleagues is every four years. And so I'm delighted to be here to talk about this project, which is to some extent an extension of the Pursuatable Voter, an aspect of the Pursuatable Voter. But I also realize that it's election day. So if we want to forego most of the conversation about some of the evidence I'm going to be showing from the 2000 for election to talk about some of the dynamics today, I'm more than happy to do that as well. Because to give you a little bit of a disclaimer, some of the things I'll be saying today, it holds in 2008. But there are also some notable differences with 2008. We just don't yet have the data back to say with certainty exactly how things will look. But I think that some of the trends I'm talking about actually do still hold in this election as well. Normally, when I talk about this broad research project I've been working on, I kind of go through an overview of kind of where the state of the literature is in the social sciences regarding the impact of information technology on different campaign actors. But I think that this group probably knows quite a bit already about how information technology has changed the behavior of the media, of the public, of politicians. And I'm happy to talk more about those. But really what I want to focus on today is the impact of information technology on politicians. And where a lot of the research that we have seen has focused has looked at how, for instance, it's changed the campaign fundraising strategy. I mean, there is no doubt that one of the reasons that Obama emerged as the alternative to Hillary Clinton was because of his success in raising money online. So there's been quite a bit of interest in that topic. There's been quite a bit of interest in new ways that technology have allowed candidates to communicate with voters, text messaging, all the things that you are all very familiar with. It has changed how the candidates have tried to get into the headlines. I mean, one of the most important things that a candidate can do is not just by television advertising, but to get talked about in the media. I mean, that is foremost for one of their goals. And the internet and information technology has really changed one of the paths by which they can get broader media attention. And I should say, in terms of, I know that a lot of people here are interested in blogs and the impact of blogs versus traditional media. And one of the things that we see is that the two of them actually work together, that when it is that a YouTube video or a blog really has an impact is when it is disseminated through the traditional media. And so you see, for instance, a link being provided on The New York Times to Obama Girl. So it's really the two of them working together as opposed to them really being kind of at odds. But what I want to focus attention on is slightly different from what I think a lot of people have talked about when they think about technology and campaigning and candidates. And that is that the information environment and the technology available to the candidates has changed not only the style of campaigning, but also the substance. It has changed who it is that the candidates are communicating with and what they're willing to say. And that is fundamentally different from what we see a lot of social science research saying. Because the theme of a lot of social science research is, sure, the campaigns are now sending out text messages and they're sending out emails. But that's really just changing the style, right? The way that the candidates communicate rather than anything really substantive about their messages. And really my argument is that in fact, that technology has fundamentally changed the substance of the political campaign. And just to cut to the chase, what I'm going to argue is that candidates are willing, because of information technology and changes in the information environment, willing to take positions on more issues and more divisive issues than ever before. And so that's what I want to get to in terms of my argument. And just to give an example, in 2004 Bush and Kerry took positions on 75 different political issues in their direct mail. 75 different issues. There is no way that 75 different issues were being discussed in television advertising, even in the mainstream media in terms of the public debate that people were having. 75 different issues, a lot of them very, very narrow. And those things were made possible because of, again, changes in the information environment. Just to pitch my book myself, thank you, Colin, for doing it as well. This is part of the work that I did with the persuadable voter. But in fact, I found it so interesting, this one part, that I'd like to take it further. And so I actually am interested in feedback. Sometimes people come and they present a book and they really don't want feedback. I really do want feedback, because this is part of some ongoing research as well. Just to give you a feel for how this fits into the persuadable voter, since the persuadable voter, I told you I'm going to be talking about the campaigns and candidates, but the title is about voters. So what Todd Shields and I do in this book is to look at who it is that is persuadable in the electorate and then what it is the candidates are doing to attempt to sway those persuadable voters. And so we have three key arguments. The first is that the persuadable voters are not just the undecided voters that you hear the headlines about. They're not just independents who oftentimes don't show up. They're not the unengaged, uninterested. I think maybe you've seen the John Stewart clip about the undecided voters. It's hilarious if you haven't seen it. I'm always going to be using that in my classes from now on. But in fact, persuadable voters are the people who actually show up at the polls and who can be persuaded are people who agree with the candidates, both candidates on different things. And so some of those are independents who agree with McCain on tax policy and Obama on abortion policy and they have to resolve this tension between their agreements and disagreements with each of the candidates. But it turns out that once you take into account the fact that many independents actually don't show up on election day and that many partisans actually disagree with their party on a number of issues, that the majority of the persuadable voters are persuadable partisans. They are partisans who disagree with their policy on an issue that's important to them. So I call them the otherwise partisans, you know, the Republican who has pro-choice but otherwise a Republican. And if what happens is that pro-choice Republican comes to believe that what is at stake in the election is abortion policy. If that issue is made salient, then that's how they'll vote. They'll vote against their party and on the basis of that cross-pressuring issue. The candidates know this. They know that they have to build a coalition between their base and a set of persuadable voters. And so what they're doing is they're looking for cleavages in the potential opposition coalition and any issues where people disagree that they can kind of pluck those people away. And