[N]eurological studies have shown that people's feelings toward a political party dramaticallyThe Political Brain, Drew Weston described a great study he and two colleagues conducted during the 2004 election. Westen and his colleagues studied the brains of 15 self-identified Democrats and 15 self-identified Republicans as they were presented with a series of slides that showed undeniably inconsistent statements by John Kerry. The partisans were asked to consider whether Kerry's two statements were inconsistent, and were then asked to rate the extent to which Kerry's statements were contradictory, from 1 (strongly disagree) to 4 (strongly agree). They then repeated the process with undeniably inconsistent statements by George W. Bush, and again with inconsistent statements by politically neutral males. Here's how Westen described the results:affect the way their brains interpret political news. In his bookThey had no trouble seeing the contradictions for the opposition candidate, rating his inconsistencies close to 4 on the four-point rating scale. For their own candidate, however, ratings averaged closer to 2, indicating minimal contradiction. Democrats responded to Kerry as Republicans responded to Bush. And as predicted, Democrats and Republicans showed no differences in their response to contradictions for the politically neutral figures.This is basically the root of the well-known "confirmation bias."
...
The results showed that when partisans face threatening information, not only are they likely to "reason" to emotionally biased conclusions, but we can trace their neutral footprints as they do it.
When confronted with potentially troubling political information, a network of neurons becomes active that produces distress. ... The brain registers the conflict between the data and desire and begins to search for ways to turn off the spigot of unpleasant emotion. We know that the brain largely succeeded in this effort, as partisans mostly denied that they had perceived any conflict between their candidate's words and deeds.
Not only did the brain manage to shut down distress through faulty reasoning, but it did so quickly -- as best we could tell, usually before subjects even made it to the third slide [which asked them to consider whether the statements were inconsistent]. The neural circuits charged with regulation of emotional states seemed to recruit beliefs that eliminated the distress and conflict partisans had experienced when they confronted unpleasant realities. And this all seemed to happen with little involvement of the circuits normally involved in reasoning.
But the political brain also did something we didn't predict. Once partisans had found a way to reason to false conclusions, not only did neural circuits involved in negative emotions turn off, but circuits involved in positive emotions turned on. The partisan brain didn't seem satisfied in just feeling better. It worked overtime to feel good, activating reward circuits that give partisans a jolt of positive reinforcement for their biased reasoning.
So given that (1) journalists are overwhelmingly Democrats, and (2) party affiliation dramatically affects the way our brains interpret political news, is it really possible that there isn't a liberal media bias?
No empirical study of newspaper stories or talking heads on TV is ever going to be able to objectively determine whether there's a liberal media bias, because what people think constitutes "liberal bias" depends on their party affiliation also. I don't perceive a liberal media bias, but then again, I'm a Democrat, so my brain would presumably interpret political news the same way a biased liberal media would.
But if we know that the inputs are heavily biased, it's very likely that the output is biased as well.
One of the more interesting results of current neuropsychological research is that some scientists think that, at least for hot-button issues, we reason backwards: we decide what we believe based on our emotional needs, and then figure out a reason that we should believe it. Regardless, I think EofC makes an excellent point: based on what we know about journalists and political cognition, it's very unlikely that there isn't substantial bias in both academia and the media. I also think there's no way to develop any direct test of media bias that will satisfy people who want to disbelieve in it.
I forget the technical term, but the human mind develops an interpretation and then looks for facts to fit the interpretation. Given that, it is not surprising that someone of one political bent will ignore or explain facts that doesn't support their political views.
Name names!!! What liberal bias is there in the media(Besides KO and Maddow)? I don't buy your conclusion at all. Do you read The Daily Howler?
Combine this with Andrew's recent post about "The End of Expertise." Could be that, given the fact that experts have to communicate via media, and that the various different media sources select their "experts," people might have a built-in filter against expertise from sources that disagree with them.
I heard that Al Gore said he invented The Internet, and that George Bush, Jr. is a regular guy who likes peanut butter sandwiches. But the Medias is Liberal!!! Yeesh.
I'm hoping that someone will point out that only repiglicans, glibertarians, demoquizlings, and constinatalists would believe that there's any bias out there.
