Metaxas is a professor of computer science who researches the way the web is changing the way we think, decide, and act.

Interview by Amita Parashar Kelly ’06

Can social media predict the outcome of elections?

What people and candidates say on social media sites like Facebook and Twitter does not predict the outcome of elections. It’s maybe a little closer than it used to be, but nowhere close to actually having any predictive power. Traditional polls depend on good representation. Social media is a different species: It’s an open forum where anybody can say whatever they want, as they want.

In computer science, we have the so-called incomputability theorem that says there are some things you cannot compute. Even if social media were able to predict elections, there would be so much interest in the influence of these mediums that it would become unpredictable. You know, it’s not a real theorem, but it gives us a sense of why it’s not going to work.

What [social media] is used for, very effectively, is to cheer your troops, to make them feel really excited, and to make them feel that they have a purpose and that you listen to them, as opposed to just measuring how many [followers] you have. If you like to hear from [a candidate] from time to time, when he or she says we have to do something, it can enhance your desire to support the candidate and make you more likely to go out and volunteer. It can also make you more informed, which is important so that you could have, and win, an argument—which is what we humans care most about.

My current research has to do with a tool [we developed] called TwitterTrails: We have found some remarkable ability in actually figuring out, through the power of the crowd, whether something that is spreading is true or false and what the level of skepticism is. It’s based on the idea that when people are not emotionally charged, it’s unlikely that they will try to fool you. They might either choose not to repeat what you said and not propagate it further, or they might question your motives and insert some kind of skepticism. But when it comes to elections and politics, that’s not the case. We see huge polarization for the American elections.

You can also expect a lot of excitement on social media just before the elections. If you spread a lie just two days before elections, chances are it’s going to work. So I am sure there are zealots out there who are designing their strategy. In particular, they try to do what we call micro-targeting—try to find the 40-year-old who lives in a contested area of the country, and try to influence that person.

What are you watching for 2016?

It is early to make any kind of prediction on 2016, but we are collecting data. More generally, I’m curious to see how the large field shakes out. I’m interested in trying to study the moment in which suddenly, a candidate will start rising above the rest. I’d like to know how these changes might be represented in the way they communicate, and I’m interested in knowing what contributes to that moment. I suspect it’s going to be some national news or event, but it could be something else.

Hahn, associate professor of political science, researches civic and political engagement, as well as environmental politics.

Interview by Amita Parashar Kelly ’06

Is field organizing making a comeback?

In the 19th and early 20th centuries, campaigns were focused almost entirely on field organizing, because that was all they had. There was no direct mail. There was no television advertising. If you ran a campaign, you basically pounded the pavement.

Around the mid-20th century, TV became a bigger force in politics. In the late 1970s, we saw the advent of direct mail. All of a sudden, mass communication was reaching lots of people more inexpensively and easily than going door-to-door. TV ads became much more popular, and a lot of campaigns focused on trying to raise the dollars they needed to buy big ads or send out mailers.

Around 2000, campaigns started asking whether all this mass communication was having the impact they really wanted. If you look at the research, TV ads might have an effect in the first day or two after people see them, but the effect wanes quickly. Republicans were really leading a lot of the work on this shift [back to field organizing] in 2000 and 2004. In 2004, they had a big volunteer operation, trying to get out the vote in the last few days of the campaign.

When Barack Obama came along in 2008, his campaign rode this growing wave of greater interest in field organizing. And Obama himself had this background as a community organizer. The campaign was able to effectively meld principles of community organizing into an electoral context in a way that hadn’t been done before.

‘What I’ve been watching, and will continue to watch, is the development of [each candidate’s] ground game. Where are campaigns putting their field offices? How many field offices are they putting in? How are they stacking up?’

