Nieman Foundation at Harvard
Browser cookies, as unkillable as cockroaches, won’t be leaving Google Chrome after all
ABOUT                    SUBSCRIBE
March 11, 2011, noon

Pablo Boczkowski: The gap between what reporters write and readers read threatens news orgs’ future

We’ve written several times about Pablo Boczkowski, the Northwestern professor who studies news production and how it is changing in a digital environment. So when I was invited to serve as respondent to a presentation of some of Pablo’s new research at the MIT Communications Forum a couple weeks ago, I was very happy to.

Here, Pablo talks about a series of studies he’s done looking at the gap between the kind of journalism that news organizations produce and the kind of journalism that consumers consume. He’s gathered data from a number of different countries and time periods to see how those interests match up (or don’t). I think we found a few interesting points of disagreement — it’s worth a listen.

I’ve posted a transcript below of both Pablo’s comments and my own (but omitting the very interesting Q&A which followed, which is also worth a listen). For anyone wanting to skip ahead, Pablo’s talk begins at 7:50 in; my response starts at 37:10; the Q&A begins at 57:45. (And if anyone wants an MP3 version for their morning commute, here’s the link.)

Thanks to Pablo for sharing the forum with me (and for saying nice things about the Lab), to David Thorburn for the introduction and the invite, and to Knight Science Fellow Jason Spingarn-Koff for moderating.

Pablo Boczkowski: Thank you for the kind introduction. Thank you. I’m going to stand. I was standing before. I don’t know how to talk and stay still. So I’m going to move around a little bit. And I need to hide this, I’ve been told.

Thank you very much to David and to Susan for the invitation to be here. It’s always great to be back at MIT. I had a fabulous five years here and I’m always pleased to return, and in particular to share work with you and get your input on it. And I’m very much looking forward to the response — I’m a big fan of the Nieman Lab and Josh’s commentary as well. So hopefully he will not rip me apart.

So without further ado, let’s move to the topic because David said — I was planning to talk for 40 minutes and David said 20 to 25. So let’s go into it.

So this book — News at Work, as Jason just said, is a story about the increasing role of imitation or copying or replication in news production — the idea that more and more and more, we have the same news across different outlets.

The problem that I faced in that research project was whether perhaps I did — most of the explanation has to do with changes in the production of news and the behavior of news workers and news organizations.

But perhaps, it occurred to me, through the research — news organizations, which are actually market-driven organizations, were responding to changes in the nature of demand. Perhaps all of us want the same news and therefore news organizations are giving us just what we want? They are optimizing their product to match the nature of demand. Was that the case or wasn’t that the case?

So, to discard or not that alternative explanation, I did a series of studies that were reported in this book that came out with Chicago in October. And I found out that it is not the case, that actually there is a huge mismatch, as I say here, between the supply of information and the demand of information.

Basically, that means that the stories that news organizations consider to be the most newsworthy ones, the most important ones at any given point in time for any day or hour — the stories that are above the fold in print newspapers, or at the top of the hour in the television newscasts, or that they are in the top screen of the website — those stories that are the most important ones for them are not necessarily the stories that consumers consider to be the most important ones.

And this ties back to a long-standing debate between journalists and scholars about the stories that journalists say we need in order to function properly in a liberal democratic tradition, as citizens of the polity — information about national news, international news, business, economics, et cetera — versus the stories that often times we want to read, that are stories usually about sports, crime, or entertainment — that are not necessarily uninteresting or unimportant stories. As for example was the case a week ago, when my former advisor, my buddy and a great guy Trevor Pinch authored — you know, a professor at Cornell University — authored what was the most popular story on at that point in time, which was a story — a very interesting one — about the role of embodied cognition in artificial intelligence and the relationship between us and machines.

So this was a very important story that piggybacked on “Jeopardy,” I think it was the television show, where the computer — right? So you can sneak in some great science study stuff through television entertainment. But, as Robert Park described 70 years ago — a famous sociologist, former newspaper person, founder of the Chicago School of sociology — this is something that journalists and scholars have known for a while: that “The things that most of us would like to publish are not the things that most of us want to read. We may be eager to get into print what is, or seems to be, edifying, but what we actually want to read or watch or listen to are stories that are interesting.”

So this debate has been going on for more than 70 years. And it’s a crucial one, not only for the industry, but also for the role of media in democracy. Interestingly enough, given how important and how longstanding this debate has been, there has been a relative dearth of empirical studies about this subject.

And the few real research studies that have systematically parsed the evidence that have existed have suffered from at least three major limitations.

The first one is that, so far, before the web, most of the studies focused on aggregate measures — ratings or circulation measures or responses that people would give you in surveys. But not necessarily the behavior that tracks how people read or consume or watch, which is story-driven, rather than circulation in a day — weekday versus weekend, et cetera.

