Colloquy with Maxwell McCombs at the University of Texas at Austin: agenda setting, a limitless theory in a connected world

Abstract Professor Maxwell McCombs is a widely known authority in Journalism. Much of his historical and intellectual background, as well as his contribution to journalism and political communication, is found in this interview, done in 22–23 May 2017 in Austin (Texas) by two qualified interviewers: a corporate communications professional, and a professor of communications theory. The long conversations with the interviewee allow them to extract the best of his academic background and, one may also say, of his amiable character. Professor McCombs is currently the Jesse H. Jones Centennial Chair in Communication Emeritus at the University of Texas at Austin. He has been visiting professor in many different US and worldwide universities. After the original Chapel Hill Study (North Carolina) with his colleague Donald Shaw he coined the term ‘agenda setting’ in 1968. He received his M.A. and his Ph.D. from Stanford University and his B.A. from Tulane University.


Introduction
When I was asked by the editor of Church, Communication and Culture about the possibility of interviewing Professor Maxwell McCombs, many things came to my mind: I had read one of his books several years before, and studied Agenda Setting theory. When I was working on my final research for my MSc. in Social Communication at the Universidad de Costa Rica (UCR) I had to compare the differences, if any, between second level Agenda Setting and Framing. I wish I could have known way back then what I know now after this interview. In any case, I started to prepare the interview and planned my trip to the US.
I found a wonderful collaborator in Ivan Lacasa-Mas, Associate professor of the Universitat Internacional de Catalunya (UIC), who was doing a visiting period of studies at Moody College at the University of Texas at Austin, precisely where Professor McCombs worked and where the interview was going to take place. In 2017, Moody College was ranked #5 Communication School by the CWUR World University Rankings, #2 Best Journalism Program in the US by the USA Today; and #3 Best Journal School for Undergrads (Testive).
Our interview took place in Austinin the heart of Longhorn territoryon May 2017 in a couple of sessions. Professor McCombs was always smiling, with a great sense of humor, and he made us feel that we had known him for years. After one of the questions he prefaced his reply by saying: 'Not everything has to be very serious stuff, because you need to have a little fun in life'. He really does.
Beginning and evolution of agenda setting theory How did the Agenda Setting theory get started? What were the insights and intuitions you had way back in the early 70's that made you undertake the research with which you finally developed the theory?
As an undergraduate I had read parts of Walter Lippmann's book Public Opinion (1922). When I went to Stanford for a master's, one of my personal projects was: 'I'm going to read the entire book'. So, I bought a paperback copy. This was so long ago … I think I paid three dollars for the book … I read it and I was very struck by his arguments about the importance of the press in the shaping of public opinion and particularly his concept of the 'pseudo-environment' that says that people are not responding to the world that is there, but to the world that is presented to them by the news media. The title of the chapter that opens the book is 'The World Outside and the Pictures in Our Heads'. Of course in his argument, the media is the bridge. In many ways, that sums up the idea of Agenda Setting. Again, I was very struck by that.
Then we have to flash forward for about four to five years. I was on the faculty at UCLA in the journalism program and we got into a discussion one day about how the impact of a news story can change. If it was the only big story of the day, it was the lead in the newspaper. But if there were several competing stories and it got push down below the front fold, or even inside the newspaper, does that lessen its impact? The summer of 1967 I moved to the University of North Carolina at Chapel Hill where I met Don Shaw. We decided to do some research around those ideas. With time, we came up with the Chapel Hill study.
At that time, not long after Joseph Klapper's book The Effects of Mass Communication (1960) selective perception was riding particularly high in the field. It basically said there really were not many effects. So, we thought, if there are not many effects, where are we most likely to find them in the election of 1968? We thought that the optimum place to look for these would be among undecided voters, because the main argument about selective perception was that people put up psychological barriers. If they favored the conservative candidate, they would tend to emphasize the conservative news about that candidate; if they favored the liberal candidate, they would favor the news of the liberal candidate. So that was the focus of our survey of voters in Chapel Hill. We only interviewed undecided voters.
We used a Gallup poll question that at that point had been around for 30-35 years. The question is still used today and many other pollsters use it also: What do you think is the most important problem facing this country today? It's a totally openended question. We decided we would ask that question of undecided voters and compare their responsiveness to the news coverage in the preceding weeks to see to what extent is there a match between the pattern of emphasis of the news coverage and the pattern of answers to the Gallup poll's most important problem question.
Well, as you know, among the undecided voters in Chapel Hill there was an almost perfect match. So that of course encouraged us to go forward. We were very excited about the election. We had all our data by November of 1968, and during the holidays we wrote a paper and submitted it to the AEJMC (Association for Education in Journalism and Mass Communication). What happened? About two months later we received an answer saying that our paper had been rejected! [laughs] But we still thought this to be very interesting. So, we organized a larger scale study in 1972 in Charlotte, North Carolina, which is a mid-size American city. This time we drew our sample from the entire voter rolls: Republicans, Democrats, and Independents. We were not looking at just the undecided voter, we were looking at everyone. And we confirmed the hypothesis that we had originally tested at Chapel Hill. That was the beginning.
The different levels of the agenda setting theory It's been a long time since the original Agenda Setting theory was developed by you at Chapel Hill (North Carolina). First you just talked about plain Agenda Setting, but afterwards you came upon Attribute Agenda Setting which you ended up calling 'second level agenda setting'. In recent studies, you speak about Network Agenda Setting (or Third Level Agenda Setting). Could you please explain why is there a third level and what is its main analytical contribution? Is Agenda Setting a limitless theory? That is, sooner or later, perhaps will there be a fourth or even a fifth level? What kind of Agenda Setting studies do you see in the near future?
