Election 2012 Roundup: Four Faculty Members on This Year's Campaign

Meghan Berry
Tanya Domi
Georgette Jasen
Adam Piore
October 17, 2012

By any measure this presidential election will be close. Battle lines are drawn on the main issues defining the race: the economy and jobs, health care reform, taxes, the deficit, education, foreign policy and terrorism, and more. The Record talked to faculty across the University to gauge the broader trends of this election cycle and to look beyond the issues to the actual mechanics of voting and campaign craft.

Nathaniel Persily, Professor of Law and Political Science, explains the impact on the race of voter ID laws, absentee ballots, voter registration and redistricting; Political Science Associate Professor Dorian Warren discusses key differences between the 2008 and 2012 presidential elections; Computer Science Professor Kathleen McKeown talks about software she created that can detect how people react to specific news events; and Political Science Professor Donald Green describes how his field research unleashed the armies of canvassers who go door-to-door in battleground states to get out the vote. 

Q. This election cycle has seen a great deal of controversy over voter identification laws. How significant are they in the election process?

Because Democrats and civil rights groups have been extremely successful in getting voter ID laws struck down or suspended for this election, they are unlikely to have much of an effect. Even in those states where we have seen voter ID laws applied in recent elections, the effect on turnout has not been great. For example, in Indiana and Georgia, which adopted Voter ID laws prior to the 2006 and 2008 elections, voter turnout in those states increased, even among African Americans. In other states, where a lot of people don’t have drivers’ licenses, such as Pennsylvania, perhaps ID laws would have had a greater impact both on aggregate turnout and turnout of certain groups. I’m more concerned that these laws are going to be enforced in an arbitrary way, that there’s going to be very spotty enforcement, and there will be long lines at the polls and confusion. If anything, such laws may lead to greater numbers of provisional ballots and absentee ballots being cast, given that absentee ballots do not require ID. Because fraud by way of absentee ballots is much more common than so-called “voter impersonation fraud” prevented by voter ID laws, the ID laws could have the perverse effect of increasing the risk of the fraud they are supposedly intending to prevent.

Q. What is the impact of more absentee ballots?

Absentee ballots are the ticking time bomb of our electoral system. They are cast with errors at a higher rate than in-person votes, andthey are more ripe for fraud than polling place voting. There are states where we have seen absentee ballots with error rates in the four to five percent range, partly because of the steps involved to ensure that a vote is cast and counted correctly. Although fraud of any sort is relatively rare in our elections, prosecutions for absentee ballot fraud are much more common than most other types of fraud. While it’s very difficult to get hordes of voters to go from one polling place on a single day to the next impersonating someone else, it’s much easier, for instance, to go to a nursing home over a week and gather 50 or 100 absentee ballots and have an impact on an election.

Q. Is there a potential for fraud in get-out-the-vote campaigns by special interest groups?

The U.S. has outsourced the process of voter registration. Political parties or interest groups are the organizations going out and registering voters. Interest groups often pay people per name they register to vote, which can lead individual workers to just write down names. Voter registration fraud is different from voter fraud, but it does clog the rolls and create administrative difficulties. One of the main factors depressing American voter turnout is a registration system that puts the burden on the voter to re-register every time he or she moves. In other countries, the government takes the initiative to register all voters and keeps better tabs on eligible voters. Unlike other countries, the United States does not have a national list of citizens or a national ID card. This arises from our tradition of respecting voters’ privacy, but it makes voter registration more difficult.

Q. What is the impact of recent redistricting?

For this election, I don’t think redistricting will have a major effect on increasing one or another party’s share of seats in the House of Representatives. For the most part, the effects have been a wash between Democrats and Republicans. However, at the margins, because the Republicans in the 2010 elections were so successful in gaining seats in state legislatures, they have used the redistricting power that they had this time to draw districts that protected their incumbents. Another important area of redistricting that’s going to get a lot of attention over the next year has to do with the issue of race and redistricting. The Voting Rights Act passed in 1965 has provisions that regulate redistricting, including Section 5, which applies mainly in the South. Those covered jurisdictions have to get permission from the federal government for any laws that they pass with respect to voting, including redistricting. The Supreme Court, likely this year, is going to consider the constitutionality of Section 5 and basically ask whether things have changed sufficiently in the last 50 years to justify striking it down as unconstitutional.

Q. What impact will the money pouring into this year’s races, estimated at about $1 billion, have on elections in the long term?

This election will be known as the "Citizens United" election, named for the Supreme Court decision that opened the spigot for corporate spending on candidate-specific television ads. I tend to think the effect of that decision itself is overrated. Most of the big money coming in is from rich individuals or their closely held corporations. Large publicly held corporations have not been as active, although much of the money spent in this election is coming from undisclosed sources. "Citizens United" was in some way a decision before its time. That case was about a video made available on-demand for downloading on cable TV. As the Internet becomes the main mode of political communication, it’s going to be impossible for any government agency to regulate these types of videos, ads and other forms of communication. It’s going to become increasingly difficult to regulate the flow of money into politics.