The second limitation is that most of the few empirical studies that are out there have either examined the supply of information — what are the stories that are important in the day — and contrasted that to secondary evidence about the stories about which people are interested in. Or examined through surveys, through focus groups, through examination of ratings information, et cetera, the stories that people consider the most popular and use secondary evidence to look at what we think are the most important stories of the day for journalists — the editorial criteria.

And finally, the third limitation is that by and large, almost all the studies have conceived these preferences or these choices as static — as not changing depending on changes in circumstances. They have taken that for granted.

So what I did, after I got my curiosity piqued doing these few studies for News at Work, is I got some grant money and designed a series of studies that have tried to first determine whether there is actually or there isn’t a gap between the supply of information, consisting of the stories that are the most important ones for journalists, versus the demand for information, consisting of the stories that are the most popular for consumers, where consumers vote with their clicks — in seven different countries, 20 different online news organizations and that have tried, by research design to get rid of all these problems. So, focus on the story as a unit of analysis — focus on the story as the unit of analysis — looking at the choices of both, or the preferences of both journalists and consumers, journalists and the public concurrently, and also looking at the role of contextual variation, contextual circumstances, where they affect or not this preference, to get a sense of whether these choices are then static, or these preferences are static or not.

So I am trained as an ethnographer. Most of what I do is ethnographic. But I’m just going to give you numbers. This is a first for me. It’s an interesting challenge to write the book based just on numbers. I’m going to present you four of the papers that came up from these studies. I have a few more. You’ll have to tell me whether you think the overall story makes sense or not. The papers are almost all published or in the pipeline. So I have little doubt about that aspect of the research. But I’m trying to get a sense of the overarching narrative that is basically a tale of four studies.

So study number one is a study where we laid the methodology for this. We did the following. One day, for a number of days we repeated the same thing. We went, together with a group of collaborators, to four online news sites in the U.S. — CNN, the online news site of the Chicago Tribune, the online news site of a paper that now doesn’t exist anymore, the Seattle Post-Intelligencer, and the news operation at Yahoo.

At three times of the day, at the same time, we grabbed the top 10 most prominently displayed stories, which are basically the stories that appear on the first screen, as an indication of the stories that journalists consider to be the most important ones of the day.

At that very same time, we took the top 10 most-clicked or most-viewed or most popular stories on each of these sites. Whenever possible, I also collected information about the most-clicked and the most-commented — I’m sorry, the most-emailed and the most-commented stories. I have analyzed that as well. I’m not going to report it today. But in the Q&A I’m happy to tell you. There are a couple of papers out there. I’m happy to give you a gist of what that says.

So the methodology is very simple. We take the two rankings — the ranking of what the journalists consider to be the most important news and the ranking of what the consumers consider to be the most important — the most popular, interesting news.

Then we contrast. We say, of these top ten stories in one ranking, how many are about what I call public affairs news? News about politics or national news in this country, international, foreign news, business and economics — one side. The other side — the rest, which is usually weather, crime, sports, entertainment. Okay? I measure whether the level of preference is similar or is different. It’s a very, very simple process, a very crude measure.

But they give remarkably consistent results across sites. On the left side here, in this turquoise color, for each of these sites is the prevalence of public affairs news — the news that we need to know in order to function properly as citizens in a liberal democracy, presented by journalists among the top 10 in the four sites.

On the Northwestern purple side, the color here on the right-hand side, you have the prevalence of public affairs news among the top 10 most-clicked stories in each of these sites. In all cases, despite the fact that you have two metro papers here versus two sort of global/national organizations at the two ends, that one is a pure player and the others come from traditional media — it doesn’t matter.

You can change as much as you want. In all cases there is a double-digit gap between supply and demand.

Audience member: What is the y-axis?

Pablo: The y-axis is the percentage of public affairs news among the top 10. So in all cases — and all of these figures that I’m going to present today, everything is statistically significant. So we have about 1,200 stories for each of these sites — hand-coded, not machine-coded, hand-coded by people trained beforehand, blah, blah, blah. In all cases, you have a gap.

Now it is possible, we know, that our choices, when we read a site, we are influenced — or when we read a newspaper — we are influenced by editorial decision-making. That is, if this story makes it above the fold, we are more likely to pay attention to that story than if it is buried in page 15. If it is placed at the top of the screen of a website, it’s more likely that it will attract hits or at least some visibility than if it’s buried several screens deep.

Conversely, I am not the only one who is looking at this. Journalists in news organizations are constantly tracking this information. They use now a tool called Chartbeat. For a couple of hundred bucks a month, you get very detailed information about the heartbeat of a site. “I saw the Trevor Pinch story is generating lots of clicks, so we move it up in the site.” Possibly or possibly not.

So it is possible not only that our behavior as consumers is influenced by the journalists, but also that journalists’ behavior is influenced by consumers. So we did a second analysis of this information.

Of the top 10 stories, there were some in the two columns, there were some that showed up in both. That could be evidence that journalists have placed them at the top because consumers had actually clicked a lot or that consumers had clicked a lot because journalists placed them at the top.

