Mitchell McKinney, professor of communication and director of the University of Missouri's Political Communication Institute, is one of the nation’s most respected authorities on presidential debates. During his long career, McKinney has often consulted with the Commission on Presidential Debates, weighing in on how events might be structured to better educate citizens about significant campaign issues — guidance that was influential in the creation of now-familiar “townhall” debate format. In addition, he has advised international, national, state and local debate-planning committees and he has co-authored and edited dozens of debate-related research publications. 

Televised debates are a relatively recent development in the history of American presidential politics, but today they seem absolutely essential to the process. How did this happen?

Actually, the rise of television and the advent of presidential debates go together. It’s no coincidence that the very first time presidential candidates engaged in general election debates was in 1960 (with the Kennedy and Nixon debates). By 1960, television and the original broadcast networks of ABC, CBS, and NBC had emerged as a significant force in American life, and it was the network news executives that first floated the idea of bringing the presidential candidates together for prime-time nationally televised debates. This innovation in political campaign communication was seen as the most effective way of reaching and educating the American public. In fact, it was estimated that approximately 80 percent of the U.S. adult population viewed, or listened to, at least one of the 1960 Kennedy-Nixon debates – which at that time was the largest audience ever assembled for a political broadcast. John Kennedy first accepted the invitation to debate with Richard Nixon initially declining. After Nixon was seen as hesitant to take on his younger opponent, the Nixon camp agreed to debates. The rest, as they say, is now history!

You’ve said that primary debates tend to be different — and more useful to voters — than general-election events. Why is this? 

Both in terms of informing and also in terms of helping debate viewers make their voting decisions, primary debates are far more influential than general election debates. By the time we get to late September or early October in a presidential race that has been going on for more than a year, very few citizens are still undecided or uninformed about the major-party candidates. This is not to say, however, that the general election debates that take place in the fall are unimportant. The fall debates serve more to reinforce one’s candidate choice; and also to help those who might be weakly committed — I like to call these folks the “persuadables” — to more firmly commit to a given candidate. Our research has consistently found that general election debates convert hardly any partisans (someone changing their vote choice from one candidate to another) and there are very few who claim to be “undecided” late in the fall who will commit to a presidential candidate based on debate performance. At most, we’ve found that a general election debate may influence as little as 2-3 percent of those who view the debate. Yet, a very important benefit that we’ve found from general-election-debate viewing is that those who watch one or more debates are more likely to show up in early November and vote. The reinforcement and motivation effects of debate viewing are an important outcome, even if very few people change their minds after watching.

Primary debates, however, are a much different story. First, large numbers of viewers come to these debates undecided, and they’re often undecided because they’re not that familiar with what might be a large field of candidates – and some of these candidates may not be well known political figures on the national stage. Therefore, we’ve found that after viewing a primary debate, and particularly one early in the primary season, as many as 60 percent or more of these viewers who were not committed to a particular candidate before watching the debate will claim allegiance to one of the candidates after watching even a single primary debate. Finally, there’s much greater issue learning reported from viewing primary debates. Again, when citizens are exposed to candidates that are just being introduced on the national stage, individuals learn a great deal about these candidate and their issue positions. 

In 1964, Johnson said no to Goldwater. In 1968, Nixon said no to Humphrey. Will we ever again see a major-party presidential candidate refuse to debate?

I think our presidential debates have now become institutionalized through public expectation, and no presidential candidate could likely withstand the negative public reaction from refusing to debate their opponent. The 1960 debates were thought to have such an influence on the outcome of the race — by helping Kennedy and hurting Nixon — that Lyndon Johnson feared he was not nearly as “telegenic” as the youthful Jack Kennedy and therefore LBJ refused to debate. Similarly, Nixon claimed it was his poor debate performance in 1960 that doomed him, and therefore he refused to debate both in 1968 and 1972. Once debates resumed in 1976 — which, by the way, was the first occurrence of a VP debate — we’ve had at least one debate in each presidential election cycle since 1976.

Is it really possible for candidates to win over audiences over by their command of the issues? Both Kennedy and Reagan, for example, were said to have gained huge advantage by simply looking and acting “presidential.”

On this point, image vs. issue, the evidence is pretty convincing: televised presidential debates work much more on the level of candidate image rather than issue learning. In other words, the primary purpose of these televised debates is for candidates to craft and project a presidential image – someone who is seen as likeable, trustworthy, smart, even the “common person” that can relate to or understand the concerns of the “ordinary” citizen. Debates are also useful in demonstrating how candidates engage in attack or clash. Here, we see if one is able to go on the attack while still appearing “presidential” or even likeable; and, how does one respond when attacked? It’s on this front that we assess if one is simply too mean or nasty, traits that we don’t like in our president, or if one when attacked or responding to attacks appears appropriately forceful and bold – traits that we do like to see in a “strong” president.

Of course, on a very basic level, issue-image go together. If a candidate seems to be avoiding answering a question, or is evasive about policy actions they have taken or would take in the future, then we might judge this candidate as less trustworthy. In fact, the presidential candidates who come to the debate stage full of facts and figures, concerned primarily with making issue points, tend to fail. We don’t like the “know it all” candidates who seem to care more about being right on the issues than projecting a likeable presidential image. In fact, in our content analyses of all presidential debates, looking specifically at candidates’ use of evidence, we’ve found that the candidate who spout the most facts and statistics in their debate responses are actually seen as the loser of the debate. So, candidates should not approach their debate performance as an opportunity to demonstrate their intellectual brilliance. We use these debates largely to help us “size up” the candidates – What kind of person are they? Do I like them? Does this individual seem to “get it,” or understand what I’m concerned about? These questions involve a mix of image and issue assessments.

What do you think about the “instant-reaction” polling that FOX, CNN and others now do? Are these polls useful for voters? Or do they promote a kind of American Idol experience that detracts from the gravitas of the encounter?

The “pop” polls that appear immediately after the debates (and some now are even being reported while the debate is still in progress) are rather meaningless. The methodology of these polls is highly suspect; often times viewers are encouraged to dial a 1-800 number and register their “winning” candidate, still others have people select a winner online, even allowing people to call or click multiple times. These instant polls are a way for supporters to show their enthusiasm for their candidate. The more legitimate daily tracking polls, however, do usually show slight movement in the candidates’ poll standings based on a superior or poor debate performance. Again, any movement in the polls that can be attributed to debate performance will register a couple of days following a given debate.

How big a role do gaffes typically play? Any chance we’ll have an “Eastern Europe is not dominated by the Soviets” moment this fall?

We don’t usually see in a presidential debate a single misstatement or “gaffe” having much of an influence on the eventual outcome of the election. Usually, an “oops” moment … will often serve to reinforce a larger narrative about the candidate – that he is not very bright, too cold and calculating, or not fully in command of important facts. Indeed, in today’s social media world of tweets and memes, almost any debate wisecrack or gaffe will become the instant "killing of big bird" or “binders full of women” remark that circulates instantaneously; for many citizens these moments become the full 90-minute debate distilled in a second or two of internet humor.