In the book we did on the Charlotte study we introduced theoretically the idea of Attribute Agenda Setting. That is, if you think in abstract terms, considering issues as a set of objects of attentionand issues are still the dominant object of attention in Agenda Setting research, although there are many others, the salience of these objects transfers from the media agenda to the public agenda. However, objects also have attributes, they have characteristics. You could think of attributes as different aspects of an issue. In a recent study, we did on the Iraq War (Muddiman, Stroud, and McCombs 2014) we identified about 30 attributes ranging from military operations to the history of Iraq, and lots of things in between. It is even easier to understand attributes if a public figure is the object you are studying or the set of public figures such as candidates for political office. Obviously these people have certain attributes, certain characteristics, from their physical appearance to their ideology, their background, their personality, etc. We introduced the idea of attributes in the book on the Charlotte study; then four years after that, in 1976, we studied attribute agenda setting in The Three Site Study. 1 We picked three very different kinds of locations: (1) Lebanon, New Hampshire, a community where the first presidential primaries were held; (2) Indianapolis, Indiana, again a mid-size city; and (3) Evanston, Illinois, a suburb of Chicago. In this instance, we followed the voters for the entire election. We did not have big groups of people. We had small panels of voters at each of these three cities and we interviewed them 10 times across the election year; also, one of the sub-studies we did during that election was in Evanston. That was the first attribute agenda setting study. We looked at how the Chicago Tribune, the dominant newspaper in that market, described the two candidates, Jimmy Carter and Gerald Ford. To measure the public's attribute agenda for each of the candidates, we came up with the question that has endured over the years and seems to work not just in the United States but just about everywhere for any type of object or theme under study. Today, it might be expressed as this: 'Suppose you had a friend who had been away for a long time and knew nothing about Donald Trump. What would you tell him?' This is totally open-ended. Some people of course go to the ideology. Others describe the personality. People can go in very different directions. Again, we got very strong correlations between the attributes of agendas, to tie the first and second levels together. For every object on the first level there is an agenda of attributes. And just as you can rank the objects in order and compare them at the first level, you can also rank in order the attributes at the second level, starting from what is the most frequently mentioned characteristic of a political candidate. Then, you can do the same for the second, the third, and so on, and you can compare those and calculate the correlation. The analysis procedures are identical for the first and second level. You are just looking at different things: either a set of objects or a set of attributes of a particular object. And that is where things stood for a long time until we came up with the idea of Network Agenda Setting.
Can you tell us more about network agenda setting?
To introduce the concept to a broad audience, Lei Guo and I edited a book on Network Agenda Setting (2015). Let me give a quick background. Lei was a master's student at the time, and I was preparing the second edition of the book Setting the Agenda. 2 She was doing abstracts of new articles to help me with updating the book when she came into my office one day with some of her work. I said 'Lei, would you like to learn network analysis?' She asked, 'Why?' and I replied, 'Well I don't know, but I think it might be a useful way to analyze agendas.' So, she went off and came back three weeks later and she knew how to do network analysis. She asked me if we should design a new study. And I said 'No. I think we should re-analyze an existing study, because if we create a new study and we get a very low correlation between the networks, there could be two possible explanations: (a) Network Agenda Setting is not very powerful, or (b) we picked a bad place to look for it. ' We re-analyzed a data set that was based on the attributes of four public figures in Texas: two candidates for governor and two candidates to the US Senate. And we found very similar strong correlations. So, we get to the core of your question. In the past, we would pull out the elements of emphasis, for example at the first level, that is issues, and we would go through the news and ask, 'How many times are these different issues mentioned?' We were pulling them out and looking at them in isolation. We did the same for the public in how they respond to the most important question, and again the same for attributes. In the study of Texas politicians, we re-analyzed what we had pulled out of the Austin American-Statesman, that is, how many times various attributes were mentioned in the news for each of those four candidates, and we did the same for the public's response.
If you think about it, these are the isolated attributes of a particular candidate and we know these are the prominent ones but, how are they connected to each other? If you think of it in terms of how Lippmann explains the pictures in our heads, the first level asks what are the pictures about, the second level asks what are the dominant characteristics in these pictures and the third level is getting closer to the idea of what the picture is.
To put it in more theoretical terms, third level agenda setting is more like a 'gestalt' of how data come together. How are they linked to each other? In sum, that re-analysis of the Texas election studies was the beginning of Network Agenda Setting.
The open future of the theory You have added a second and a third level to Agenda Setting. But did you modify any of the original ideas that you conceived in Chapel Hill? In other words, over the years, has Agenda Setting Theory expanded without actually being modified? Is it a limitless theory where one can discover a fourth or fifth level?
If you had asked me 10 years ago, I would have said there was always the possibility that someone would come up with a new route, but I couldn't have identified it. However, it has proved to be a highly productive theory and people are always coming up with new ideas. Not long ago I read a manuscript not yet published. It does not speak of a new level, but it illustrates particularly how young scholars have exciting new ideas.
This unpublished manuscript went beyond just counting the frequency of various issues. Among other categories, for instance, it made a distinction between news stories about unexpected events and how frequently those occurred and what issue they reflected, versus news stories about more or less continuing patterns of news coverage. And it seems to me that it was maybe even proposing a third or fourth category and then analyzing the agenda setting effects of those separately. We can say that there are even more refined ways of looking at the agenda beyond just counting.
At the third level one of the key measures of a network is degree centrality. If you analyze a network, some of its attributes are not only tied to more attributes than others, but also some of these ties occur more frequently. That is basically what degree centrality measures. You can think of it as an alternative measure of salience.
This reminds one of power and influence diagrams, where there are central people in those diagrams and they are tied to more people with stronger or weaker ties.
Exactly, network analysis itself goes way back. I encountered it briefly in graduate school. At that time there was a journal called Sociometry, 3 which did exactly that, sociological analysis: typically, who's connected to whom and who is central. Another example getting closer to our field, a dissertation done here [University of Texas at Austin] a few years ago, looked at all the tweets for a year within the U.S. House of Representatives: who is tweeting whom. What was interesting about it was, of course, that people like the Speaker of the House and the majority leaders were in the center of the network, or at least the center of the Republican network. But there also were some surprises: members of the House who did not hold official positions, but were very central in the network.
