Click image to enlarge

Political scientists Bert Johnson and Matt Dickinson discuss the presidential election.

Media Contact

Ray, Sarah C.

Professor Pundits: What Did We Learn?

November 11, 2016

MIDDLEBURY, Vt. -- President-elect Donald J. Trump has begun his transition planning. The surprising (to many) results of the election have given Middlebury political scientists Matt Dickinson and Bert Johnson a lot to think about as they consider the many ways in which this election defied conventional thinking.

In their final commentary on the 2016 presidential election, the Professor Pundits discuss how and why polling misled so many people, why the presidential transition period is so crucial, and what the election results mean for the very idea of presidential campaigns.

Dickinson writes the blog Presidential Power and is frequently quoted in the national news media. He often live tweets political events at @MattDickinson44. Johnson, also a regular in the national media, tweets at @bnjohns.


What can we think of a candidate who wishes to abolish the EPA, thinks that global warming is a hoax, and appoints Bannon, who encourages the publication of the ALT right, misogynous and racist articles? We are headed for frightening times.

by Dr. R. Matilde ... (not verified)

Im not a pollitical science person or analyst but there were no models to cover overwhelming voters who were hidden because it just was not pollitically correct to vote for Trump.The pundents did warn about it but it never got a rope hold as viable lagitimate stats to the polling. Socail media who thought the tea party they dismissed as relevant to this election was dead from being kicked in the lower extremities and were kicked and beaten back long before the election and they came out in full force.

by thomas bervin (not verified)

You can think, you can reflect, on what it means that Mr. Trump was elected with those views, who voted for him, where the votes came from geographically and culturally. You might also offer sourced and vetted sufficient facts and briefly share your reasoned analysis to provide foundation for your conclusory assertion that Mr. Bannon encourages misogyny and racism.

by Ed Breen (not verified)

I've studied the "alt right" (rebranded white supremacy) for years as part of my religion and violence concentration. (I'm a historian of religions.) The election exposes the growing paradigm shift in geography and demography that splits the country into white v. diverse and rural v. urban dichotomies. What the set of groups on the radical right formerly lacked was numbers (unification and resolve to achieve a practical political goal) and a charismatic leader, uber alles. Now it has both. Fortunately, the opposition is very aware, numerous, and has resources. The price of liberty is eternal vigilance.

by Jean Rosenfeld (not verified)

It's possible that some people were reluctant to admit that they were Trump voters (naming another candidate instead), but I suspect that what is more likely is that either Trump voters, when contacted, were not responding to polls at all, or that pollsters were underestimating the chances that Trump supporters would vote. The first possibility makes sense if Trump voters are suspicious of 'the establishment'; the second makes sense because Trump voters in past elections may have had lower rates of participation than others. We'll have to examine more data as they are released before we know for sure. One thing to remember, however, is that the movement in voting patterns from past elections was very slight, overall. Larry Bartels has a good piece about this in the Washington Post ( There was slight movement in the Republican direction among rural voters and voters with lower levels of educational attainment, but the "typical Trump voter" is more or less the same as the "typical Romney voter."

by Bert Johnson (not verified)

Hi Dr. Matilde - I think many people share your concerns regarding the future of a Trump presidency, and for the reasons you cite. One thing to keep in mind, however, is that except in times of national emergency, the presidency, historically, has been a relatively weak office in terms of its effective influence on governmental outcomes. This is because to get anything done, such as abolishing the EPA, the president requires the active cooperation, or at least the aquiescence, of a multitude of other actors, beginning with Congress. This is even true for foreign policy. So my expectation is that this system of shared powers will act as a restraint on Trump's worst impulses.

by Matthew Dickinson (not verified)

Hi Thomas, In fact, there was an ongoing debate among political scientists regarding whether there was a hidden Trump vote, and if so, how large it might be. One way we tested that was to examine Trump's support in polls conducted via interactive voice versus those conducted by live interviews (that is, automated versus live interviewers). The idea is that Trump supporters might have been more willing to reveal their preferences to the automated voices. Alas, we didn't see a huge difference in the preferences expressed in either type of poll. But clearly the state level polls in key battleground states systematically understated Trump's support by 2%-5%. We'll be studying why this was the case for some time to come.

by Matthew Dickinson (not verified)

Hi Jean, I would make three responses to your comments. First, in addition to the rural/urban and white/racial minority divide, there was also a growing split based on education; relative to 2012, Trump attracted greater support among whites who do not have a colege degree. Having said that, it is also worth noting that in a state-by-state comparison, Trump tracks the Romney vote in 2012 quite closely - he won the election by boosting support among lower-income whites slightly in key states, but this election did not constitute a whole-sale restructuring of the respective parties' voting coalitions. Finally, while not disputing your characterization of some Trump supporters as members of the "alt right", my sense from talking extensively with dozens of Trump supporters at several rallies is that most of his supporters do not fall into this category. However, your final sentiment holds regardless of who supported which candidate, I would think!

by Matthew Dickinson (not verified)

Dr. Dickinson, I have a graduate level degree, am an independent voter, and did not vote for Mr. Trump. That said, I never participate in polls when contacted before elections. While I do participate in scientific surveying conducted by my professional societies and my alma mater (University of Iowa) for the sake of advancing knowledge, I do not see any contribution to the greater good of mankind by sharing my knowledge with partisan pollsters hoping to use the information to sway an election to their benefactor/client's advantage. Please educate me; why should the average citizen participate in polls run by any of the parties, unless you are a member of said party? Respectfully, S. L. Benton, B.S., M.D.

by Steven L.s Benton (not verified)

Hi Mr. Benton I suppose one answer to your question is that not all political polls are commissioned by parties or their candidates - some are run by academic or media organizations, similar to your professional societies, and are used to gain a greater knowledge of the electorate. We often use the data from these polls in our classrooms. A second answer is that there may be a benefit to responding to polls, even if they are conducted by parties or candidates. The benefit is that the survey responses may give those candidates a better understanding of voters' concerns, and may allow the candidate and parties to better respond to those concerns. To be sure, that response may simply be position taking for the sake of winning votes, which I think is your concern. But it could also reflect a genuine effort by the candidate to find out what the voters want - and to address those concerns. After all, if we want our elected officials to serve our interests, they need to know what those interests are. Polls, one might argue, are a means to discovering those interests, even if they also are being used for partisan purposes.

by Matthew Dickinson (not verified)

Dr. Dickinson, Thanks for your thoughtful answer. The pollsters who call me rarely identify who they are working for. If I received a call from someone who stated that they were gathering information for a political science department at an institution of higher learning, I would be inclined to participate, time permitting. That said, my own professional societies and alma mater approach me via paper or email, which is more respectful of my time. Most of my social group do not respond to phone solicitations. Furthermore, those of us who are independents, which I believe now reflects the plurality of registered voters, are not members of a political party in part because of our skepticism about the motives of said parties. Most of my extended family share this point of view including former union members who left the Democrat Party years ago and farmers who no longer call themselves Republicans. Thank you for hosting this discourse, S. L. Benton

by S. L. Benton (not verified)

Post a new comment

We hope to create a lively discussion and invite you to add your voice. Please keep comments civil and relevant to the news item at hand. We may remove comments that do not follow these guidelines.

Your comment will be visible after it has been approved by our comment moderators.