February 13, 2005
Media Bias: An Overview
In the wake of the Gannon and Jordan affairs, there has been a resurrection of discussions about media bias: whether it exists, and its nature if it does exist. For myself, this examination began recently with my post on the Fourth Estate in which I evidently forgot to code my html for hyperbole for calling journalists Stalinist, and was fairly called out on it by Rick DeMent of Unspecified Chatter (nee, the Rant).
More broadly, two recent blog posts have provided a reasonably cogent defense of the so-called Mainstream Media (MSM) from folks sitting within the game itself; the first, from Steve Silver (courtesy Michael Totten) gets what appears to be a fairly solid fisking from commenter JBP at February 10, 2005 09:30 AM in Michael Totten's thread. The other is by one Derek Rose, who is also a 'member' of the media, and takes the form of a rebuttal of an open letter by a Marine in Iraq. In this one, Mr. Rose is deserving of credit, if nothing else, by sticking around in the comments and replying to his detractors.
But unfortunately, neither of these defenses or much of the immediate response provides much more than a he said, she said approach and fails to apply much thought to the problem. This is, to a certain extent, to be expected since folks on both sides of the aisle seem to think that the day-to-day conduct of major media outlets is, in and of itself, prima facie evidence of bias of some sort or another.
So, let's go back to the data, and some of the probable responses to the data.
First, we've got a collection of large number of studies on media bias cataloged by the Media Research Center. Without drilling down through each of them individually, we can get a generalized sense of what they're driving at by noting the general trend of the reports. In terms of both party identification and voting behaviors, the journalistic corps is roughly as far to the left as an average Berkley, California; Cambridge, Massachusetts; or, a Washington, DC. More or less, journalists who report at the national level tend to vote Democrat by something like a 10-to-1 margin with party identification to match.
The counterargument to this might be related to something along the lines that political leanings don't a reporting bias make. Folks may have all manner of political leanings without making that a pre-eminent focus of their work. After all, we look to judges to rule without bias. Whether or not the folks on the Supreme Court do so or not, we certainly do not go into a court for a traffic ticket and immediately ask how the judge voted in the last election. Besides which, if reporters go into the game with a full intention to report the truth wholly, why is it that one would assume that they are incapable of reporting without bias any more than one might have problems assuming that a doctor act within the guidelines established by the Hippocratic Oath? If nothing else, wouldn't it be much more likely to assume that the news media are beholden to corporate interests, rather than purely political motives?
We can approach the question of bias by looking at what news outlets use for information in an effort to see if their sources reflect a non-partisan selection. A study done at Stanford and later posted by Yale (link now expired, new link appreciated) notes that media outlets tend rather strongly toward citing the same bits of information heavily preferred by Democratic politicians. Absent the actual study, I have been able to find a few key paragraphs from an article about the study (Courtesy Protein Wisdom):
"Our results show a very significant liberal bias," they write. "One of our measures found that The Drudge Report is the most centrist of all media outlets in our sample. Our other measure found that Fox News' Special Report is the most centrist." And all three papers, plus NBC and CBS, "were closer to the average Democrat in Congress than to the median member of the House of Representatives." Fair and balanced, anyone? To use a simplified example, they say, suppose there were only two think tanks, and The New York Times cited the liberal one twice as often as the conservative one. Then the newspaper's ADA score would be the same as that of a member of Congress who did the same.
The estimated ADA score for Fox, based on citations, was 35.6. That puts it in the company of Sen. Olympia Snowe, R-Maine, and a few points below the House median, 39.0. The two highest were The New York Times, at 67.6, and CBS Evening News, at 70.0. The average Republican in Congress has an ADA score of 11.2, and the average Democrat 74.1.
The authors say they expected to find that the mainstream media leaned to the left, but they were "astounded by the degree." So when people say, for example, that The New York Times may be tilted left, but people can compensate for that by watching Fox News, they don't take into account that the Times is much further from the center than Fox. "To gain a balanced perspective, one would need to spend twice as much time watching Special Report as he or she spends reading The New York Times."
But rather than trying to read statistical tea leaves, we can look at the stories covered by the media. The coverage of the Iraq War is an obvious place to start. One of the constant cries from the right is that news coverage about Iraq is hopelessly biased on disaster and therefore is relentless in it's effort to paint the Bush effort in the worst possible light. Rather than take that assertion at face value, one of the most astute experts on news coverage in Iraq, Arthur Cherenkoff, has tallied a sample of news reports about Iraq. By his count, there are 10,877 stories damaging to Bush, 123 neutral stories, and 407 that could be seen to be in Bush's favor (or 95%:1%:4%). The first line of response normally heard in response to something like this is that the situation in Iraq really is that dire, and even in cases where it's not, bad news sells. A fair enough response, so let us look at a narrower set of Iraq-related news in an effort to at least see if Iraq news itself reflects an undue focus on disaster, without respect to the broader question of whether Iraq is a mess in its entirety.