In the inaugural issue of The Agenda Setting Journal there was an article by a research team from the University of Alabama 4 who looked at the trends in Agenda Setting research from 1972 to 2015, and they found 451 articles published articles in English on Agenda Setting and, of course, a great many more in Spanish, German, Chinese and other languages: and the first level of Agenda Setting is still an active area. We have moved on to second level and third level but people are still coming up with ideas about how to do first level analysis. So, will the next kind of big contribution be the fourth level? Maybe. Or will it be somewhere else in the theory? I think it is impossible to predict. But given the history of the theory, there are still many questions that can be explored. The Agenda Setting Journal is the first theory-based journal in our field, and Benjamin Publishing (Amsterdam) will sponsor it.
Has there been any modification of the core idea? There have been expansions of the theory, but the core idea of Agenda Setting is not first level, second level or third level; the core idea is the transfer of salience from one agenda to another. This can occur at the first level with objects, at the second level with attributes, at the third level with networks. For all of them, the core idea remains exactly the same. It's the transfer of salience from one agenda to another. And because that is the core idea, while most of the research still remains in the area of public affairs, there are new areas of research some of which are quite remote from that. Phil Bantimaroudis, a graduate of this department, now a professor in Greece, has opened up an area that he calls 'Cultural Agenda Setting' (Bantimaroudis 2017). His first study was still media oriented. His question was: Does the amount of attention Greek museums receive in newspapers influence attendance at Greek museums? And the answer is yes, it does. He has done similar research with movies and is going to publish a book. All of this is applied to what he calls the cultural industries, very different from the original domain.
Probably the one that is pushing the farthest out domain is a study in which two Israeli scholars applied Agenda Setting to religious practice in Jewish synagogues in Europe. 5 They explicitly used the framework of Agenda Setting in this analysis. So, if you take the core idea you can begin to expand into a lot of new areas. But public affairs still remains the core domain. If one just thinks of Agenda Setting in terms of the core, then all these specific domains are just operational definitions of the key concept. I suppose one can even think of first, second and third level as operational definitions, and I think that core is still going to carry us forward.

Creators of agenda setting: McCombs, Shaw … and Weaver
Today the Agenda Setting Theory seems to be mainly associated with your name rather than with Donald L. Shaw's. 6 Do you agree with this statement? Why? Don (Shaw) over the years has remained very much involved in Agenda Setting. He's also devoted a fair amount of time to historical research because that is what he was trained for at the University of Wisconsin. Fortunately for our partnership, he was trained in quantitative approaches to history, so we were speaking in the same language.
That has made it very easy to collaborate, and we have worked together on many studies over the years. He has probably spent more time on history than I have on other areas. He is very active and has come up with an interesting idea that he calls Agenda Melding. The idea of trying to link agenda setting by the media as we know it with people's personal agendas, and how do those mesh.
He also worked, a few years ago, with a young scholar, Chris Vargo, who is now at the University of Colorado. It has been five years ago now that Don, Chris, Lei and I did a joint study in which we looked at Twitter during the 2012 presidential election, particularly tweets about the issues in the election. 7 This was also our first go with big data. Chris and Lei led the way on that, and Don and I did more on the theory side. They downloaded several million tweets over a three-month period about the election. They started with a very large group of tweets and then divided those into Republicans and Democrats, according to which candidate they were mentioning and linking to an issue. To make it manageable, if a person did not tweet more than once a month, we dropped them. So, we chose people who were actively commenting on the election: Obama supporters and Romney supporters. And then we had a third set of tweets from journalists in major news organizations. With this we had an issue agenda from journalists, pro-Republicans and pro-Democrats. We did a network analysis of these data and found many interesting comparisons, as well as the fact that the media still retains a lot of influence on both of these voter groups.
I would like also to mention David Weaver. He has a lot to do with Agenda Setting. We've also known each other for a long time [He shows us a picture]. There's David Weaver on the left, Donald Shaw in the middle, and you know who's on the right [smile]. Dave [Weaver] came to Chapel Hill for his Ph.D. shortly after Don and I did the Chapel Hill study and he introduced the concept of 'need for orientation' in the Charlotte study, the very next study in 1972. And so, Don and I regard Dave as a co-founder of Agenda Setting. He wasn't involved in the very first study, but from the second study on he's been there.

Agenda setting and framing
There's a recent paper, 'The end of Framing as we know it … and the future of media effects', coauthored by Cacciatore, Scheufele, and Iyengar (2016) where the authors mention that framing has emerged as one of the most popular areas of research for scholars in communication but that at the same time it is a concept with so many definitions that it's unclear what it really is. Could you talk to us about the relationship between Agenda Setting and different forms of Framing? Do you agree with, for example, the analysis that David H. Weaver did in 2007 in 'Thoughts on Agenda Setting, Framing, and Priming'?
I was delighted to see this article that you cite because as they point out, if you really look at the entire literature, everything is a frame and that's a problem. There are many different definitions of frames. For many years I have said that some of these frames are essentially the same thing as attributes, just two names for the same thing. You can choose to approach your research from a Framing perspective or Agenda Setting perspective but you are essentially looking at the same thing. But some frames are very different. For a number of years, I called them Rhetorical Frames, and probably the best example I could come up with is that old clich e question 'Is the glass half full or half empty?' If you think of them as attributes, those two frames [half full or half empty] are identical; they are speaking of the quantity of the content, because the glass is at the half way mark. But if you think of it from the point of view of rhetoric, they are very different, one is optimistic and the other pessimist. So that is a very different kind of frame.