So we extracted, removed all those stories, to get a purer sense of the preferences of each site. Guess what? The gap increases hugely. So when you have, actually left to our own devices, the preferences of the choices of people diverge, even more. This is in the case of CNN, a gap of 50 percentage points. It’s like basically running — having a bakery, putting all your products — you decide that 70 percent of what you are going to produce and put on the market are croissants, and only 20 percent of those are sold. You wonder how much longer can an organization survive with this huge gap between supply and demand.

So, okay. We basically proved that there is huge, sizable, quite robust gap between the needs and the wants, if you wish, between the supply and demand of information in the case of the States. But the States to present one of my favorite analogies, is the home of McDonalds. So perhaps people do not eat McDonalds in other countries.

We run a second study. That basically is exactly the same methodology to look at all the new sites of national newspapers — and like the previous study we just worked with national newspaper data for comparability reasons in six foreign countries. Three in Latin America and three in western Europe.

For each country except for the case of Brazil, because we were also interested in issues, not just of regional difference but also ideological differences, we looked at two sites per country. One, conservative leaning. The other liberal/centrist-leaning. So perhaps this is a thing of the right, not or the left, or vice versa.

These six countries are Germany, U.K., and Spain, in Argentina — I’m sorry, in Western Europe, and then Argentina, Brazil, and Mexico in Latin America. The first paper is coming out in Journal of Communication is saying and one just came out in Communication Research. Same ideas, same measures, similar results.

This is each site — so we have Tagesspiegel and Welt from Germany, El Pais and El Mundo from Spain, Guardian and Times from the U.K., Clarín and La Nación from Argentina, Folha de Sao Paulo from Brazil, and Universal and La Reforma from Mexico. In all cases, there is a sizable and significant gap, in some cases up to 30 percentage points between what is provided to the public and what the public actually wants to read, based on where they click.

Perhaps, however, this is more a thing of Europe, not Latin America, or Latin America and not Europe. No — not only the editorial criteria are remarkably similar across very different regions, but also consumer behavior. Here you have journalist choices. The prevalence of public affairs news in western Europe on the left and Latin America — or the three countries in western Europe, three countries in Latin America on the right. In both cases, almost 60 percent — there’s one percentage point difference. Here, preference of consumers, which are exactly at 39 percent, on average, combining all the sites together. And by the way this is not machine-coded. This is real people, people who lived in those countries for at least a year, and two people per country. We are talking a total 18,000 stories in this study, hand-coded this way. Perhaps, however, there is an ideological difference. This only applies to sites on the right. Because those of us on the left are a little bit more enlightened! Well, not actually.

You have, again the journalists here, the consumers here, this is conservative…it doesn’t really matter, right? Conservative, liberal, conservative, liberal — one percentage point difference if you group together all the conservative sides, plus group together all the liberals sides.

So, now that we know that there is gap, the gap is sizable, the gap applies in at least three different regions of the world, they got that doesn’t seem to vary depending on ideological preferences — we can start testing whether it is static or not. So, whether perhaps things change when people should pay more attention than other normal circumstances. And what better time to study that than during the 2008 election, which was not only just a presidential election, it was a major political moment. It was possibly the election of the first African-American president in the history of the country, coupled with a momentous financial transformation that you erased I don’t know how many trillions of dollars in peoples’ assets.

So if there is any point in time in which people should pay attention to politics, economics, international affairs, it’s this time. So the same methodology. We looked at six U.S.-based news sites. Two from cable, CNN and Fox. So two very different ones in terms of ideological orientation. Two from broadcast television, ABC and CBS. And two from print, Washington Post, a serious site, and USA Today, the antithesis.

We started collecting information three weeks before the first convention and we ended three weeks after the election. On average, we collected — for the normal weeks we collected between four and five times a week and around the election period, we collected for 14 consecutive days.

And then we went back exactly a year later, around those 14 days, and we gathered information during a more routine period — those same sites, same 14 days and I will report on that as well. We also gathered information during the congressional through the same 14 days in 2010 and we’re going to do a panel — we’re going to continue collecting information at least until 2012, if not longer than that.

So, I’m going to give you two cuts of the data. The first one compares what happens in the 14 days around the election time in 2008 with the same 14 days in 2009. Not the full load for ’08 — just these 14 days, those 14 days.

And here we start seeing that these choice is that actually not static. So the gap exceeds — is huge actually in Fox. So people, consumers of were not really very interested in politics around election time. It’s still fairly large on CNN and it’s also large in USA Today and statistically significant in all cases.

It doesn’t exist in ABC and CBS at this one statistical vantage point. And the interesting things at the Washington Post is that there is a gap that reverses. Okay, so eight of over 10 top stories that journalists published in the Washington Post around election time were on politics. But that wasn’t enough for the public of Washington Post. They wanted more than nine out of ten. So the answer is, are these preferences is static? No. They actually changed quite a bit, depending — and when I show you another measure where if that is the case — depending on the context. When there’s a major election and people are trying to pay attention, people actually do pay attention. What happens when life goes back to normal? Everything goes back to normal. Gap large, everything in the same direction, no changes from two years ago. No changes, I can predict, if you measure now in all six sites. So this is what they interannual comparison, from one year to the next, tells us.