so what they do is they try and identify those, you know, the Democrats try and identify those pro-choice Republicans and they tell them, huh, you should vote on the basis of abortion policy. Abortion policy is actually at stake in this election. And so they try to increase the salience of that cross-pressuring issue. And they've always tried to do that. So you look back to the Republican Southern strategy where Republicans made an effort to, you know, talk about anti-civil rights positions in an effort to appeal to white Southern Democrats. That was a wedge strategy where they were trying to increase the salience of that cross-pressuring issue. But it is risky to take positions on especially divisive issues like abortion. You know, if you have a TV ad about your abortion positions, you risk, you know, alienating all those people who disagree, taking a position on any issues, think of, you know, Daddy Bush and Read My Lips, right? Taking a position on any issue, paint you into a corner where you're a flip flopper if you change your mind. And so just holding all those constant, Canada's don't want to take positions on issues, but they're able to because of changes in the information environment, information technology, because now what they can do is they can narrowly send messages to individual voters without it being received by the broader electorate. We call it dog whistle politics, right? Where you send a message to those who are intended to receive it without anybody else finding out. And that's done through direct mail, political email, text messaging, and so on. And so the information environment is really what enables this type of micro-targeting of wedge issues. And so that's what I want to focus a long way to get to what I want to focus on today. Just to hit home the point, this is just a number of Democrats in the electorate and Republicans in the electorate. And I should have added 2008 on there because what you would have found is in 2004, the number of Republicans went slightly up, and that has since swapped. In 2008, the number of Democrats in the electorate is actually a gap of, say, six percentage points in the Democrats' favor. But the key point here is just that you can't build a, you can't win from your base alone. And one of the explanations for why it is candidates are targeting wedge issues has been they're trying to mobilize their base. And so I want to take on that alternative hypothesis. That's one of the things that is kind of an alternative theory floating out to what I'll be arguing today. Just to get to the point more quickly, again, what I'm arguing is the information environment enables candidates to micro-target different messages to different voters. And there's two key pieces here. One is that you have the technology to narrowly target a message. But you're only going to do that to the extent that you have information about the individual voter that you're going to know what they're going to be receptive to. And so the other piece of the equation is that candidates have more information about individual voters than ever before. And really, we can trace this to the computerized voter registration rolls that the Help America Vote Act has mandated. This is the first election in which all states have had to have the statewide databases, although most started transitioning and by 2004 had them. These statewide databases of voter registration rolls means that candidates have, at their fingertips, your name, your address, your party registration in most states, and your turnout history in most states. And to that, they can merge in information about any contributions that you've made, any magazines that you subscribe to, and so on. And then they combine with that some polling. And for those of you who've, I see lots of Obama shirts, so those of you who've worked on the campaigns know also that when you've gone door to door, that you've had a series of questions that get mapped back into this database so that the candidates have information about what each individual voter is interested in hearing about. And if there are any cross-pressuring issues that might help in pulling them into their camp. Yeah. So I'm guessing you are, I mean, if I had to speculate, based on the fact that we're in Cambridge, I'm guessing this is on the Democratic side. The Democrats are actually trying to catch up to the Republicans in terms of the quality of their database. There are still many errors because it's something that constantly has to be updated, right? I mean, you, and so there are definitely errors. And actually, I'll show you some evidence of how those errors actually affect the type of campaign dynamics that we see. So I just want to run through a couple of examples. This is from 2004. This was a pro-carry piece. Not sent out in Cambridge, I suspect. When it comes to politics, I care about two things, my guns and my job. I love it that my guns comes first. This will be Arkansas if you don't vote. The Bible will be banned. And this is a man proposing to another man so gay marriage will be allowed. In this election cycle, here's McCain sending out about abortion policy. Obama sending the message that he's a committed Christian, he's not a Muslim, he's a committed Christian. This was sent in Southern states, again, not in Massachusetts. So those are just a few examples. I'll go through some more later. But the key is that what we find is that direct mail in particular, but political emails the same way we just don't have as nice of a data source, is that the content of this microtargeted mail is very different from what we see in the broader campaign. And so in the analysis that we do in the book, we look at the issue content of direct mail versus television advertising and find, for instance, that 30% of direct mail pieces had wedge issues in them and 9% were just abortion, gay marriage and stem cell research, whereas 0% of TV ads were like that. And that's not quite fair, that's rounds down because there was a pro-bush abortion television ad but it was a Spanish language ad, and so it was already kind of targeted to a narrow constituency. And so we get a very different feed, we have a very different content of the campaign when the candidates are able to use information and use microtargeting. One of my pieces is missing, but there was another graph up there as well that I must have accidentally deleted, and that is who gets targeted is actually, again, they use the information they have to spend their money more efficiently and that means that they're spending their money on those people who are persuadable and then they send a little bit to their ones. So the ones of those people they know are gonna vote and they know they're gonna vote for them, those people are not ignored, but most of the money is going to the persuadable voters. And like television advertising, they're going to competitive states, but more than just the state, they also look at the individual and they're in particular looking at those individuals who are going to vote. And so the active or non-active vote history, these are all people who are registered voters but the average number of pieces of direct mail that somebody