I was also delighted to see in that article that they came up with the notion that there are two kinds of frames and that, for the field of Framing to make progress, we need to distinguish between these: they call them Frames of emphasis and Frames of equivalence. Frames of emphasis are also in effect attributes: how much emphasis is being placed on certain aspects of a situation or in the description of an object. Frames of equivalence are more like Rhetorical Frames. To quote from that article: 'First there is considerable disagreement over what exactly constitutes Framing. This is perhaps most readily apparent in the different operationalization of the concept, particularly apparent is the difference between Equivalence Framing, a form of Framing that involves manipulating the presentation of logically equivalent information, and Emphasis Framing, a form of Framing that involves manipulating the content.' Thus, Emphasis Framing refers to attributes that are manipulating content, according to which aspects of the situation or object you are talking about. Equivalence Frames are logically equivalent but present different perspectives. A glass half full and a glass half empty are the same, but they communicate very different perspectives.
I hope that helps clarify the Framing field because I have felt for some years that because there were so many of these competing definitions of Framing, it might suffer the same fate as Uses and Gratifications. This is the advantage-disadvantage of being around in the field for a long time because you watch it evolve. Back in the 70 s was when Uses and Gratifications (from now on, Uses and Grats) really took off. At that time everyone said Uses and Grats was the new paradigm of effects, that it was the new direction. What happened was that Uses and Grats did not really become a new paradigm that displaced media effects, and after a while, people got into an endless debate about how many Uses and Grats are there. One person says there are five; another says no, there are eight, and someone else says no, there are really only three. So now when you encounter Uses and Grats, you find different versions. It is still a highly useful perspective, but it is fragmented and has lost a lot of its momentum. For a long time, I thought, this might happen to Framing, that there would be all these little fiefdoms where there is my definition of Framing, your definition of Framing, somebody else's definition of Framing; this sort of thing makes it much more difficult to get a coherent body of knowledge, to advance in terms of a scientific theory.
Agenda Setting does have a coherent vocabulary that ties studies together. You can say this is first level, second level, or third level. You can say, these are the sources of the media agenda and these are the consequences of the media agenda and this is the psychology underlying Agenda Setting effects. Agenda Setting has tighter definitions of its components than Framing does.
That's kind of a quick answer. But I think the concepts of Equivalence Framing and Emphasis Framing are very important because they make a nice boundary between two sets of literature. It divides the literature in a way that makes it much more manageable. How does that fit with Agenda Setting? I think Emphasis Framing and second level Agenda Setting are essentially the same concept.
Agenda setting, corporate reputation and fake news Is there a relationship between Agenda Setting and Corporate Reputation? Are powerful international corporations ruling the media and therefore setting their reputation and, more generally, setting national and international agendas?
That's one of the areas where Agenda Setting has expanded into a different domain. Basically, I discovered that research 10-15 years ago now. Business schools are active in Agenda Setting research which basically looks at how frequently corporations or other institutions, but mostly corporations, were mentioned in the media and how well known they were among the public. Also, their investigation has taken them to the attribute level: how they were mentioned in the media, and how that impacted their reputation. There is a big well-funded institute in The Netherlands called the Reputation Institute that does an annual survey looking at corporations. 8 Their analysis is very straightforward Agenda Setting at both the first and second levels. After the traditional public affairs area, reputation is one of the new domains of Agenda Setting, and probably the most active area because business schools have lots of money.

Do corporations have impact on national and international agendas?
To the same extent that they have an influence on Congressthey lobby for certain ideas. But it is a very limited impact. I think there is more if you just look at reputation in the media. I am rather wary of pronouncements that businesses make about particular issues because obviously they have a vested interest. For example, when a big coal mining companies speak out about climate change. Why did they do that? Because efforts to control climate change are impacting their business. I remember, many years ago, the discussion about tobacco causing lung cancer and other diseases. And it was one of the national television network reporters recalling a new study showing smoking cause cancer. It had been a large-scale study. And of course the Tobacco Institute, the trade group, had a response saying 'Well, it's not quite proven yet'. The journalist was on a phone conversation with another colleague saying: 'I guess we have to call the Tobacco Institute and get a statement'. They knew what they were going to say: 'We need more study'. However, for years the research had become more and more clear that if you smoke you are highly likely to come down with lung cancer and other diseases. So, the journalists decided not to make that call. The lesson from this story is that, once an issue becomes well defined, when it is clear-cut that it is tilting one way or the other, the influence of corporations becomes truncated because they have a self-interest.
There is also an interesting study in the Network Agenda Setting. Greenpeace took on the Nestle Corporation over palm oil production. Nestle's influence was pretty minor and the Greenpeace influence was, and continues to be, very strong, because the evidence is rather clear that the impact of palm oil production on the habitat of certain animals in the environment is very negative. I think one needs to look very carefully at the corporate influence, because for the most part it is very constrained.
How important are the audiences in the Agenda Setting? Do they really matter? Has there been a tipping point in the way mainstream media consider the audiences?
It is very important, as I mentioned in the Charlotte study. David Weaver introduced the concept of 'need for orientation', so that ultimately the audience controls the process. If they do not find an item that's in the media relevant, it is not likely to have much of an Agenda Setting effect. Also, if it's what Zucker calls an obtrusive issue, 9 where people have personal experience, the media coverage does not have much effect either. Media networks -I hopeare attuned to that. They certainly need to be. An example of an obtrusive issue is inflation. One doesn't need to watch television or read the newspaper to know if there is inflation in the economy. Whatever the media's agenda is, there is a pretty minor impact there. So, yes, ultimately the audience controls the process. People are not just a blank slate waiting to be filled in by the media. People turn to the media when they want information about something that they cannot obtain through personal experience. For example, the week after a very strange 2016 presidential election and with a lot of real fake newsnot what Trump calls fake news, but really fake news, The New York Times gained 100,000 new online subscribers. And the Wall Street Journal, also gained. This means that people are looking for really solid information about what's going on out there. In this case, we have two major trusted journalistic institutions and they've got huge increases in their online subscriptions. People will not follow them as the only source, but they turn to the The New York Times and the Wall Street Journal in addition to other sources, to know what's happening out there. Certainly, the media does pay attention to the online traffic, but overall news values prevail.