Q. Your research on field experiments is said to have changed political science research, and election themselves. In what way?

My 2000 paper (co-authored with Alan S. Gerber), “The Effects of Canvassing, Telephone Calls and Direct Mail on Voter Turnout,” revived a long-forgotten research method of field experimentation. We created a number of nonpartisan get-out-the-vote messages and used them to see what boosts turnout. Before then, I had done extensive research in campaigns, but this was my first foray into experiments. The paper, which showed the effectiveness of old-fashioned face-to-face mobilization, sparked a cottage industry in political science and revived the parties’ interest in door-to-door canvassing. It was after that paper that you started to get the roving armies of canvassers going to battleground states.

Q. You also did research on the efficacy of direct mail, using four postcards to study the impact that different messages have on voter turnout. What did you find?

The idea was to progressively amp up the social pressure applied to voters. The basic mailing encouraged people to do their civic duty. The next announced that they were going to be part of a study that would monitor whether they voted. The one after that told people that voting was a civic duty, a matter of public record, and displayed their own voting records and the records of others in their household. The final, which was the most heavy-handed, not only displayed voting records of the recipient and others in the household; it also displayed the voting history of the neighbors. That was the one that produced this enormous effect. Voter participation was about 8 percentage points higher, and people in that study not only voted at higher rates in that year, they have also voted at higher rates in subsequent elections.

Q. Is there any validity to the arguments we see around poll results?

Arguing about polls is an inevitable feature of any presidential election. There is no question that it has gotten more difficult to poll with the advent of the cell phone. Response rates are falling, and it is becoming increasingly difficult to draw random samples of likely voters in battleground states, because voters with reachable phone numbers are also getting a vast number of calls from campaign-run phone banks. Pollsters are all wondering when traditional polling methods will misfire and generate inaccurate predictions. Standard methods may perform adequately in 2012, but we are probably going to see a fundamental shift in polling methods during the next couple of election cycles.

Q. Last year, you did a field experiment on the effects of televised campaign ads. What did you find?

We randomly assigned the volume and timing of TV and radio ads in 18 Texas media markets and then gauged opinion every day using tracking polls to see how rapidly one could detect effects and how long-lasting those effects would be. What we found is that you did see a fairly substantial effect, roughly a 5 percentage point gain in a trial ballot. This study basically silenced the critics who said you could never do a field experiment inexpensively and quickly, and get fairly powerful results on both the impact and enduring effects of a campaign.

Q. It seems that the Obama campaign is employing the face-to-face approach to a greater degree than Romney. Have you found that true?

In 2008, Obama ran a textbook campaign. The Obama campaign of 2012 will implement many of the same canvassing tactics, perhaps with a little bit less enthusiasm from its activists. On the Romney side, much of the canvassing will be left to allied organizations funded in large part by super-PACs. Conservative Christians announced that they were hoping to knock on two million doors. Both presidential campaigns will still spend the bulk of their money on mass media, but in the closing days of the campaign, they will focus increasingly on the slow but steady votes that can be generated from massive on-the-ground efforts.

Q. What kinds of experiments are you thinking about now?

I think that the great frontier is to study large-scale mass media advertising in the context of presidential campaigns. My suggestion would be to randomly assign different amounts of media to different markets and examine the extent to which opinions move in those markets, the extent to which donations move. To study not only the effects of one advertisement, but perhaps a series of ads that are designed to counteract your opponent. It’s only a matter of time before people start doing exactly what I just proposed because the stakes are so high.

Q. Tell us about your project and how it relates to the election.

Over the course of the last year, we have been collecting data from online political forums, Twitter and different kinds of online discussions. We have also collected news. The program we have created identifies influencers. We look to identify who was able to change people’s minds and how; where sentiment shifted; and what arguments people found persuasive. And we look to correlate shifts in sentiment with events as reported in the news.

Q. How does it work?

The systems detect influence using several programs that perceive components of influence. For example, one program can detect in online discussions whether a participant feels positively, negatively or neutral towards a topic and whether other people on the forum agree or disagree. We use machine-learned algorithms that recognize sentiment based on the way language is used—for instance, negative or positive expressions such as “I like” or “I hate.” The computer uses all of these cues to identify that this person is in favor of a particular claim or this other person is opposed.

We have another algorithm that can detect whether someone is making a claim about a particular topic. And we can detect when someone is attempting to persuade people to go along with their opinion. Then we can look at how those arguments change the sentiment of others or the direction of the conversation.

When we put them all together, we can determine whether somebody is being influential and when they change other people’s opinions. We are still building these systems and plan to look at other kinds of information, such as age and gender.