who had an active vote history, which means they had voted in both of the last two elections, they had 7.3 pieces of direct mail on average compared to 1.3 pieces for those people who lived in non-battleground states who were registered to vote but weren't active voters. And so this is from 2004, one of the things that is slightly different about 2008. And I think that again the media message has perhaps been a little bit overblown about the extent to which the candidates are now going over, mobilizing new voters. It's been a little bit cheaper and easier for Obama to mobilize even people without inactive vote histories because he knows that with young voters and African-Americans that they're going to break so much in favor for him that if he can get them to the polls that it's a positive. And so we see a slightly, I expect once we get some data that we'll see that this election cycle there was a little bit more effort plugged into on the Obama side actually mobilizing people that don't have an active vote history. But even still the majority of money is going towards those people that just need a little boost to get them to the polls. Looking at specific issues, I kind of gave you the breakdown, just wedge issues and so on. Again, what we can do is look at direct mail versus TV and again see that the content is very different. And the issues that are being talked about in television are very broad problems rather than divisive issues. So it's things like healthcare and education and jobs. It's things that nobody disagrees on the outcome that we want. We just disagree on the means to get there. And so what you find are these very broad messages and direct mail has some of that but it also has things that candidates would never take a position on in television advertising. Okay, so this question, yeah. On that last chapter, were those all majority issues or were some of those wedges because you said healthcare education jobs are things that nobody agrees about the role so it's not a wedge issue. But what about the other issues that the list, they are wedge issues, right? Right, so this is actually just kind of the top issues that we're being talked about rather than my definition of wedge issues versus not wedge issues. And actually, I mean, you could think about their education. Everyone agrees that we want an educated electorate. There are some education issues that are wedge issues. Vouchers are a wedge issue. And so this is just a very broad, yeah. Okay, so this question of, is this really going? Are all these kind of targeted messages really going not to the persuadable voters but to the base? If it was really all about the base, then there's a few things that we might expect. Number one, that the messages are being sent not because they need to persuade these people but because they want money from them or time from them or right to get them to the polls perhaps. But you find that almost none of the direct mail pieces are actually pure mobilization appeals. If a candidate is sending out a message, a piece of a campaign message, they never say go to the polls without also saying here's how you should vote when you get there and here's a motivation for you getting there. And so almost all campaign messages have persuasive content in them. And it's also not the case that most of the direct mail pieces, the ground war campaign are explicitly asking for volunteer appeals or asking for fundraising. So for those of you who might be part of a base, you may have gotten some of those. But the vast majority of ground war messages are actually not, again, they're not going to the base to try and get them more activated. Instead, they have issue content. They are an attempt to persuade those pivotal voters. In fact, I like the party labels. This is just whether or not there's, because you might think that if it's really just a base strategy, all they need to say is vote for the Republicans or vote for the Democrats or remind somebody that you're a Democrat and so you need to vote, get out and vote. But oftentimes they don't even want to admit their party. This is especially the case in this election year. And so I think you look on the Republican side this election year, especially at congressional level and you'll find that none of their campaign messages advertise their party affiliation. But they do, again, they contain a lot of issue content. This is a little bit of a complicated graph to interpret, but the key thing here is that to identify whether or not direct mail pieces were going to the base or to the persuadable people in the middle, what I did is I just looked at strength of party ID from strong Democrat to strong Republicans, so the dark blue to the dark red. And you look, for instance, at the pro-Bush moral appeal. So this is the percentage of strong Democrats all the way to strong Republicans who received pro-Bush messages that included a moral appeal. It was a pure base strategy. What you'd expect is that strong Republicans were gonna be the ones receiving the most appeals about gay marriage, abortion, stem cell research. But in fact, what we find is the blue line and the yellow line are actually higher than the Republican lines. It is the case that Democrats and independents were more likely to receive that message than were Republicans. And again, that just reflects the fact that these messages were being used to try and peel away persuadable voters who they thought they could capture on these moral issues. And just to give some more examples, here's one that's very explicit in that appeal. I feel safe with President Bush. I've always been approached by a Democrat, but party loyalties have no meaning when it comes to my family safety, right? So the point of the message is to say what should be salient in your vote choice? We know that you have conflicting pressures, but what matters in this election is the foreign policy aspect. And you see this kind of on a broader campaign basis as well where each side is trying to focus attention. Obama's trying to focus attention on the economy. McCain, his hope was to focus attention on foreign policy. This is from the Florida Republican Party. The Dale Castro endorses Obama. I love this guy. This is actually not true. I mean, it's not a photoshopped picture. The picture is true, but the endorsement is not. I'm going to skip that and talk about some of the consequences. So in terms of the consequences of this microtargeted strategy, one is that what you found is that because the candidates are spending their money more efficiently, they calculate for each individual voter, your predicted probability of voting and your predicted probability of voting for them. And they map those two together and they cut a threshold, right? They don't need to spend a lot of money on their ones and they don't want to waste their money on people that have no chance of voting or no chance of voting for them. They're trying to target to that middle. So what's happened is, and that's kind of hard to see, but what's happened over time is that this has meant that they have ignored those people, especially who are unlikely to vote. And so this just breaks whether or not somebody gets contacted