Then we did an analysis looking at the behavior both of journalists and consumers during the election cycle. So this chart gives information about the journalist’s choices starting from week one, which is prior the first convention, going through week 18, three weeks after the election, with week 15 being the week of the election. Each site has a color coding. Basically — if you do as we did through a regression — you can predict that from one week to the next, there is basically on average a three percent chance that as the election nears journalists are going to give us more public affairs news.

Fairly predictable, and the more predictable things that the journalists’ choices basically move in a pack. They all go; it doesn’t matter where you look. It doesn’t matter whether you look at on Fox or Washington Post. They are more or less going to give us the same diet.

So have this picture in mind — look at us. This is us. This is the public from one week to the next, right? Huge disparity. So in some sites, you can actually plot a regression. Most of the states you can’t predict from one week to the next. This is incredible. This is Fox. This is absolutely incredible. Okay?

Audience member: [Off mic question]

Pablo: No, the election is week 15. This is how much the election drew the attention. The best of three out of 10 stories in that week. Among the top 10 most clicked ones. The interesting thing or the most noticeable thing is that compare this with this. This with this. This is an industry that is fairly static. That doesn’t move depending on the context.

That doesn’t change. They have a formula, they push it. One, two, three, four, five. This is us. We change our moods. Today I want croissants. Tomorrow I want bagel. But I’m actually going to same bakery — the same bakery always giving me croissants. After a while, my appetite is going to be not satiated or not satisfied.

So eventually, which is one of the implications I think of all this, eventually people are going to start leaving these sites. There is no industry that can survive, especially in a competitive environment with this huge gap — [Jason signals two more minutes] I have two? Ha. I come from Argentina so with hyperinflation we say we eight, how about that?

So, okay, the moral of the story is that we can tell, okay, these choices are actually not static. They’re dynamic. They change depending on the environment, and of course this is a major environmental transformation, a major contextual variation in the case of the States — something that people had been clued in for months.

Would the same changes happen during a major crisis that erupts like this, is totally unforeseen, and vanishes quickly after that? So we run a fourth study that came out in Press/Politics last year, too. We basically were fortunate enough that during the day of election period for the international study, in Argentina there was one such crisis, a major political crisis where the economic minister had to resign, the government almost was collapsed, et cetera. And so what we did is we looked at the evolution — because we had 24 weeks of data, we looked at the evolution of covering during those 24 weeks. We also, the summer after we collected data, we also went to these organizations and a few others and interviewed a bunch of bunch of editors and also interviewed a number of consumers.

Long story, short, since I don’t have a lot of time, the choices are static only when there are no major contextual transformations. Even in case of a crisis erupting like this one, it only lasts for five weeks. Consumers really pay attention to topics that journalists consider politically important in those weeks. And then, when the crisis is over, people go back to life as usual. So this is the evolution of coverage during the 24 weeks, weeks 18 through 23 are the weeks in which the crisis was taking place. If you look at the lines, the blue lines are the…the solid one is the journalists’ choices for Clarín, which is the largest newspaper in the country. The broken ones are the consumer’s choices for the site, and the same for La Nación, which is the second largest news site in the country. They basically don’t move a lot together during normal periods. When there is a major crisis, they converge. And as soon as the crisis is over, they diverge.

That’s another way of showing the data. So, why? Why, given that we all know — I know it and people in the industry know it and people in these sites know it — given that they are publishing stuff that, a lot of it goes unnoticed, it’s not paid enough attention, why do journalists continue publishing this information? Why is their behavior like that?

One of our editors told us, “We cannot ignore what people are interested in, but we cannot make the news as a function of those interests either…I believe that journalism has a role that is different from following those [consumer demand] trends.” Another one, different organization, Clarín, said, “We are guided by parameters that have to do with taking care of the brand. We cannot publish anything that is out there.”

So basically this is saying that there is an occupational logic that, so far, trumps the market logic. We know what the demand is, but our occupational logic is such that will trump the nature of demand.

What happens on the consumer side? Why do consumers first click on public affairs or pay attention to public affairs news during crisis periods, but not during ordinary periods? One of them said, “You know, we have the agricultural producer strike,” which is this political crisis, “so now I follow the headlines from this story and if there is something I can read about it, I read it.”

“But during normal periods,” another said, “I don’t like politics very much, but I have to pay a bit of attention to this. Whereas, normally, I read bizarre news to take a break from current events.”

So basically, just to summarize a lot of interviews, people consider political news or public affairs news anxiety-provoking, demanding a lot of cognitive effort. And they feel somewhat unprepared to interpret it. All of that moves them away from this information. So we have two very different cultures and two very different logics here.

Jason: I think we should, if it’s all right with you, save the rest for the discussion.