Q. How would this relate to the national discourse?

We might use it to analyze how people reacted to particular events, which events were important in determining opinions, and how in online social media, influence played a role in determining outcomes. So we might look at whether changes in opinion are correlated to a particular event in the news, like the recent attack in Libya.

There might be a discussion about health care on a forum, and you might have two people taking different sides about whether Obamacare is helpful or not. By looking at what claims are made and who agrees or disagrees, we can identify whether one person was able to influence other people.

Also, we can look at the influence of Obama or people who work for him, what they are reported to have said, how often it was mentioned in the forums and what effect that had on people’s sentiments.

I would note that our programs are not perfect, they have error rates. These are hard problems, and they are not easily solved. Our error rates can go from 10 to 30 percent depending on the component and the data.

Q. What is your overall goal?

Ultimately, we want to be looking across quite a few forums to see what is happening. We can aggregate that information to find out how influence is expressed in different genres, which forums are most influential, how influences changes over time, who within a forum is most influential and how do they get people to agree. Ultimately, we want to be able to detect influence for a variety of purposes such as advertising or online problem solving.

Q. How did you get interested in this field?

I did my undergrad at Brown in comparative literature. So I had an interest in language at that point. I did both French and English, and looked at all time periods. I loved James Joyce and Baudelaire. A lot of my friends were doing computer science and were very enthusiastic about it. So towards my senior year I began taking course in computer science. I learned at the end of undergraduate that there was a field where you could combine computers and language. I heard University of Pennsylvania had a great program in it, and I went there to do a PhD with Aravind Joshi one of the pioneers in the field of natural language processing.

Q. What are some key differences between the 2008 and 2012 presidential elections?

There are three main differences. Organized labor is much weaker now than it was in 2008, especially in key states like Wisconsin and Ohio, where there have been attempts to roll back public employee collective bargaining rights. The second major change has been the role of money. Since the 2010 Citizens United decision, we’ve seen a wave of money come into this election cycle. The third difference is that voters—especially black, Latino and youth voters—are much less enthusiastic this time. The Obama campaign has a huge challenge trying to reclaim the excitement and the mobilization of 2008 and getting them to come out again in 2012.

Q. Who is investing in Obama’s campaign this time around?

Organized labor spent more money than they ever had to get a president elected to the White House and to get a Democratic Congress in 2008. Labor is probably going to end up spending more money this time to keep pace with the super PACs that have emerged mostly on the right. If labor spent about $150 million in the 2008 election, we can expect upwards of $250 million, maybe more, spent on this election.

Obama had a lot of big donors from Wall Street last time. But with the loss of much of that support because of financial reform and a perception that he’s antibusiness, it’s unclear where he is going to raise the additional money. And even though the Obama campaign raised a lot last month, overall the amount of super PAC money supporting Romney dwarfs the amount supporting Obama. So in terms of total money to be spent in this election, Obama is at a disadvantage.

Q. What role do minorities play in this election?

In key states, they’re going to make all the difference. Black voters are heavily Democratic, and they support the president upwards of 90-plus percent. Latino voters are around 70 percent in favor of re-electing Obama. Nevada, for instance, has a lot of Latino voters. Obama will need an overwhelming majority of those voters if he wants to win the state. The same goes in Florida, where there are a lot of Jewish voters who vote 70 percent or higher for Democratic candidates. In North Carolina, he needs exceptional turnout from African American voters.

Q. What role does organized labor play in this election?

The labor movement is one of the core constituencies of the Democratic Party and has been since the New Deal era in the 1930s. It serves as the party’s chief mobilizing force. For Democrats and for President Obama’s reelection campaign, labor is the biggest source of troops on the ground, as well as money. In a two-party system, the Democrats have traditionally been the party of labor, and the Republicans have been the party of business. Mitt Romney’s position is essentially the Republican position. Over the last few decades, American unions have lost hundreds of thousands, if not millions, of members. A lot of that is due to the shifting nature of our economy—we used to be a manufacturing economy, and now we’re mostly service-based. At the same time, unions have also lost political power. The recent wave of legislation taking away collective bargaining rights from public employees in Wisconsin is an example. Having said that, American unions still punch above their weight in the electoral cycle. Even though unions only represent about 12 percent of the American workforce, they make up about 25 percent of the electorate.

Q. What has your research on Wal-Mart taught you about business in politics?

I’ve been working on a book about Wal-Mart, especially Wal-Mart’s attempt to enter big cities like New York and Los Angeles. In the 1990s, Wal-Mart avoided politics altogether. But in the last decade, we’ve seen it become much more engaged in the political system—making contributions to candidates and a lot more lobbying efforts. Wal-Mart is, in many ways, illustrative of the broader role that business plays in politics. We’ve seen lots of businesses get much more engaged and put much more money in the political system. It creates quite a different dynamic when you have a weakening labor movement and a stronger business sector engaging in politics.