On methodological questions
How does a 'classic' author of the discipline see the present of the communication research field? What added value do you think younger scholars are 'bringing to the table'? Do you miss something? Do you see any shadows or even significant gaps? With this question we are thinking of more general issues such as, for example, young scholars obsessed with publishing or having a different mindset that is reflected in the focus of their studies or on how they do theories or science. Do you think some peace (and calm) is needed to elaborate a big theory? Is there a real problem here?
I think the field is gradually coming around to the view that, when you go up for tenure, what's important is not just how many articles you have published but whether there is some coherence with the rest of your work. In the past, it was more about years and number of articles published, without much attention to the coherence.
My advice to students in graduate school is that it is a great time to explore a lot of different areas. You should take advantage of that, and you might find some interesting theories, three, four, even five. But you cannot keep going and keep up with five areas. Often someone takes a couple of seminars in a field and then doesn't like to let it go. My suggestion is to prioritize your interest: write some conference papers, if possible publish them and then close out that area, so you don't feel that your time there was wasted but that you got something. at some point you need to come down to, probably ideally, two theories or areas. If you choose one too prematurely you may find yourself in a dead end. But with two areas you are hedging your bets, so that if one doesn't work out too well, you've got another one. In the experimental sciences, that is exactly what people do: they find a particular niche within a field, whether it is chemistry or biology or another. Theoretically those fields are more developed and move forward faster than things do in the social sciences, but over time you need to focus much more intensely in one area.
If you do this, by the time you are ready go to full Professor, you should have enough research done to do a really good book that will look very impressive, because you have significantly developed some area.
In sum, I hope that more and more people will try to shape their research programs staying with one or two related areas. The problem is that there is still people who are wandering around because there are many interesting areas. However, too many isolated studies like that do not really do much for the field. That has happened also with Agenda Setting. It has attracted a fair number of people, but some of whom just passed through and did a study or two. But there is also a handful of researchers who kept at it, and that is how it has been developed as a coherent and expanding theory.

In your opinion, what is the methodological contribution of agenda setting theory?
Agenda Setting has become one of the major theories of communication research. It is approaching 50 years, and has generated a lot of empirical research.
I go back and look at the article in the new The Agenda Setting Journal, 10 in which Kim, Kim and Zhou identified hundreds of published articles on Agenda Setting. That is a substantial body of literature and it extends across the world. What is impressive is that the theory started in North America, then it moved or expanded to Western Europe, and now you find Agenda Setting studies throughout Latin America, in East Asia, increasingly in Eastern Europe and even in the Middle East. I have a visiting scholar here from Egypt and she has mentioned that there is a huge Agenda Setting literature in Arabic, which is of course inaccessible to most of us.
So, it has become an international theory. Just a final thought on that: there was a seminar some years ago in Taiwan, at which we discussed where the theory applies or does not apply. And the conclusion we came to is it will apply in about any country that, first of all, has a reasonably open political system, which, as a loose definition, would be if elections matter in that country. Secondly, it applies where the press is an open system or at least part of the press is an open system, meaning it is not under the thumb of the government or a dominant political party. All of that testifies to the evolution of Agenda Setting as a major theory. Methodologically, the original design used content analysis and survey research. Although Don Shaw and I cannot say we were the first to pair content analysis and survey research, we certainly made it much more widespread. Of course, the methodology of Agenda Setting studies has spread well beyond that original Chapel Hill design, so you find regression analysis and quadratic equations and, now on the third level, network analysis. None of those were original at all with Agenda Setting; they were widely used methodologies. And I should not leave out qualitative methodology, although there are not a lot of qualitative studies. So, the original design was a moderate contribution. In those early years I remember that, at the University of Minnesota, in the Methods Course they always did an Agenda Setting project, because the professors who taught it said it was a chance to get the students involved in both content analysis and survey research on one project. So pedagogically it was a useful area to study.
So, it brought about a moderate contribution methodologically, but let's not forget that most of the methodology is the general methodology of the Social Sciences, applied to professional and dominant areas such as Public Relations, which includes political campaigns where professional communicators have the goal to try to get items on the agenda. Sometimes they are successful, sometimes they are not.
In fact, the public relations area from early on has always liked Agenda Setting, also because it was a way to evaluate their work. One of the early studies looked at the Public Relations effort of the state government in Louisiana, and found the items they tried to get out to the newspapers, the major newspapers of the state, got into the papers about half the time. If public relations were a baseball game that would be phenomenal. They would like to have a .999 batting average but a .500 batting average is great. It gives them a model for evaluating their work and that is probably the major professional application.
Where do you place your work in that quantitative-qualitative spectrum?
The way Don Shaw and I approached Agenda Setting was obviously quantitative, because that's the way we were trained and, overall in the agenda setting field, there is more quantitative than qualitative research. There are a number of reasons for that.
One, it is a scientific field where replication is particularly important. If I do a study and say 'I know X and Y are clearly related, but that is one study done in one place at one time', I could be way off base because every finding needs to be replicated many times. My best friend from high school ended up being a professor of pharmaceutical chemistry. That is a very precise scientific field but he told me that it is filled with failures to replicate. And that happens in a lot more precise kind of measurement than we have at our disposal in Social Sciences. So, replication is important and quantitative results are much easier to compare.