Pablo: Give me two minutes.

Jason: Two minutes? Okay.

Pablo: Two slides. I only have two more slides.

Jason: We’re definitely getting into hyperinflation mode here.

Pablo: I know. So what does this mean for the industry? One, the media industry is an industry that grew and developed within the natural monopoly or oligopoly position. You had to come to them — you’re an advertiser, you had to come to them. They gave us what they considered was important, and basically when you are one or two, at the most three players in the market, you can actually ignore the nature of demand.

Now the industry has moved to a much more competitive environment. So this gap that has existed, according to Robert Park, for at least 70 years, now becomes terribly pressing in terms of the economic viability of the industry.

Second, this highlights the growing tension that you see in each and every newsroom that I’ve been to between the logics of the occupation and the logics of the market that I said before.

Third, the Washington Post finding is, in part, an interesting one for me, the reverse nature of the gap, because it tells us that the niche sites, the sites that have a specialty, are siphoning interest from the generalist sites. People who are in St. Louis or people who are in Seattle who are interested in politics go to Politico or the Washington Post — they don’t go to their local news organization.

So the bundled product strategy, the generalist strategy that dominated the industry for a long while, actually might no longer be so feasible. And since Jason is looking at me, “Come on. Come on. You have to finish.” I leave you with that last slide so that you can read and I can talk about it. Thank you. [applause]

Jason: It’s fascinating. So that’s brand-new research.

Pablo: It’s brand new.

Jason: Are we the first to see it?

Pablo: No. [laughter]

Jason: Great. That’s exciting. That’s going to be part of a new book?

Pablo: Yes. Hopefully.

Jason: Okay. Josh.

Joshua Benton: Terrific. Well, thanks very much. Thank you, Pablo, for that presentation. It’s a great honor to share the table with you. It’s a great pleasure to be here at MIT.

While I’ve been at Harvard for almost four years now, I really do come at a lot of these questions fundamentally from a journalistic perspective as opposed to an academic one. So while I can’t match Pablo in academic credentials, I do think journalists — traditionally their task is to ask questions. So I’m just going to raise a few questions about some of the research that we just heard presented that I hope can spark a fruitful discussion.

I appreciate all the work that Pablo has done to illustrate the gap that exists between the production of journalism and the consumption of journalism. That said, I’d be surprised if anyone here was really shocked by the idea that, in fact, the things that we consume are not the same things that we would like to see produced or that professional journalists would like to see produced.

I suspect we’d find a similar gap if we looked at the amount of broccoli you’re supposed to eat versus the amount that people actually do eat. Or the number of hours that Americans spend listening to Justin Bieber versus the number of hours they admit to listening to Justin Beiber.

I think we can all, even from our own experience, acknowledge that our own consumption of media is not always optimized for seriousness or for civic worthiness.

I’d actually be interested, if you’re looking for something else for the last chapter of your book, to do similar research looking specifically at the consumption habits of the same journalists who are producing the stories — because I’m sure at the end of a long day of muckraking a lot of them will go back to go read a trashy novel about vampires or zombies.

I, last night, used another tool to try and gauge the interest of the Internet-using public — this is Google Trends which is matching what people are searching for at any given moment. The top five there include a woman named Melissa Molinaro who, from what I could tell is only of interest because she’s in an Old Navy commercial and looks something like Kim Kardashian.

The word “forehead” — I was completely unable to figure why people were searching for forehead last night, but perhaps someone in the crowd can crowdsource that for me.

Two searches related to the death of the star of a reality television show, “The Deadliest Catch.” And “Scott Walker prank call,” which — that last one, that fifth one could be described as public affairs journalism if you’re willing to have a broad definition. But I’m willing to invoke the prank-call exception to anything actually being public affairs journalism. I think that sort of moves it into the entertainment category for most people who are consuming it.

The Internet certainly makes it very easy for us to quantify the public’s wandering attentions. But I would note that it’s not exactly a new phenomenon for there to be concern about the civic information levels of the American public.

I remember, as a kid in the ’80s, hearing about all the Americans who couldn’t find the Soviet Union on a map — which is really hard because the Soviet Union was pretty big, pretty hard to miss. There have always been people who, while they may have bought a very serious newspaper, skip straight to the comics or the sports section or the crossword puzzle or who, quite frankly, just wanted the ads. Anyone who’s worked at a newspaper will be able to tell you that Sunday paper sales are largely driven by people who take the newspaper, set it aside, and then just go through the circulars.

Joel Kramer who was the publisher of the Minneapolis Star Tribune and now runs a nonprofit news site called MinnPost, he told me that their readership research at the Star Tribune consistently found, over the years, that about 10 to 15 percent of their audience were people he would consider serious news consumers. People who sought out news as part of their civic identity, people who viewed it as sort of a life priority to get different perspectives and to stay informed with the latest happenings with government.

It’s easy in hindsight to imagine some idealized past where everyone was a dedicated public affairs reader — and maybe that was true in Cambridge, I confess. But I just want to raise a few questions.