by a campaign by party registration. And you probably can't see the line, but the 20% and 7%. So it's always been the case that those who are registered to vote have been more likely to be contacted by a campaign. That's always been the case. But that gap has grown from a 7% and 20% gap to a 14% to a 58% gap for those who are registered to vote in a battleground state. So we've had this explosion in the inequality in mobilization attempts. So the people who are not registered to vote are essentially getting left out of the process and nobody's asking them to join the process. And there are the nonpartisan groups that are trying and there are all the caveats about some of the things going on the selection cycle, but even still, right? Where is Obama putting the majority of his money in terms of mobilizing youth vote into college campuses, not into the uncollege educated and so on. And so this inequality in mobilization attempts is something that although it might have shrunk a tiny bit this election cycle, the overall trend is one that seems quite undemocratic. Another consequence is this 75 different policy issues that were being talked about in the campaign. And again, the Fidel Castro example, I think helps to highlight the fact that this is not, this election cycle, a similar thing is going on, right? You know, where the headlines are talking about the economy on the ground, there are many other messages that are going out. And so there's this fragmentation of the policy agenda so that some people are being told the issues they care about are a priority to the candidate and then your neighbor is being told that their pet issue is also a priority to the candidate. Well come time to govern, not all of those issues can be a top priority and it's going to create, it does create problems when it comes to governance. And we certainly saw this with George Bush where nobody in his coalition was kept happy. The Christian conservatives were told that moral issues in their direct mail and they were told the moral issues were a priority. And so they were thrilled when Terry Shiva was being the topic of discussion but upset when Social Security was the first policy initiative that Bush tried to take on in his new term. Whereas the people who were being told that small business issues were really his priority, were upset with Terry Shiva. So the coalition that has to be built for an election, it has consequences come time for governance. And then you also have this focus on issues that someone are your superficial. Not to say that snowmobiling policy is not an important issue to some people but this was the focus of direct mail and phone calls to a number of people in 2006 and 2004 because there was a list of people who owned snowmobiles and so the Republican party was sending messages that said the Republican candidate's going to be the best candidate for snowmobiling policy. Now this might upset the rest of the country to hear that this is what is being sold as a top priority rather than at the time the Iraq war and the economy. And so there's this concern that through micro-targeting that there's really a focus on issues that are important to a small group but again might not be important to the broader community. So where does this leave us? In terms of 2004, interpretations that really was all about voters, the values voters, really was a misinterpretation of the election. The vast majority of people cared about Iraq and not about gay marriage. But it's not surprising that the traditional media came to this conclusion because for some, for those Christian conservatives they were being told that these moral issues were in fact a priority to Bush. And so for them that's what the election was about at the same time that some 30 other segments in every stage. So every state had, all the battleground states had about 30 different target groups. They were all being told that some other issue was also at stake. And so it makes it difficult for us to say after the election is all said and done what the will of the people really was. Because different people were being told the different issues were important and so we're voting on the basis of different messages that they were sending to the candidates. And so when it's all said and done, we would like to believe that there's a clear message being sent as a result of an election. That there's a mandate that if one candidate wins we know exactly where that candidate has been instructed to do based on the votes of the people and that's no longer possible given the diversity and complexity of the campaign agenda that they are selling to the American public. Okay, so I think I'm going to stop there and open up discussion, especially because I didn't really talk about this election much and I'm happy to. So. Thank you for this relates to what your sense is just sort of a plea for you to continue extrapolating what you talked about in the context of email and of the current campaign. It would seem like email and the growth of computing power and the ability to manage large data sets would mean that for marginal cost you could just mind the heck out of these kinds of databases send an infinite number of messages for basically zero amounts of dollars compared to what it takes to send direct mail with a stamp stamp on it. So is that prophecy being fulfilled? Is that basically what's happening or is there a better story? So there's something about the email that makes it a little bit interesting and I should say that we have a little bit of data from 2004. I was hoping to be able to do the study in 2008 and didn't have the funding and so we don't have kind of the detail of data that we had in 2004. But there's something that's slightly different because email is so cheap you can do one of two things, right? You don't have to spend your money, your email as efficiently so you can in fact send it to people with a very general message. So on the one hand you do find very much so the micro targeting in political email. But a couple different things are going on. One is that a lot of the emails are going to people who have signed up online and so number one the people who are signing up online tend to be the base not persuadable voters and it's now very cheap and easy to contact those base people and so it's not necessarily a clear indication of overall kind of campaign strategy. And then in terms of email spam there are some internal studies going on that I know on the progressive side are trying to figure out to what extent do we hurt ourselves, right? To spam people. And I don't know what we, we don't know the answer yet. Videos of things that just happened. So there's a lot of more interaction across the lecture too. How do you see this impacting? The research you've done before and the insights you came up with given this changing landscape. So at the same time that it's become interactive, right? And we, the candidates are very fearful of that, right? And where this has become most apparent is it's definitely true that you, Obama girl, right? People say, wait, is that supposed to be pro-Obama or not, right? Like is this helping him or hurting him? And so there's always that concern that you give the message to somebody else and it's not going to be a message that you want. And so both campaigns where this was, I think really obvious is both campaigns tried to really crack down on people contributing money to third parties, interest groups, right? 