Some good research on coverage of the Iraqi museum looting after the fall of Baghdad has been done by Jeffery, at Iraqi Bloggers Central. He follows the evolution of the looting story quite closely, and notes, essentially, that the original claims that the entire inventory of 170,000 artifacts had been looted, turned out after saner heads compared notes, that some 2% of a total of 501,000 items have yet to be accounted for. No small difference that. So, while this might seem to confirm the thesis that stories of mayhem, death, and destruction. But before we use these anecdotes as a demonstration that the "drama sells" thesis is a lock, let's take a deeper look.
If we examine reporting on Joesph Wilson and the subsequent coverage of the fact that he lied about his allegations regarding Iraqi attempts to purchase Uranium in Niger. Rather than looking it as the number of stories about Wilson's veracity, we can flip it back around and look at it as the number of stories backing up the allegation that "Bush Lied"â„˘ versus the number of stories exonerating the sitting President from behavior bordering on treason. In either case we get a range of ratios between the two categories of stories Âis from 48:1 to 18:1 (for individual outlets: the total cited was 302:9 or about 33:1) on stories about Bush's perfidy, versus the willing distortion of the truth in an effort to unseat the President.
Here we have two ways to parse this. On one hand we get an a convincing demonstration that bad news sells - 33 stories of Bush scandal to every story clearing things up. But flip that around to look at how the stories were originally sold: 33 stories of good news (or certainly good news to those who were seeking proof of Presidential perfidy) to every counter story. Does this support the notion that perceptions of bias only derive from the natural tendency of the media to focus on disaster? Much of bias is in the eye of the beholder and I leave it to the reader to review some of the specifics. Does the disproportionate interest reflect an interest in selling stories, or tarnishing the president?
It must be noted that the media certainly didn't wear kid gloves when following the Lewinski affair. Perhaps it's just that Presidential scandal attracts the eye more quickly than other sorts of bad behavior. But with respect to Lewinski, some on the right might point out that the interest in marital infidelity, rather than the more serious charge of perjury. However the centrist explanation about the coverage of Lewinski ties in directly to the centrist explanation of why the focus wasn't on perjury: scandals, particularly those related to sex, sell. So, with this, we have nothing particularly conclusive one way or the other. Further, it's certainly difficult to compare metrics on "scandal" since not all scandals are made alike.
One of the blog links that I've been trying to find recently, but have been unable to find was one that I felt gave a reasonable comparison of how the media deals with differing Presidents. Specifically, it addressed the historical mandate debate by comparing when the media felt that the elected (or re-elected) President really had a mandate. Some of the more interesting points made notes that Clinton was granted a mandate by the press in 1992 with his 43% win, while there was some significant noise about Reagan not having a mandate, despite his 49 state win in 1984. I was, however, able to find this bit:
When the Republicans took over control of congress in 1994, Peter Jennings said that this was not really a "mandate for change" ..."It's clear that anger controls the child and not the other way around," stated Mr. Jennings. "The voters had a temper-tantrum...The nation can't be run by an angry 2-year-old." (The actual statement was said on a radio commentary in 1994, and reported again in the South Jersey Courier Post on November 27, 1994.) The Republican victory was also stated by Steve Roberts, a US News and World Report writer on CNBC's Equal Time - "They (voters) are not voting Republican...They are voting against a lot of unhappiness in their own lives."
Funny - the same media was hailing Clinton's 1992 victory as a "mandate for change" (with a whopping 43% of the vote) and that Republican ideas were dead, that the Democrat ideas were what the people wanted. So, when the voters vote Republican they don't know what they are doing, but when they vote for a Democrat they have more credibility?
So, with this and my memory we may have evidence that on a directly measurable basis, that the news media may actually have a bias.