One, before I go into it — one that just came to mind listening to Pablo’s presentation. It’s certainly true that some of the listing of the most popular stories and the most promoted stories on news sites may have some taint between them as news organizations realize, “Hey, this story is popular. We should move it up.”

But I think it’s difficult to argue that the presentation of the stories on a front page or the first few stories in a newscast have ever been a pure reflection of journalist’s judgments on what are the most important stories.

Page 1 meeting of the newspapers, the environment I’m most familiar with, usually had a mix. “We have to get the right mix — a few hard stories and a nice feature story about some teen that’s done something spectacular.” The presence on the front page was never a full statement of “this is the most important story of the day.”

But that aside, I want to just raise three questions that hopefully can be of use in our conversation. The first one is: Are we evaluating the right news universe? To the extent that I have a quibble with some of Pablo’s work and with News At Work, his last book, it’s that to my mind it doesn’t quite reflect the reality that the news universe is no longer limited, either from a consumer’s point of view or a producer’s point of view, to the large news institutions — either CNN or The New York Times or even the newcomers like Yahoo.

That’s the one question that I would push back a bit on from Pablo’s presentation is I’m not 100 percent sure that supply and demand is exactly the right metaphor here, because it’s really hard for me to look at the Internet and say there is a shortage of supply of anything.

It’s hard, whether you’re just limiting it to one news organization or going beyond — if you want to find out about what’s going on in Libya, you can easily spend the next 24 hours pouring through lots of information. It seems it’s less a matter of supply and demand and more one of distribution and promotion.

Also, along those lines, looking simply at the top 10 stories at any given moment, it’s not a really strong, to my mind, statement of supply. It’s only even a rough statement of consumption. If you’re looking at a site the size of, say, The New York Times or The Washington Post, even the top 10 stories that are consumed at any given moment are still only a tiny fraction, or at least a small fraction, nowhere near a majority of the amount of news reading that is going on at any given time.

If you just looked at just purely the public affairs production of something like The New York Times or even CNN, it’s certainly a list of stories much longer than 10 every single day.

When we’re comparing print to online, I think it’s very important for us to not just carve off the serious part of the print world and make that the point of comparison to the enormous, glorious confusion of the online world — or vice versa. Even in print, before the days of the Internet, more people read People than read Time, more people read Ladies’ Home Journal and Good Housekeeping than either of those. And more people, today read Game Informer magazine and Better Homes and Gardens than any of the other ones that I’ve just mentioned.

The web, particularly thanks to social media but not exclusively, does a really terrific job of connecting quality news and information and the people who might want to consume it. If you ever just look at the things that your friends are sharing on Twitter or on Facebook, they tend to be a pretty good reflection of your interests — assuming that you pick your friends correctly, which is always a different question.

As a result, there are lots of outlets that existed before the Internet — in some cases thrived before the Internet — who are now reaching all new audience that they would not have been able to achieve without the connecting power of online media. There are many, many more people reading the journalism produced by The New York Times newsroom today than has ever been the case. The Atlantic is doing great gangbusters on the web, generating lots of revenue and lots of readers by having a very smart digital strategy that is aiming at high-quality content. The New York Review of Books, of all people, is doing quite a good job online and reaching all sorts of new audiences who would never have gone to the one newsstand in their Midwestern hometown that would have carried a copy of The New York Review of Books.

There’s lots of terrific, serious criticism of the arts and books on book blogs, on little, nerdy startup sites. Because the web is such a good distribution and connection platform, that sort of quality content is able to reach new audiences. I would love to be able to see a way that that broadened effect, both on the high-end and on the low-end in terms of high and lowbrow — to see that reflected.

If we’re going to take the next step and evaluate the impact of these changes in the news universe and how we consume and produce news, it’s important to acknowledge these sorts of changes. As Ethan Zuckerman at Harvard has said often, for subjects like foreign affairs, it really is not so much a supply question as it is a demand question. It’s not how can we get more information about Libya. It’s how can we get more people to be interested in information about Libya.

The second question I wanted to raise is: Are news organizations public institutions or businesses? Now, of course, I can easily answer that for you. The answer is both. But which lens we view them through should inform, I think, the way that we view this gap between the public’s interest and journalistic production.

I think viewing news organizations as being primarily civic-minded institutions, is really only tenable if you ignore the vast majority of news that has been produced in the history of the world. Even before the Internet, an average American newspaper would only spend about 20 percent of its revenue, of its budget on the newsroom.

In that newsroom were lots of people — sports reporters, movie critics, other people who were writing perfectly valuable stuff, but stuff that it’s hard to say was critical to democracy in a lot of ways, or critical to being an informed voter. Even The New York Times spends an awful lot of resources covering the Yankees and fashion and pop music and whatever hipsters are up to in Brooklyn.