—Hahrie Hahn

When the Obama campaign first got on the ground in Iowa, they realized if they used the usual campaign tactics that everyone uses—identify the party faithful in Iowa, have Obama do sit-downs with them, and then try to get them to support the candidate in the caucuses—they were going to lose because Hillary Clinton had the entire Democratic Party pretty much wrapped up at that point. So in the 2008 primary, all the different primary states became little laboratories of experimentation. What the campaign found was that the strategies that were most effective were from community organizing—relying on volunteer leaders, developing the commitment and leadership those volunteers, and then using them to organize their neighbors in their own community. The whole philosophy behind the campaign was that, win or lose, they were going to leave something behind in the community. So it wasn’t just that they wanted to get Obama elected, which of course they did, but at the end of the day, they wanted to have an infrastructure of volunteer leaders and activists developed through the campaign that would remain in the communities.

What are you watching for 2016?

What I’ve been watching, and will continue to watch, is the development of [each candidate’s] ground game. Where are campaigns putting their field offices? How many field offices are they putting in? How are they stacking up? What is the extent to which they’re relying on staff to run their field campaign versus trying to recruit volunteers?

Jeffries is an associate professor of American studies whose research focuses on race, gender, politics, identity, and popular culture.

Interview by Gail Russell Chaddock ’72

Is “Black Lives Matter” activism making a difference in campaign 2016?

One thing you’ve seen already is a greater willingness to address criminal justice and punishment reform than we’ve seen in previous election cycles. This is a direct result of the activism around the Trayvon Martin murder, the Eric Garner murder, Ferguson protests, the nonindictment of Darren Wilson, all those things.

Everyone understands that all lives are valuable. But what Black Lives Matter activists are calling to our attention is that there are specific injustices or problems that the black community is suffering from that other communities are not in the same way. To have someone like Bernie Sanders affirming black lives matter, speaking the name of Sandra Bland, that sort of rhetoric during election cycles has been notably absent in the past.

Remember that Bill Clinton during the 1990s made quite an effort to convince voters that he was just as tough on crime as his Republican opponent, because he was governor of a state that had the death penalty. He got a crime bill passed that many of his advisors have admitted was a catastrophe for criminal justice and racial equality in this country.

The phrase “Black Lives Matter” first appeared in a tweet from Alicia Garza, a longtime activist and advocate based out of the Bay Area in California. … You shouldn’t be surprised by some of the extreme and ugly portrayals from the far right media of the movement. But even more moderate outlets haven’t covered it as accurately as they should.

In some ways, I understand the confusion and focus on the phrase “Black Lives Matter.” It’s not that these activists are totally unconcerned about anyone who falls outside the category of blackness. They are focused on the unique forms of oppression that black folks are subject to, and they believe that black liberation is a prerequisite for social progress. But in other ways, we can view such resistance from conservative and mainstream media as a distraction from the concrete demands and institutional changes so desperately needed to achieve a more sensible and just society.

The cell phone and social media have been essential to these movements. Look at the University of Missouri protesters: The students actively refused mainstream media coverage and trusted they could get their message out otherwise through access to social media. It challenged the power of mainstream news to set the terms of the debate. It challenged the power of the campaign staff to set the terms of the debate, as well.

What are you watching for 2016?

Right now two things have my attention. The first is the extent to which Sanders has been able to articulate what he means by democratic socialism. Young voters have not heard this articulation of democratic socialism. It’s a simple principle that the disparity in wealth and social inequality—the distance between the haves and have nots—we have to understand as a sign that something is deeply wrong with the way our economy functions and, by extension, what our cultural values are.

On the other side of the ticket, [what has caught my attention is] the sort of naked intolerance and disrespect that have come out of the mouths of people like Donald Trump and Jeb Bush, who says that he favors forcing refugees to prove their Christianity before being admitted to the country.

Burke is a professor of political science who teaches a course called Health Politics and Policy.

Interview by Amy Mayer ’94

To what extent will health-care policy play a role in the campaigns and in voter decision-making?

It’s typically not one of the top issues. It’s kind of unusual when it’s a top issue. And arguably, where it tends to be an important issue is where a Democrat’s running and uses it as an issue to galvanize the liberals. So, like [Bill] Clinton in 1992 and then Obama in 2008, to some extent. That’s when it’s important.