But, if all I have is a set of empirical data, that does not a bias make. Correlation does not imply causation. So, what theory do I have that would tie this bits of observation together in a coherent whole? Well, the idea was first brought to my mind in another blog post about (as I recall) "moment media". Alas, this is a post that I cannot find either, but the gist of it was that journalism was, for decades, a profession built around the objection recording of truth - sort of stenographers of record, if you will. In the post-Watergate era, not only did advocacy journalism start to rise to the fore, but so did "moment journalism." The explanation given is that not only did the media have an obligation to report the news, but should also try to create an emotional "moment" with each news story. In an effort to make the news relevant and real for the news consumer, journalists started looking for a hook, a story, something to rouse hearts and minds - not just a dry recitation of facts. I was able to find a few articles speaking to how journalists view their own responsibilities with respect to creating a narrative:
Some city room diehards may still deem narrative journalismâ€™s concern with voice and theme, protagonists and story arcs, mere staff-draining, news hole-eating incursions onto fictionâ€™s traditional turf.
But Mark Kramer, who brought his annual conference to Harvardâ€™s Nieman Foundation last year, said tough economic times have made editorial management more receptive to some aspects of narrative because they see it as a way to attract and hold readers.
Or, as is noted here:
Hardboiled reporters don't routinely seek to engineer the sequential emotional responses of readers. They don't mess much with their readers at all. Storytellers do. The two roles are in conflict. But the conflict has often been resolved, even by some of those hard-boiled reporters.
So it would seem that this increasing tendency to look for the engaging emotional hook and the resonant story, journalists first look for the sensational. This is reasonable as the sensational story is an easier one to write - it takes a lot more work to make a story about a family returning home after a mediocre vacation than it does to create a gripping narrative about a Peace Corps volunteer lost in a tsunami.
But past that, does narrative focus lead to bias? I would say yes, in those cases when the profession itself has a strong set of political leanings. In that kind of environment when editors and reporters are looking to pick out the point of interest and refine the gripping details that breath life into otherwise "hardboiled" reporting, they have only one reliable metric to work with - whether or not they themselves are interested in the story. Does a given presentation of data generate outrage, move one to tears, or cause a sense of warm satisfaction? Well, that would depend on the people reading the story. And, in the case of journalists, the people making the first call on whether or not a story gets written, and if it does, how it is presented, would be people who will be (at least at the national level) overwhelmingly left-leaning.
But why is this? This, in my estimation is a perverse result of the attempt of media itself to be objective. Without delving into the origins of an objective media in America, one will note that an objective media implies a role for media in the democratic process. Specifically, an objective media takes on the responsibility for educating and drawing the public's attention to those items which are most relevant and important for their democracy. In other words, in this world view, the media provides a role which augments the role of government in a patriarchal system. But why doesn't this show up as right-side social patriarchialism?
It's because the left and right are not mirror image political philosophies. The left has, among its intellectual foundations, the idea of Marxist "false consciousness." This is the catalyst that really makes the concept of narrative journalism really react so very dynamically with the idea of objective media as a necessary component of democracy. A philosophical underpinning of many of those folks who make up the left today is that the people are mislead by those who would take advantage of them, and that it is the responsibility of those on the soapbox to speak truth to power. And the new mission of media is to tell the story compellingly and steer clear of a simple recitation of data that won't engage people in critical discussions about their own fate at the hands of corporations and theocrats.
So does this pan out as a viable theory? Hard to say, but in closing I would note what it is that journalists themselves have to say about the state of their profession. Well, Christopher Hitchens would seem to agree with the assertion that the proclivities of the press made the prospect of a Vietnam-like quagmire and exciting prospect indeed:
There were no more than three Bush-Blair sympathizers in the Kuwait Hilton during the days of the "southern front" in last year's Iraq war, and I know this because I was in that case in the minority. One doesn't have to be an "old hand" to detect the signs of a conscience collective or, if one doesn't care for it, a "herd mentality."
It's now fairly obvious that those who cover Iraq have placed their bets on a fiasco or "quagmire" and that this conclusion shows in the fiber and detail of their writing.
But past that, how about someone a little bit less of a firebrand, how do they see the same things in Iraq coverage? Strangely enough, highly-respected reporter John Burns goes beyond my argument of an unintentional drift towards the tantalizing to something a bit more insidious and more than a bit disheartening. Especially in light of Eason Jordan and his revelations about CNN's behavior in Iraq.
Although I could make the case that this kind of pandering to dictators is simply a manifestation of the effort to destroy a false consciousness through aggressive pursuit of narrative. But really, come on now, at what point do I have to keep making excuses for some of those sorry sons-of-bitches simply in an effort to appear even handed.
And oh, by the way, don't even think about picking up bias in talk radio as a worthwhile counterpoint to bias in the MSM. Or at least until you start showing me how people rely on talk radio for their news.