So what percentage of an average newsroom would pass the “Is it critical to democracy” test? It’s a lot less than 100 percent, that’s for sure. And even given that newspapers do an awful lot of important public affairs journalism, which they certainly do, why do they do it? Did they do it purely out of a civic responsibility? Or did they do it because they felt it was good business?

Again, the answer is both. But historically speaking, the story of newspapers in the 20th century was one of consolidation. Going from cities with dozens of newspapers to cities that would have, in most cases only one, and occasionally two. And if you live in New York, you get a few more.

That consolidation, in a variety of ways, led to spread of distribution in coverage area. You went from, if you were at my old newspaper, the Dallas Morning News, you were no longer covering the area immediately around Dallas. You were suddenly covering this sprawling metropolis of six million people.

Within those six million people, there are enormous disparities in wealth, in race and ethnicity, in interests, in socioeconomic status. So the economic incentive, if you want to be a monopoly player under this model and to reach as many people as you possibly can, is to focus your coverage on the broadest possible common interests. That’s why The Dallas Morning News wrote an awful lot of stories about the Dallas Cowboys and why the Globe writes a lot about the Celtics and the Red Sox.

It’s also one reason why the Globe writes a lot about Deval Patrick and the Statehouse — because when you have a large metropolitan newspaper that is trying to appeal to a large, very diverse audience, it’s efficient, from an economic point of view, to focus on the things that appeal most broadly.

It’s also the reason, on the other hand, why newspapers have traditionally been pretty darn bad, I would argue, at covering niches and subcultures and the small — the little things going on that a broad metro newspaper, you can almost always count to not do a particularly good job on.

From my own experience, I remember the day when I was looking through the newspaper’s archives in Dallas for information about the bar where the Velvet Underground recorded a live album in 1969. I realized that, after spending lots of time looking for references, that The Dallas Morning News essentially chose not to cover the 1960s, chose to ignore the counterculture, and treated as something they were not particularly interested in.

We had the example a few years ago, where the Lexington Herald-Leader published a front-page apology saying, as a correction, “We’re sorry that we forgot to cover the civil rights movement.” There are lots of ways in which newspapers, by trying to appeal in the broadest possible ways, focus their coverage in ways that leave open lots of niches to be exploited by new, more agile players.

Also remember that newspapers are not just selling news. They are also selling an image of their readers. They are selling that image to advertisers of course — “you should want to put your goods in front of these wonderful people.” But they’re also selling it to readers themselves. People’s identities are influenced by the kinds of media they consume.

I remember the great saying — I’ve been trying to figure out who said this, if anyone knows, please talk to me afterwards — that there are three kinds of people in the world: those who don’t read the New York Review of Books, those who do read the New York Review of Books and those who no longer have to read The New York Review of Books.

These sorts of identity issues are important to the ways that people consume their media. While there are lots of journalists who of course feel a civic duty to do serious reporting, and there are publishers who feel similarly, there are also economic motives, trying to get a more upmarket audience. It’s something which a lot of metro newspapers are very much focused on, and other news outlets as well.

Also, that sense of responsibility, that civic gravitas that the classic publisher had — well, not to demean that in any way, but it’s also a pretty easy sense of responsibility to feel when you have a local monopoly that will produce 30 percent profit margins without trying all that hard. When the question is, do you have a 30 percent profit margin or a 20 percent profit margin, and how much journalism you are willing to invest in to make that difference, the questions are a little bit different. So these business factors and economic motivations are also important, I think.

My third and final question is a little bit more — I hope there are good political scientists in the room: Is it okay for people to not be informed about the news? How much does it really matter?

I’m not surprised at all that the closest connection that Pablo found between news produced and news consumed was around the 2008 election, when, logically speaking, interest in public affairs should have been at their highest point, by most theories of what an informed public is supposed to be.

When I saw the numbers for Fox, I was reminded actually of Nick Lemann’s great profile a couple of years ago in The New Yorker of Bill O’Reilly, where he noted that while people who don’t actually watch Bill O’Reilly may think of it as being primarily a political show, but if you sit down and watch it, there are a lot of crime stories, a lot of “child abducted” kinds of stories that the political focus almost seemed secondary on a lot of days.

Jay Hamilton, who is an economist at Duke — his book All The News That’s Fit to Sell, a terrific book I would recommend — he outlines what he calls the four kinds of information demands, the four reasons that people would seek out and consume information.

The first is producer information. That means information that would improve or increase your production. So if you’re a stock trader and you find out that a company in which you’re interested just got a big new contract or is facing a lawsuit, that’s information that you can use to make you do your job better. So The Wall Street Journal would be producer information for you. The same is true if you are an agent in Hollywood and you are reading Variety. And also, to get to Pablo’s data, I think it’s also completely true if you are a reader of The Washington Post, and therefore probably affiliated with the federal government in some way, and it’s a few days before your megaboss is about to change, it’s very much tied to your production in your job.

The second type of consumer information is information that lets you make better consumer decisions. So, should I buy the new MacBook Pro that was unveiled this morning? If I want to risk injury to go see Spiderman on Broadway, will it be worth it? Reviewers of all kinds are the obvious suppliers of this kind of consumer information.