Obamacare is such a complicated thing that I’m not sure anyone other than nerds really has much of a sense of what it really is. Some [candidates] might say, we have to start over, but we’ll take some parts of it. That’s what a lot of Democrats in more marginal districts say, if Obamacare is an issue. But to be honest, I’m not sure it’s going to be much of an issue in this election. It’s going to be overshadowed by a lot of other things. I don’t think Republicans are going to run on it.

I don’t think most Republicans really would like to repeal it. This is somewhat cynical, but if you got rid of it, the next day it would be chaos. It’d be chaos not just for all the people you’d be cutting off, tens of millions now, but it’d also be chaos for all the people in the marketplace, businesses and insurers that are premised on it continuing. A lot of business enterprises and nongovernmental nonprofits have put a lot of resources into it. I’m not sure Republicans would want to be responsible for [ending] that. Also, most of them haven’t come up with an actual plan. They talk about repeal and replace, but look around for the replace part and you’ll find it’s pretty meager.

Political scientists will tell you that always the No. 1 issue, unless we’re at war, is the economy. Political scientists have found that if you ask people whether the economy is going up or going down, people are pretty sophisticated about that. And when things are headed down, the people who are in the middle, who are not pledged to either party, they tend to vote on that because it gives them a sense of how the incumbent’s doing. If the incumbent party, they think, is doing badly, they’re going to punish the incumbent. Or if they think the economy is on the right track, then they will reward the incumbent.

‘We’re in an amazing period in American political history. We’ve had a bunch of really competitive elections in a row without a dominant party, and we had a period over the last 40 years of divided government.’

—Tom Burke

What in particular are you watching this year?

We’re in an amazing period in American political history. We’ve had a bunch of really competitive elections in a row without a dominant party, and we had a period over the last 40 years of divided government. For a few years of this, it was, “The Republicans own the executive, and the Democrats own Congress.” And now, it’s “Democrats own the presidency, but they struggle with Congress and are really doing very poorly at the state level.” They only have unified control of seven states. Republicans have unified control of 23 states. So at the national level, when people say, “The Republican Party’s on a disastrous path,” well, really? I mean, they control Congress, and they have a good shot at retaining control of Congress. And then at the state level, they’re cranking.

So it’s a really interesting situation, and what is going to change it? What is going to either lead the Democrats to do better at the congressional and state level or on the other hand, what is the Republican Party going to do to finally start attracting a national [presidential] majority that they haven’t done up until now?

Pattanayak is a statistician and director of Wellesley’s Quantitative Analysis Institute.

Interview by Amy Mayer ’94

In 2015, Gallup, the venerable polling operation, announced it will not conduct surveys to predict presidential primary winners. Instead, Gallup polls will focus on issues. What drove them to that decision?

One of the reasons that Gallup is reviewing its policies and its priorities is that in 2012, they got the answer really wrong. They concluded based on their review that their method was the best method that is possible right now: a telephone survey that attempts to reach people who tend to vote, and tries to be representative. Gallup also concluded its sample size was not the issue. And still they decided to back off and focus more on the issues than making predictions. My opinion is that’s a reflection of the fact that it’s just getting harder and harder to do a good poll, mostly because of changes in how we use telephones.

The history here is that in 1936, Literary Digest magazine conducted a very large survey. Gallup conducted a much smaller one, but it was more representative. Gallup made the correct prediction (Franklin Roosevelt was re-elected). That was the beginning of the Gallup organization’s leadership in the polling industry. There’s a trade-off any time you gather data between having a big sample size, and having representativeness.

‘Every person who votes should make the decision based on what they think of the issues, and not based on who they believe is most popular among all the other voters.’

—Cassandra Pattanayak

The Literary Digest poll tended to include people who were more wealthy and had telephones. At the time, that was not a good way to go about getting a data set. Then telephones became very popular, and for many, many years, a telephone survey was a great thing to do. But now in 2015, 2016, it turns out that’s not such a great thing to do anymore, because it’s no longer the case that every household in the U.S. has exactly one telephone.

