Conversation: 5 questions for Jon Krosnick


The Stanford political psychologist gives his take on this year’s extraordinary election

By Sara Martin

October 2016, Vol 47, No. 9

Print version: page 27

This is a very striking, even shocking campaign that has set political psychologists back on their heels, simply because we haven’t seen a phenomenon like this,” says Jon Krosnick, PhD. “This year, we can’t rely on our long-standing assumptions about the way Americans will make their decisions.”

Teasing apart how people form political attitudes is Krosnick’s forte. As a professor of psychology, communication and political science at Stanford University, Krosnick has explored many of the factors that influence voter behaviors.

The Monitor asked him to share his insights on this year’s wild ride.

What is particularly striking to you about this election?

What’s really amazing is that these two candidates were selected as a result of the normal nominating process, yet somehow both parties have ended up with candidates with higher negative ratings than any other major party nominee in the history of polling. This is like walking into a restaurant and seeing nothing on the menu that you want to eat, but you’ve just finished a 15-mile hike and there’s no other restaurant for 20 miles, so you’ve got to eat something.

As a result, it’s going to be really interesting moving forward, particularly because the approval of Congress is at an all-time low. Americans’ faith in the federal government has plummeted. Add to that the fact that we have two presidential candidates who most people don’t want, and we’re set up for lack of confidence in government and in our future as a country.

What does that mean for voter turnout?

One possibility is that there will be unusually low turnout because people feel so disappointed. But there could be high turnout, because it feels like people are talking about politics more than they normally have, which of course from a political psychology point of view—and the health of democracy point of view—is a really good thing.

It’s also possible that an outright fear of the opposing candidate will cause more people to turn out. My work with Allyson Holbrook and John Cacioppo shows that turnout is enhanced when you hate a candidate (American Journal of Political Science, 2001).

What might explain today’s support of more anti-establishment candidates?

Given the profound disappointment Americans feel about the state of our federal government, it’s understandable that they would say, “Wait a minute, Donald Trump just said what I think! He said that the country’s government is broken in language that captures what is in my heart!” As anyone who has studied psychology knows, when a citizen feels aggrieved and a leader says, “I understand why you are angry,” that citizen feels validated, and there are tremendous positive consequences for their self-esteem and sense of mastery.

You’ve also studied how polls can influence voter behaviors. What have you found?

One reason that polls influence people is because when data show that most other people believe something, there is good reason to believe that something is true. Polls also let us know when we are out of step with others. For example, if a person supports a certain candidate, and that support is unusual, he or she will be less inclined to continue to support that candidate. But if a poll shows that many others also support the candidate, the result gives credibility to the individual’s point of view.

While the appetite for polls has grown tremendously during the last 20 years, the quality of polls has declined dramatically as well. With the arrival of the internet, it’s cheap to do surveys, but the respondents in most of those online polls are not being selected from the population randomly, which means there is no scientific basis for assuming the particular respondents will accurately reflect the population. Look for example at the recent online Brexit polls not using random sampling—some had the “leavers” up by 10 and some had the “stayers” up by 10. But in Britain, pre-election polls that use random sampling continue to be very accurate. This should be no surprise to psychologists, who understand the tremendous value of randomization. But we should all focus our attention on the pre-election telephone polls that used random sampling and called landlines and cellphones. Sadly, the media have an appetite for polls often regardless of the quality of their methods, and there is more and more dissemination of unreliable results. It’s not hard to imagine that public confidence in survey research might decline over time as a result.

What aspect of elections do you think psychologists should explore further?

One potential focus is political advertising. Some people believe that negative ads are inevitable and consequential, but there’s actually an empirical dispute in the literature on this issue.

The one thing we know for sure, though, is that negative ads are discouraging and disappointing for Americans and undermine their confidence in government and the political process. You might wonder, then, shouldn’t candidates resist running negative ads? Yet there has been no restraint at all. I think psychologists have a lot to offer in thinking about issues like this. It’s one example of how more people doing political psychology could help democracy by conveying a more powerful message about whether this strategy ought to be stopped; whether it’s good for the public; and more generally, how and why and when advertising is consequential.

And, of course, this year’s election creates an opening for psychologists not only to help the country think about what is happening, but also for us to learn from what is happening.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s