Entertainment information is what Lindsay Lohan is up to today, plus also a Philip Roth novel. It is information that is primarily aimed at entertaining the consumer in some way or another. An awful lot of what news organizations produce, I would argue, falls directly into this category. Journalists may not like to admit it. But I think it’s true.

Then finally, the fourth category is voter information. That’s information that lets the reader make better decisions about his or her government. Should I vote for Obama? Is he handling Libya correctly? Was the stimulus worth it?

And what Hamilton argues is that, for the first three kinds of information, there is an immediate return to the consumer. You make more money. You avoid Spiderman. You have a good laugh or a good cry.

With voter information though, knowing doesn’t give you an immediate return. Knowing about the stimulus doesn’t mean that the stimulus will have a different impact on you. It simply means that you’ll know more about it. Because the chance of your individual vote being the difference maker in any given election is so small, you don’t derive the same direct benefit from that knowledge that you do from the other classes.

Hamilton argues that, for this reason, it can be a perfectly rational economic decision for someone to choose not to follow the more broccoli side of the news. And frankly, these choices are something that we all — even if you are the biggest news omnivore in the room, there are still lots and lots of things that you choose not to read about every day. You may be completely up on the latest impact of the stimulus, but perhaps the Botswanan economy is not your major focus. We all make these choices.

Hamilton says that he views the elements that lead people to pursue a lot of voter information as what he calls the three Ds — duty, diversion, or drama. I think a lot of cable news would fall under diversion or drama, and not many other people would fall under having the sense of duty or following it.

So the question I really raise is, particularly in the kind of political system that we have, where the political science literature tells us that the best predictor of whether or not someone is going to be reelected as president is the level of unemployment right before the election, is it not potentially a rational choice for people to be choosing to focus less on public affairs news?

Is it okay for us instead to have a class of people, whether that’s politicians, but also journalists, lobby groups, academics, as well as those people driven by duty — to have this class of people who do the heavy lifting and pay attention to these things?

Speaking from my own experience, when I was writing about education policy in Texas, some of my stories I knew would reach a very broad audience and would be of interest to a broad audience. But a lot of them I knew were kind of targeted to a few dozen bureaucrats and administrators and legislators, for the hard-core policy questions that weren’t going to reach far beyond that.

How important is that element of mass in the mass media?

Jason: Josh, I just want to give you a warning.

Josh: Yes. All right. I am duly warned. When we talk about agenda setting, which I think is a slide that you may not have gotten to, I think one of your last slides, we live in a world where Gawker can take down a congressman, and Talking Points Memo can get an attorney general to resign, and Wikileaks can set the whole world talking. The ability to set the agenda is much more broadly spread than ever before.

And I think the impact we’ve seen in the economic structure of news organizations is moving away from the world of duplication to a world that is much more about niche. When you look at Washington coverage, there are fewer people than there were 20 years ago doing broad Washington coverage, of the sort that the Washington Bureau of the Des Moines Register would have done at some point in the past. But there are many more people doing niche coverage, about the elements of federal policy that are important to them.

For the people who are really interested, the people who are reading Politico and The Washington Post, there is no shortage of blanket coverage. It’s just that they’re reaching different audiences instead of trying to appeal to the broad element.

It’s really hard for me to look at the 700 feeds in my RSS reader and see more conformity or a new bottleneck for diverse subjects or diverse points of view.

But finally, to conclude, I think that the example of Wikileaks is instructive, because Julian Assange realized after working for several years to do what he wanted to do, at a certain point he needed to work with news organizations like the Times and the Guardian to have his work make the maximum impact. Those news organizations still have the enormous reach of their audience — an enormous weapon in the battle of what kind of media is going to be produced and consumed. So I think in the end there is going to be lots of room for lots of different kinds of players. The ability to judge the actions of the audience, as one unit, will get a lot harder.

We going to have to start looking at more narrow slices of the news production and consumption question. And I think in the end there will be room for all kinds. Thanks.

Joshua Benton is the senior writer and former director of Nieman Lab. You can reach him via email ( or Twitter DM (@jbenton).
POSTED     March 11, 2011, noon
Show tags
Join the 60,000 who get the freshest future-of-journalism news in our daily email.
Browser cookies, as unkillable as cockroaches, won’t be leaving Google Chrome after all
Google — which planned to block third-party cookies in 2022, then 2023, then 2024, then 2025 — now says it won’t block them after all. A big win for adtech, but what about publishers?
Would you pay to be able to quit TikTok and Instagram? You’d be surprised how many would
“The relationship he has uncovered is more like the co-dependence seen in a destructive relationship, or the way we relate to addictive products such as tobacco that we know are doing us harm.”
BREAKING: The ways people hear about big news these days; “into a million pieces,” says source
The New York Times and the Washington Post compete with meme accounts for the chance to be first with a big headline.