Most surveys, if you look at the footnotes, include about a thousand people. And they work very hard to get surveys that are representative of the U.S. That’s worked for a long time. But you just can’t get a representative sample very well via telephones any more. Young people are less likely to have landlines. In households with high socioeconomic status, there are multiple phones. [And] the area codes of cell phone numbers no longer correspond to where people actually live. So you can’t even use the phone numbers for making sure you have represented all the different parts of the country. Typically, we can prioritize size or representativeness. It’s really hard to do well on both. Polls like Gallup have it right in that they are collecting small samples and working hard to be representative; the problem is that it’s very difficult to collect a representative sample of any size. And that’s the trade-off that underlies the type of problems that Gallup and other organizations are running into.

What are you particularly watching this election cycle?

I’ll be watching for media focus on the issues, and less on daily updates of the poll numbers. Sometimes people tend to over-rely on numbers because they’re numbers. There’s one particular statistic you can cite, and that is a very concrete way to summarize what might be going on. But the truth is, there are a lot of uncertainties about those numbers generated by polls. And another truth is that every person who votes should make the decision based on what they think of the issues, and not based on who they believe is most popular among all the other voters.

Just is a professor of political science who studies elections, politics, and the media.

Interview by Gail Russell Chaddock ’72

How do you assess the role of the media so far in campaign 2016?

The media have fallen victim to a problem that they’ve had before—that is, when a horse race is competitive, focusing very heavily on that. Because it’s not obvious where the GOP contest is going to go, the focus is on the Republican side. Every week, it’s another poll and another news cycle. Every few weeks, there has been another debate, which is another news cycle. So the polls and the debates have become a preoccupation.

It hasn’t been good for Democrats. They decided to schedule their debates on Saturday nights when there were no viewers, and so they didn’t get the coverage that the Republican debate had. The Republican debates occurred so frequently that they dominated the airwaves.

“Gotcha” questions were very much on the minds of debate moderators, as different channels grabbed the opportunity to pander to themselves and highlight the moderators. Jake Tapper, before the CNN debate, said he wanted to get the candidates asking questions of each other and attacking each other because he thought it would be great television. I didn’t think it was great television. It didn’t necessarily help the audience understand what was at issue. When the media questioners were called to account by Ted Cruz, he was able to feed on the general distrust of the media among Republican voters. The questioning backfired.

The media played an enormous role in the debates. Fox’s decision to sponsor a debate and to decide who made the upper tier and who was in the bottom tier, or “undercard,” was extremely manipulative. Not surprisingly, the “children’s table” did not get much viewership. And those candidates consigned to the undercard struggled to get any traction in the polls.

In all of the media manipulation, Trump still came out on top. The media have been frustrated. When they pointed out that Trump had lied or misstated the facts, they thought this would have a dramatic impact on his polling support. But it did not. Many people in the media cannot understand how someone could support a candidate who lies as vigorously and unapologetically as Trump. His support seems to be based on the view that he tells it like it is, and he’s tough.

Women candidates often get short shrift from the media, but I don’t think that is the case with the presidential candidates. Hillary Clinton is not being treated with disdain or lack of attention, mainly because she is the front-runner on the Democratic side. Carly Fiorina may have been better prepared than others on the undercard debate, so she was promoted.

But giving the media the right to promote or demote candidates, it’s not helpful. It certainly opens them to charges of bias.

What interests you the most about this year’s campaign cycle?

The enormous Republican field, the number of debates they’re having, and the lack of political experience of the front-runners on the Republican side. This poses an interesting set of problems for the Republican Party. There was a great deal of discontent among the Republican primary voters: They had control of both houses of Congress but they then did not get the kinds of legislation they wanted—or the repeal of legislation that they hated. That created a great unhappiness, a frustration with business as usual in the Congress, and a rejection of establishment figures. These outsider candidates have been able to capitalize on that anger and disappointment on the Republican side.