527s. And both had sent the signal, we want money to come to us centrally for McCain that was the, you know, the RNC. For Obama it was to his campaign. We want it here. We don't want it in the hands of the third parties. Or in the, you know, the interest groups. At the end of the campaign though, and actually you see this, so that a couple of weeks ago, I think the totals were like 150 million was contributed to interest groups, whereas in 2004 the number was 450 million. And you get, you know, a lot of, if you're the NRA, you are kind of very much going to focus your message on your issue and not necessarily concerned about kind of the overall theme of the campaign. At the end of the campaign, both candidates signaled that they perhaps wouldn't mind if people started spending money on their behalf. But at that point there's no way that I think that we'll see the catch up to the level that we saw in 2004. So as a way of maybe not answering your question very well, but there is no doubt that the campaign has become more interactive, that the swirl of information that people have received about the campaign has become more interactive and is not just coming from the candidates, but the candidates themselves want to try and control that message as much as possible. And I think that's going to be an ongoing tension. Okay, Sean, because the reason that I'm thinking that might be important in what you're talking about is there has to be some threshold which this micro-targeting is going to break down because people are going to say, well, I got an email saying he's for abortion or whatever. And so I'm just wondering if that's something that's interesting to you to study to find out how much this strategy can really work. Absolutely, and so the question is, can, is there a way for those messages to filter up? And we've seen a little bit of that, but still not very much because it's so much more difficult to track emails and the source of emails, right? Compared to television ads where you can respond the same day and get your message out. And so there is an attempt, right? And so we've seen where, whether it's candidate to the presidential level or certainly at lower levels where they are trying to do the viral campaign, right? To get friends on the blogs, blog favorable things and to make it look more voter to voter than candidate to voter. And both sides are fully aware that all research shows that the most persuasive appeal come from people that you know. And there is broad recognition of that. And so there is no doubt that the campaigns have tried to incorporate voter to voter to get their message through voter to voter communications in a variety of different ways. So everything you've talked about so far has been, I guess, messages where the party or the campaign was clearly identified as the source of that message. Right. I'm interested in dirty tricks. Yeah. It's a dirty person. And in the last few years, we've seen things like phone jamming in Ohio or fake flyers and I'm pretty sure it wasn't. Phone jamming, though, was actually party sanctioned in that we have here. But in both those cases, the FBI was able to figure out the source of the activity even if they weren't able to send guys to jail or at least identify the source within a few days or a few weeks. And as we shift to more internet-based technologies, we have the possibility of using these zombie botnets on the internet to send spam or anonymous voice-over IP calls from India or China that can then be used to send out misinformation and voter repression efforts or suppression efforts. So I guess I'm wondering as we shift to these technologies where candidates and parties and other groups can send out false and anonymous messages, how does that change the playing field? So, I mean, I think that we've certainly seen it in this election cycle, right? In a pretty major way. I have a survey that I have to finish up today and one of the things that I'm asking about is where people receiving forwarded emails. In the past, when we've asked about have you received a political email, we just did it very generally. But it's important to know the difference between whether it's a voter-to-voter personalized message versus this forwarded chain mail that is coming from a source that we don't necessarily know. And we don't know. So we do know in terms of just political persuasion, that number one, negative messages are more effective than positive messages. That messages from someone you know are more persuasive than from someone you don't know. But we don't yet know, right? If that negative message comes forwarded from someone you know, but the original source you don't know, do people discount it or can they even distinguish? Because as I'm sure some of you saw these chain emails, they were actually written so that it looked like it was just forwarded once, right? A friend of mine sent this along and that same tagline was used to go all over the country where my sister's aunt told me that Obama had such and such and it was worded in such a way to try and make it look much more personal and familiar than... I think it's an area that we don't know enough about but has absolutely changed the dynamic. And I think that there's not even a consensus on from the campaign strategy side, whether you ignore, acknowledge, counteract, what the appropriate response is and I don't think there have been any internal studies that... It's interesting that you mentioned that. My colleagues at Indiana University in a study where they sent fake emails to people pretending to be from their social network and friends, so they mined Facebook, they typed in people's social graphs and then they sent phishing emails to people trying to trick them into really using them as passwords and they saw a 400% increase in effectiveness of phishing attacks. So if you really wanted to see if the campaign could mine Facebook and then send false forwarded messages to people pretending to be from their friends, see if that would have... Sorry, I'm not a bad guy, but see if that would be more effective. Yeah, that probably wouldn't be approved by the Human Substitution Review Committee at Harvard, but it's a great idea. I've been dealing with this in my work with indigenous populations across countries and the way I have handled it and it's not perfect, but it's a teaching standard where you verify possibly responding and they look for alternative media, actual media where you place a call into one of the campaigns and that becomes the teaching process and as much as we all love the internet and believe me, I do and I use it, it has been very helpful. For this campaign, I found out that on the MSN website for Massachusetts, they had the wrong clothes of registration for becoming a registered voter. I found it on a Friday, it was fixed because I knew what the process was and if we start teaching the kids what they start using the internet