That said, there is nothing inherently quantitative about Agenda Setting theory and most of the papers in my Spring seminar a year ago were in fact qualitative analyses. One was a comparison of the photos designated by the AP as the top photos of the day and how frequently newspapers used those rather than those not designated top photos, and an analysis of the characteristics of the photos that were used. If he had undertaken a huge sample, maybe you could have imposed a quantitative analysis on that. But this was one student with a limited amount of time. He did a nice qualitative piece. Another student also involved in photojournalism observed the Austin American-Statesman newsrooms editorial conferences, interviewed the photo editor and analyzed the way the newspaper handled photographs, especially how they made their decisions on what photos go on the front page of the paper. So, there is ample room for both kinds of studies and sometimes you cannot do quantitative measurement.
Some years ago, I met a young Chinese scholar in Australia who had a fascinating story about an Agenda Setting study in a city in China where the government wanted to do a big expansion of a chemical plant. The people who lived there did not want it. It was bad enough as it was and they did not want it to become bigger. A few academics and some journalists began to write about this and the discussion gradually evolved into a bit of a local movement over a period of time. About a year later, the government withdrew its plan. So, you could interpret that as Agenda Setting by a combination of academics and media who got the public concerned, and concerned enough, and got the government to back away from the expansion of the chemical plant. China is an interesting case because journalists are asked to walk a very fine line. That was very much a local issue. You can challenge the government locally; you certainly do not challenge the government nationally. But there are now more quantitative studies emerging from China many coming out of Hong Kong. Although it is now under the mainland Chinese government, it has inherited a very different culture. It still has some strong vestiges of British culture. I don't think I have ever seen an Agenda Setting study from China that was national in scope, whereas in the U.S. there are dozens of them, particularly during election times, where there are dozens and dozens of national polls. If someone wants to do an Agenda Setting study, one can take the survey data and can do a study of content analysis. I think in time that will happen in China.

Applying agenda setting to internet and social media
In the book Agenda Setting in a 2.0 World, edited by Thomas J. Johnson (Routledge 2014), the first chapter mentions a study that compares the presence of the Tea Party in the main stream media and on Facebook. Is there a different approach when you study Agenda Setting on Facebook? Is it applicable to Twitter? How does the Agenda Setting theory work in a world connected by internet and social media? Is it possible to talk about the same Agenda Setting theory in the internet world?
My view of the new media is that they are part of the overall media. If the new media are reasonably within the mainstream, they are applying essentially the same values. The metaphor I use is that they are all looking out the same window, and they more or less see the same thing. But because news values are something inexact, we can't have the same views. Moreover, journalists also tend to look around to see what other journalists are writing about.
When I worked on the Times-Picayune, if we wanted to know what the States Item, 11 the evening newspaper, had, we actually had to go out on the street and buy one. 'What did they cover? Is there anything they are covering that we should be?' It has become easier now, as Pablo Boczkwoski has noted. 12 He wrote an interesting book 13 on the two leading newspapers in Buenos Aires (Argentina) and noted over time that the lead stories they ran on the front page became more and more alike. In the contemporary newsroom there are TVs mounted on the wall, our website, their website, so you know almost minute to minute what the other paper is doing. If it is important, we need to get on to that too.
That is the area now we refer to as the Intermedia Agenda Setting. Media monitor each other. Moreover, in most countries there is a kind of a pecking order, a degree of status. I would say in the U.S. The New York Times and The Washington Post are the papers that everybody else looks up to and tends to follow, and so to a considerable degree they set the agenda. But that idea goes way back: many years ago Warren Breed, who was a professor of Sociology at Tulane University, described the Dendritic Theory of news coverage. 14 His metaphor was the sap system of a tree that flows from the roots through the various part of the tree. In the press system the flow is downward from a few national newspapers, maybe also from the Associated Press, to the big regional papers and then to the smaller papers. Kind of cascading down. In the original Agenda Setting we studied nine different media: The New York Times, The Washington Post, at least two of the three national television networks, one news magazine, and several local North Carolina newspapers. In the published study, we included a matrix in which we correlated the agenda of each news medium with the agenda of all the others. The correlations are all positive and high. They are the highest when you are comparing newspapers with newspapers, and the difference between national and local is not that much. Correlations are little less when comparing TV and newspapers because they are operating in different circumstances, but there is a consensus.
There's a lot of talk about polarization in the U.S., but in the Twitter study that Don (Shaw) and I did with Chris Vargo and Lei Guo across multiple Twitter sources, there was substantial agreement on what the major issues are. 15 Which side of the issue different groups came down on was different, but everyone was on the same page.
That goes back to Harold Lasswell's three functions of communication: surveillance, creation of consensus in society, and transmission of culture, which is less related to journalism. Consensus in society is important. If people are talking past each other on totally different issues, you are going to have chaos and society cannot function in chaos.
Have social media really democratized public opinion? Do you think the best educated and mostly wealthy young adults dominate online audiences? Could we refer to those people in Twitter who have many followers as real micro-agenda setters?
Has social media democratized public opinion? No! [laughs]. There is still very much a digital divide. It is larger in some countries than others, but it certainly exists in this country [USA] and even for people who have access to all these new media, they do not necessarily use them, particularly older people.
Probably most people now have access to them. Increasingly libraries have computers where you can do e-mail or access various social media, but some people just don't want to use them. Others do not use them simply because they are not interested. I would describe the users, say during an election campaign, of Twitter, Facebook, Reddit, etc., as people who have a high degree of political interest. Fifty years ago, if you had a high degree of political interest it probably meant you talked to a lot of other people, but you would talk to them face to face. If you really had a lot of interest maybe you would write a letter to the editor of the newspaper.
Nonetheless, describing only the people who use Twitter as persons with high political interest, leaves a lot of people out, either because they just are not interested in using technology or they do not have the means, or they do not have access to these technologies. Even some of us don't have access because we do not want to. Indeed, I don't understand how some people seem to have all this time to tweet. In A Really Big Lunch, a book more related to my son's culinary interest than to mine, written by a food writer Jim Harrison, he said if all these people who spend hours and hours on their cellphones would spent that time drinking wine they would probably be much better off [smile].
Let's take another example from the World Association of Public Opinion Research, in which I have been involved in for years. They set up a website where you could comment on whatever you want to. I noticed there were a dozen or so persons that seemed to have something to say at length on things every day. Do these people have a job? … I mean, maybe if you were independently wealthy … But I ask, do you have an hour every day to be composing your thoughts on all these issues? I certainly don't.
One can look at these people and also a few scholars as opinion leaders in the new version of Two-Step Flow. 16 Perhaps to some extent they are absorbing lots of media. They are commenting on it. Maybe by using Facebook they are distributing this to a lot of other people. That is an interesting area of study. Particularly Gabriel Weimann 17 in Israel has done some interesting work in that area but, has it really democratized public opinion? No, inasmuch as a lot of people are left out of the conversation.

Religion, fake news and mainstream media
Turning to a major social arena: you have written that religion plays a significant Agenda Setting role in the lives of its adherents. Could you explain us why and how? What advice would you give to an organized religion such as the Catholic Church?
We talked a little bit about that previously. Organized religions clearly have an agenda of beliefs and they have various ways of communicating them, beginning with week to week or day to day messages if they have religious services and they have followers. It does not matter whether you are Protestant, Catholic, Muslim, or any other religion. They are clearly trying to set an agenda and they are communicating it.
This brings to mind the Chabad study (Rashi and McCombs 2015). A few years ago, I was approached by one of the Israeli scholars who did the Synagogue historical study of Agenda Setting to work with him on a study of Chabad. It is a very conservative Jewish sect. There are a number of similar groups and most of them in terms of communication regard television and film as great evils that should be avoided. What was different about the rabbi that headed Chabad? He turned it upside down. He said all these mass media are great ways to communicate to adherents scattered around the world. The group is mostly in Israel itself, but there are also many adherents in New York and Los Angeles. We needthe rabbi saidto make effective use of the mass media, mostly television and radio. Instead of telling his adherents to stay away from these evil channels, he told them to use these channels, and they have been very successful at doing that.
Larger religious groups do not have this problem with a very scattered group of adherents. But the television is still useful. About 40 years ago, a very popular television program in the U.S. was Bishop Fulton Sheen's. 18 It not only reached Catholics, but a wider audience. That was a very effective use of television for him to speak about Christian beliefs. So, the communication media can reach members of various religions or perhaps even people who show little interest in a particular religion.
What advice would I give? Basically, I would say think beyond the traditional methods of books, newspapers, and radio programs, especially where there are large numbers of adherents. I am sure that there are now websites as well. Think more broadly about how can you distribute your message effectively and most of all to people who are already your adherents, or to younger people, beginning with children who are growing up.
If Catholic parents send their children to Catholic schools they will learn the catechism. But many children of Catholic parents do not go to Catholic schools. And depending on how faithful their parents are, they may not darken the church door very often. So, you may need to think more broadly about how you reach your audience. There were some ads during the Super Bowl, towards the end of the year of the football season, with the former football coach at Notre Dame. I don't really want to call them ads, let's call them infomercials. He was talking about the importance of religion and primarily addressed Catholics, including Catholics who had fallen away from the church, saying they were welcome back: 'Catholics Come Home.' So, here's a former football coach and at that point also a well-known commentator on the sport programs, a celebrity not so much talking as a spokesperson of the church, but as a lay person within the church. I think he was an ideal spokesperson for a message like that during the football season. So, I think the short answer is think broadly. Follow Chabad [smile].
In the age of the fake news and the post truth, it seems that everything is oriented to cause an effect or set an agenda. One might think that public communication is a matter of imposing one's agenda. Is Agenda Setting compatible with freedom and peace? How to explain that the world is not a war of agendas or what can be done so that it does not end up being it?
My fake news or your fake news? [smile] … It is true: there is a lot of fake news. Is public communication a matter of imposing ones agenda? No, it is a matter of presenting a thorough picture of the day's most important events. You can't report everything that's going on there; and every society needs an institution-in modern times it is the news media-to sift through all that's happening every day and say, here are the most important things. The foundation for this goes back to the Hutchins Commission on Freedom of the Press in the 1940 s in the U.S. Interestingly, it was paid for by Time magazine. Henry Luce, who founded the magazine, was still very much in control, and it was not an ultra conservative, but a conservative magazine. So Luce paid for the commission to do its work, and the commission had a major prescription: you need an accurate and representative picture of the events of the day. I think for the most part the press does that; sometimes they get off on the wrong path or-well, over the past year I think they were totally thrown off balance by the Trump campaign. I think they have got their feet now and, particularly The Washington Post and The New York Times, are fulfilling their watchdog function of reporting about the things that are going on within the government that the public needs to know about. They are not just saying, 'We like these issues, so we're going to push them'. They are putting out issues that they think are important and that are subjected to verification by lots of other news organizations.
How we explain that the world is not a war of agendas?
I think people need to understand how our free press operates. Understand that they do not end up with the same stories every day because the media are in a gigantic conspiracy, that they all get on the phone and say today we are going to push so-andso. Particularly in today's news environment even if they wanted to, there's no time to do so.
People have to understand how a free press operates and that some stories may make a big splash. News that is not all that important just dies. The expression 'Does the story have legs?' means 'Will it last over time? People are turning to trusted sources because they need to understand what's going on. Mainstream newspapers are not the only ones, but a good place to start.