and using the emails that because of technology, there are a few gaps in this. You want to trust friends and double their funds. From a newspaper, a phone call, something else, a radio program, something else because that has not all been coordinated to be manipulated. One of the things I'll be able to see is one of the projects I'm working on right now is it's been a multi-wave panel over the course of the campaign tracing individuals starting a year ago and so I've been asking every wave, we're now in wave 10, have you been to a website? Have you been getting these emails? And so we can look to see if, for instance, there's more error, we asked one question was, what's the first word that comes to mind when you hear Obama, when you hear McCain? And we can look to see if there's more references to some of the negative rumors among those people who were more active online and those weren't, who weren't. And I haven't looked to that yet, but I intend to. Here's your header. Thanks. So there's been quite a bit of debate about increasing political polarization within the US. Bill Bishop's got a book out, big sort, basically announcing that none of us actually know our neighbors anymore. We've all sorted ourselves into homogenous districts. We're actually pretty good evidence of this. It's pretty hard to press to find McCain t-shirt or long sign or anything around here. You seem to be suggesting that that's wrong or that it's more complicated than that. I think you're absolutely right that the way that I would pitch the story is that it is absolutely the case that our campaigns, our candidates, our politicians have polarized. And to some extent they are talking about the cultural issues and polarization because there's an incentive to, not because the public has polarized. And in fact, when I, and I didn't go through all this evidence but in the persuadable voter, what I show is that the issues on which people disagree with their party, people are more likely to disagree on social issues like abortion and gay marriage, instance all research, than on economic issues. And that means that there's more of an incentive for candidates to talk about social issues than economic issues, even though most people might want to focus on economic issues. And so there's this perverse incentive structure that's created because of the technology but also even just the electoral system. And so I think that by and large what you find is that partisans are correctly sorted. You agree with, if you're a Democrat, you agree with the Democrats more than the Republicans. You agree with them more on economic issues than social issues. But most people, still there's something that they disagree with their party on. An opportunity for a smart politician here somewhere to break out of this trap of increasing polarization, which seems to be the strategy that's coming to play. Let's ignore presidential for a moment and deal with House of Representatives, for instance, where we see increasing evidence of ideological polarization. If you're suggesting that there's actually a large mass of voters and perhaps a majority of voters who are not in lockstep with their own party, isn't there the possibility for someone to show up and say, hey, I'm really moderate. And I listen to people and I get along and I might sort of change my mind and I'm squishy and mold me in your own fashion. Why aren't we, if they're right about persuadability, why aren't we seeing that? What we saw in 2006. So keep in mind that the micro targeting really got started in 2004, a little bit 2002. But 2006, what we saw is that the way that Democrats often took wedge issues off the table was to put up candidates who were pro-life, right? If that was going to be a potential issue that could be raised in their district, then they tried to take it off the table. And that's less likely to happen in the presidential level, but even there, we've kind of seen it this cycle. And it's something that frankly surprised me and a lot of other political scientists that you have McCain and Obama much closer on immigration than you would have expected given the potential for that to become a wedge issue. And so you certainly, at the congressional level in particular, that if it's clear that your potential coalition is vulnerable to some type of wedge appeal, that you can undercut that by adopting a more moderate position. I'll leave it a little bit to your question. I have two questions about how do you see the way that wedge issues evolve over the campaigns? Because one thing, I guess the particular ones that have been really striking to me, one is gay marriage that you had Sarah Palin and Joe Biden both being like, yeah, well, fine, whatever, cool. And I mean, that's really different from the tone, at least, of the national exposure of that in the last election. The other one being in the longer past, welfare and what used to be a huge, huge wedge issue just gone really as something. And then also social security, which is more of the technical issue, which was a huge buzzword in the 2000 election is just not at all discussed right now. It seems like even though it's still a really huge problem. So I'm wondering about those, maybe specifically the evolution. And then as an even broader question, I'm wondering about your perspective as a political scientist and an academic researcher, looking at this and not having a concern necessarily about the partisan outcomes, but about governance outcomes in general and looking at what are the recommendations that come out of this sort of research about mass communication, about issues and about governance mandates and how do we understand what the public wants, what's good based on election like this. Right, so let me take the first one first because actually both are terrific questions. So the emergence of an issue is very interesting and I traced a few in the book. I think stem cell research is a nice example. So the evolution of stem cell research was in 2000, nobody knew anything about stem cell research. It was never talked about. It was not mentioned in the debates. It was not mentioned in the platforms. It was not mentioned on the candidate websites. There was no mention of stem cell research except that Bush sent a letter to Catholic bishops saying that he was opposed to stem cell research. So he made the first move. He made the first move in an appeal to Catholics, a traditionally democratic constituency that he was trying to pick up by saying, aha, you're the only people out there. They're thinking about stem cell research and abortion and other things but that was the really the first emergence of stem cell research in the campaign. Nobody knew about it. Poll numbers were finding that 50% of people said they didn't have enough information to say if they had an opinion or not. This emerges as an issue over the course of the first term where there are congressional hearings, there is movement in terms of policy but at that point, Bush has wedded himself to a particular position. If we look through all the internal documents about whether or not he was going to kind of stand by his opposition to stem cell research and one of his very first television addresses to the nation was to say that he was going to cut off funding for stem cell research except for existing lines. The very first, right, and it was actually debated whether or not that was what he should do because as the issue started emerging, as the public started to become aware of the issue, it was clear that he was on the wrong side of the issue and yet he was already wedded to it because of a move that he made in the campaign. This created the potential for a wedge issue for the Democrats. So what do we see in 2004? Somebody who was speaking prime time at the Democratic Convention was Reagan's son and he explicitly said, right, I'm not making a political speech here, I'm making a speech about stem cell research. Yeah, okay, it was a political speech. It was an appeal to those pro-stem cell research Republicans who now, right, this was a wedge issue. And so I did the same thing on racial policy, civil rights policy. But the emergence of an issue, it depends, right? You know, some things there is, you know, gay marriage, for instance, there was the Massachusetts court ruling. It's also the case that there are the interest groups who are trying to force issues onto the public agenda. So ballot initiatives are a new tool that interest groups and activists use to try and get an issue that they think is going to be an advantage on the ballots and into the public discussion. Minimum wage was one that the Democratic activists were pushing on to state ballots and of course gay marriage is one that Republicans were. And so it creates this interesting dynamic where as the public either reveals their opinions or comes to learn their opinions, that you come to find out whether or not something is going to be a winnable wedge or not. Gay marriage, I think this particular election cycle is, I think, interesting from the fact that, especially in broadcast messages and television, that there is such a sense that people don't want to talk about little issues, right? They don't want to talk about superficial politics, that there is concern for the backlash that, wait a second, don't talk about the thing, like focus on the things that we care about. And so you still see it in the ground or communications, but I think the candidates are being, and more than we've certainly seen in the last couple of election cycles, you know, you have the issues like immigration and I mean, the candidates were the same on gay marriage last election cycle too. But okay, so the other question was about a really tough one that the political philosophers deal with all the time, right? And that is, you know, what does this all mean normatively, like in terms of democracy and what we should take away from an election? There is no doubt that elections have always been a very blunt measure of the public will. And candidates and the winning candidate supplement information from an election with information, right? Like both sides are saying, look, there's a chance I can win because look at how many people are at the rally and what the people are saying at the rally. Well, the rallies are not a representation of the entire nation. And so just to not really answer the question, maybe to highlight something that's important to me and that is survey research, that one of the things that we see that is critically important has been for many decades is that polls are one of the few ways that everyone has an equal voice, right? Is that you can find out public opinion. But it's important to remember that at the congressional level, the president is the only person who has the entire nation as his constituency. And what matters for policymaking, right, are the constituency preferences, the people that actually go and show up at the polls and what their opinions are in individual states and individual districts. And so policy outcomes, there are many different actors who are all interested in the same thing, which is reelection, but it sometimes is going to pull them in different directions. And so looking at the kind of prospects for governing for the next president, they're pretty crummy. And that is even if Obama was to get a Democrat, keep a Democratic majority in Congress. Of course, the important thing is whether or not there's a filibuster proof Senate. Even if you get a filibuster proof Senate, so you get the 60, you then have to contend with the fact that there are a lot of Democrats, conservative Democrats who are not going to go for, you know, liberal policy changes. And so, and we saw this, you know, Clinton had a Democratic Congress when he was first elected and the public, I mean, so one of the questions that political scientists are asking now is, you know, really who wants to take, take over at this point because you're really going to, there's so many challenges and so many hurdles to kind of actually making policy change. Yeah, I don't know if that helps to answer, gets everyone depressed here on election day. I have a question from a European perspective. So I think this kind of micro-targeting and we don't see that as much in European countries. And I guess one of the reasons is that we have privacy regulations that put quite heavy restrictions on processing of data about people's political opinions. So there is, however, like a growing industry of campaign experts that are looking at the US situation. I think this would be, of course, a great thing to be able to micro-target much better. And I can see in like, at the moment we have kind of like the start of the renegotiation process about this privacy directive. And I can see that probably also this industry is going to try to get some kind of, like a lower standard for this process of political things. And so I wonder how does your research fit into that debate about it? So I can see the one, like you expressed your concerns about the effects of the micro-targeting. On the other hand, it does give political parties and others like a better way to connect to voters. I was wondering if you could say something about that. Yeah, and I mean, you're exactly right. That one of the key pieces is having information about individual voters. And so the key is the fact that the campaigns and parties have this information about each voter. And you could say, why is it that they get access to the voter rolls? And all the justification for giving access to voter rolls is it's good for democracy because then you mobilize, because then it makes it easy to mobilize. But what it doesn't take into account is how only subsets are being mobilized. And so there is, I think I've seen some legal scholars in fact talking about micro-targeting. I'm trying to remember who exactly it was for those of you who might be interested in this. But Michael Kang, who's a law professor at Emory, has a piece talking about how great it is that all of a sudden you can have mobilization efforts going to more voters. And we know that this is a good thing for democracy to have more people mobilized. And what it misses is the fact that money is being spent more efficiently. It's not just across-the-board mobilization. I think if there was a big push, then this could at least be more of a public debate. But the people who would