You have written that the tradition of gatekeeping research was transformed by the theory of Agenda Setting. In 2017, can we still talk about gatekeepers? Who are the new gatekeepers, if any? I do not think gatekeeping has been so much transformed by Agenda Setting as incorporated into that portion of the theory that deals with the construction of the media agenda. Intermedia Agenda Setting is part of this. Editors still decide what's going into the paper or not, and what gets page 1 versus page 20. So, I think there still are gatekeepers, and the traditional gatekeeperswho are the editors of the main news organizationsare still the dominant ones; in part because they work for news organizations that are doing most of the original reporting. And that makes their job particularly important. The gatekeepers that we would have talked about forty years ago are still dominant. The social media are much more disorganized. In most instances there is no clear person or set of persons who are the gatekeepers, but if they come up with something new or original it gets picked up by the mainstream media. To that extent, they play secondary gatekeeper. For most of the time, the media is a chorus, but occasionally there is a soloist. Most often the soloist will come from the dominant organizations but occasionally a soloist will come from Facebook or Twitter, who will get on to something, will get it out and will be picked up on the agenda. Overall, much the same positions are still dominant, but their ranks have been enlarged to some degree by social media.
Social media don't do much original reporting. They mostly comment on things that are in the mainstream media or they comment on things that are outside the civic arena. A funny but significant example of that: I remember in one of my seminars we were just starting one day when one student looked over at another student, and said, 'Carolyn, you've got an iPad?' and she said 'Oh yes'. He said: 'You must not have been using it. I have no idea what you had for breakfast this morning'.
There is an awful lot of stuff moving across social media. So, the ranks of the gatekeepers have been enlarged. They are gatekeeping for more channels now because newspapers are producing print editions and online editions and maybe also blogs and Twitter streams, etc.
Need for a more sociological reporting Many people thought that Brexit would never succeed. Also, many thought it was impossible for Trump to win the past US election. Were those results really unpredictable or are mainstream media losing their objectivity, fairness, or balance? What should they do differently in the near future?
In the U.S. election the polls were not fake at all. They predicted the election quite closely because they were looking at the outcome of the popular vote and they said Hillary would win by three points. She won the popular vote by a point and a half. That's pretty much on target. Where they were not looking at in more detail was how particular states would go and how the electoral college would go. They missed it in that sense. They were not being unfair and intentionally inaccurate; it was their traditions and their inclination to be looking at the national horse race that led them to miss the point.
Why it missed Brexit, I really don't know. More generally I think the problem with the general reportage for both Brexit and Trump's victory was the beat system, which is how most news organizations cover the majority of the news. This goes all the way back to Walter Lippmann. He notes that there are only a few reporters who are stationed in places where important events might surface. Most news coverage is coverage of institutions in an organized way: meetings, contacts with representatives of those groups, etc.
In the book Buzz Merrit and I did, 19 he came up with an interesting metaphor of an ecological system. His criticism was that the news media cover the ocean and the big bays and the main rivers going into the big bays, but they do not cover creeks and the marshes and the swamps which is where many issues originate (This is using 'swamp' in a different way than it's been used recently with regard to Washington). But pursuing that ecological metaphor, there were a lot of people out there in the swamp who were very unhappy, who had been left behind by the rapid changes in the economy. Suddenly people who had good well-paying jobs in manufacturing were out of work. When I grew up in Birmingham, Alabama, 50% of the people there worked for one company, U.S. Steel, and my father worked for another steel company. So, the bulk of the people in Birmingham at that time worked in the steel industry. Ninety percent of that is gone. Times have changed. People used to make 20-25 dollars an hour and then suddenly these plants were closed down. Iron mostly from China can be sold in Birmingham cheaper than iron that is made here.
If you get more involved in the swamp, you can sense some of these trends among the public coming. I wrote a piece some time ago saying every local newspaper should have at least an editor who routinely reads the neighborhood websites to know what are they talking about … On those websites frequently you get things like 'Could you recommend a good plumber?' but there are also comments on the public schools or zoning in the area. If it's primarily a residential area, they may ask 'Do we want to allow five story apartment buildings in the neighborhood?' That's one version of the swamp. If you read those websites on a regular basis you would see what the rising concerns and issues are. We need to have more reporting that gets down to that area or even to something like what in sociology is called 'participant observation,' where you hang around with people in different groups. You know it might be hanging around a bowling alley. Occasionally, they get into serious conversation. What are they talking about in barbershops and other places where people gather?
A good example is a piece published recently by The New York Times on people who drive long distance trucks. The journalist basically spent many days at a truck stop in Indiana where two interstate highways cross. This truck stop has room for 300 trucks. He just went around talking to truck drivers. Increasingly driving trucks is a less and less viable occupation, and he emphasized the hardships. If you drive a truck back and forth across country, you are isolated, you spend your days alone. That's why many drivers have dogs. At least they have some companionship. Occasionally also you will see a husband and wife team driving a big truck. I think we need more reporting like that. One of the interviews featured a woman from Los Angeles (California), a Hispanic woman. The journalist asked her, 'Why are you driving a truck?' and she said, 'Well I have a daughter who was admitted to the University of California at Davis, and we did not have the money for her to go. Based on the low paying job I had in Los Angeles I could do better driving a truck.' So, she is out there driving a truck, far from home. Her daughter of course is at college. And she said, 'I'll do this till her college debts are all paid off'.
We need moreto put it in an academic waysociological reporting that gets down to people who are impacted in a negative way by changes in society; to what is happening in the swamp, and what is going on in the neighborhood because that eventually will impact what is happening in the bay and the ocean. It is a nice ecological metaphor.
How well do the priorities of the media agenda, as reflected in the daily practice of journalism, correspond to the larger social value and utility of those items? Which do you think is the most important ethical question for journalists today? Does that issue have any direct connection with your Agenda Setting Theory?
The priorities of the media agenda for the most part do correspond to larger social issues, to what is relevant to people. We could write a wonderfully researched piece about Kazakhstan, but I am not sure how many people would read it unless we can show why people need to know what is going on Kazakhstan. Sometimes news organizations do get caught up writing about topics that the public just doesn't care about.
For the most part, there is a reasonable fit between the media agenda and the public agenda because overall the media have a good sense of what is relevant to their public, both in the general sense of what the public needs to know and in terms of personal values. The ethical question for journalists is how to fill their agenda with stories that meet these standards.