have to decide are the same people who have an incentive to keep the information, right? But it's not only the voter roll, Michael. So I had the impression that they buy in to all the massive databases that are there, like the consumer information. Yes, absolutely. But again, from the European perspective, about this pros and cons of mobilizing the base or something broader, well, basically here, if you think about micro-targeting, it's pretty much to me, at least, as a European, a way of reinforcing the structure of power, because you essentially send the message to these people there. Plus, if you use, but that's an advantage for campaigning. Plus, if you use the technology for receiving more money, then you spend this money on, well, TV advertisement, for instance, right? Again, to reinforce the message from the top. So essentially in Europe, you can't really micro-target because of the data protection laws. You can't really spend that much money on TV ads because of the financial structure of how you finance the campaign. And then, well, I don't really have my point on this, but I wonder whether there are more pros or cons for using this technology in a way which actually influences the base from the top more than it's possible in a European situation, for instance, because the structure in Europe does not really allow you as a politician to influence the government. And a lot of other political systems, you do the coalition building after the election rather than before. And so I mean, I think you're both absolutely right to point out that there are these institutional and structural differences that make for a completely different campaign environment. In terms of lining up the pros and cons, I'll leave that to somebody else. I mean, I just want to point out that there are a lot of people who just say, oh, technology, the impact has been all great. It's allowed people to become more active, more easily if you're interested in politics. And one of the things that it's allowed people to reach out to people who haven't been involved in politics before. All those things are true. And so I'm not sure that we can balance off and say which is better at the end. Is it all good or bad? I just want to point out that this dynamic is going on. And at the end of the day, the thing to also remember is people make up their mind not just on the basis of that text message or that forwarded email or even the email appeal from their friend. They're making up their mind on the basis of all the information that they learned during the campaign. And so you look at 2008. And damn it, this is an election cycle where it looks as if. McCain has not been able to really utilize wedge issues to any great extent because whatever is being said on the ground that people are still overwhelmingly concerned about the economy. And so there are limp. And I guess the key thing I want to point out is just that even if in 2008 that at the end of the day, these sophisticated campaigning techniques don't have a big impact on voters, they have an impact on the candidates. And that's an important thing to recognize. The major advantage of the system here is that after all, it rises the level of awareness among the constituents, comparing at least to the European system where pretty often people just guess going to the poll. We'll leave a party, right? I mean, especially in a parliamentary type system. There's not the same incentive structure because it's more kind of party-based. But yeah, I mean, there's pros and cons. Can I ask two kind of wrap up stuff questions? One, what do you think the biggest surprise thus far has been in the 2008 cycle and a prediction for 2010 or 2012? What do you think another, like a future iteration of this is when 2012 was too far out, but sort of recognizing the. So I'll do the first one, the last one first. And that is one of the things we've seen is beyond just merging in information from consumer databases, now companies and campaigns are looking to scrape other information they can gather about individuals. So one company has advertised that they're developing technology to scrape resumes on the internet. And so that way you can put into the database, right? Not just your turnout history and not just your magazine subscriptions, but also information from your resume that my sociology majors are more likely to be democratic and business majors are more likely to be Republican. And so again, more information. There is no turning the clock, I think, on the type of information society, information economy that we have. And so I think what we'll find is just more and more information about each individual voter is. And I think getting back to the point that sometimes it's wrong, what candidates are going to find is that they sometimes are going to run the risk that even somebody who owns a gun, they might not be all that supportive of the Republican on a different set of issues. And so they're going to run into, at some point, so much admiration. It's not clear exactly how they'll incorporate it. OK, the surprising thing from 2008. And I don't know if this is the most surprising, but I just think it's hilarious for this group. So Obama has so much money. And so he has done some things that, number one, because he knows if he can get young people out, that this is a group that breaks very much for him relative to McCain. And so he's done advertising on video games. Is it the second life? So there's a big kind of, not big, right? This is cheap relative to other campaigning. But it makes for a nice headline. And it is this great comparison to McCain who doesn't know how to work Google. So there was a friend of mine who's been working on the Republican campaign, asked about their technology efforts. And was talking about Obama's efforts on video games and stuff. And they said that, in fact, that there was a McCain avatar in Second Life. And so they had one. But the Obama supporters in Second Life had double parked in front of his apartment. And so he can't get out. And so the entire campaign, the McCain avatar, has been stuck in his apartment with the Obama supporters spray painting on the side of his apartment building. And so I think that probably this election cycle, the kind of the technological gap between the candidates, has been very striking. And I think that to some extent all the talk about young voters I think has been a little bit overblown. The turnout rate of young people is so, so pathetic that even if you have record breaking numbers among young people, it comes nowhere close to the numbers of other cohorts. And so I suspect that the headline might be there were record numbers, but when you start at 28% of 18 to 20-year-olds are voting in 2000, you can increase that by 10 percentage points. And you still are not getting up to the 52 to 60% turnout rate of their parents. And so, but again, Obama's had the money to still play and appeal to this group. And so I think that has made for some interesting campaign stories, if nothing else. So the biggest surprise is the Second Life voters saved the election for Obama. All right, so everyone, please join me in thanking Sunshine for her love. Thank you. Thank you.