Ah. There's that missing sentence. Wondered about that.
I can't stand the Liberal bias on Fox News, Washington Times, Wall St. Journal, NY Post...it makes me crazy, these damn liberals are everywhere!
A few points:
1. It's possible that a professional who has reached a high position as a journalist at a major network or newspaper is able to overcome their bias much more effectively than the average person.
2. I know it's conventional that most people in the media are Democrats, but has anyone looked at it in a systematic manner? How about just TV - a breakdown of all the major talking heads on CBS, NBC, ABC, FOXNews, CNN, MSNBC - what percentage are Democrats or Republicans?
3. The influence of the corporate owners has to be considered. I don't know what Tom Brokaw's politics were before he worked for NBC, but his views might be shaped by decades of positive and negative reinforcement by those in charge of the network.
4. Liberals are always going to think that the media is too conservative. Conservatives are always going to think that the media is too liberal.
I think it's very hard to define liberal/conservative bias, one of those things if you are on the right you see things being left in the media and vice versa. Take a hot example now, is the media for or against any one of the politicians? If a person is pro-Obama they will remember the stories that criticize or praise McCain and forget the opposite.
And the results of the study aren't really all that surprising if you've read much about how neural pathways develop. I'd imagine that developing a new phobia happens much the same way.
But like I said, not really surprising. What'd be far more interesting and informative is a genuine look at how swing/moderate voter brains light up. What kind of reactions do you see from someone who tends to vote both ways?
I knew I was biased left; my response was to question my leftie-assumptions twice as hard. As a result, most of the local readers I knew assumed I was Republican, with a right bias.
While I knew some journalists who didn't consider the impact of their personal bias on the writing, most did.
Personally, I think bias is more a product of a news organization then a individual; it's based on assignments, editing habits, and headlines as much as the bias of any individual story. Even the act of putting a biased reporter on payroll (Colmes, anyone?) is an editorial-level decision. Remember, every news organization has methods of balancing bias within a given story/issue/hour; each internally measures it's own bias as it strives to be "fair and balanced."
Their argument is rendered irrelevant and incomplete by the obvious fact that the media is clearly not liberal.
Its interesting to me that the neurological roots of partisanship go so deep, but not surprising. I'm genuinely a little more surprised by the people who are "ZOMG!! People use emotion rather than reason for political decisions!!"
Uh. Yeah. Also, water is wet.
As far as media bias.... Meh. First off, I think theres a difference between someone who nods their heads and says "Yeah. That makes sense." and someone who thinks we need to get rid of presidential term limits and make _______ our Emperor-For-Life. Just because they agree with policies doesn't mean they are excited about the guy (Kerry) and just because they are excited about the woman, doesn't mean they agree with her policies (Palin). This is the definition of Charisma. If you don't allow yourself to fall into partisan fervor, you largely won't.
My personal opinion is that we are wired this way by sociobiology. We are built to fall into line behind our chief. Out on the veldt, internal strife weakens the tribe so that other tribes can take it out. The chief is chief and you do what he says. Even in democracy, its really just a civilized form of tribal warfare. Only instead of aligning with a chief, we align with an ideology.
Why there is a keyhole in our brains for ideology and religion is a whole OTHER question.
Personally, I think bias is more a product of a news organization then a individual; it's based on assignments, editing habits, and headlines as much as the bias of any individual story. Even the act of putting a biased reporter on payroll (Colmes, anyone?) is an editorial-level decision. Remember, every news organization has methods of balancing bias within a given story/issue/hour; each internally measures it's own bias as it strives to be "fair and balanced."
Wouldn't it be better to admit to the bias and have a variety of opinions in the newsroom? We could call it diversity.
Seriously though, our legal system works by having advocates of different positions engage in an adversarial process with the intent being that the truth will be uncovered. Why isn't that a good model for news coverage?
I was with this article until this line....
"So given that (1) journalists are overwhelmingly Democrats,"
Really? I mean, I know there was a survey done a couple of years ago that looked at the campaign contributions of journalists. If I recall, it looked at something like 100 journalists, and 90 of them gave to Dems. However, it basically looked at people like a sports columnist for a local paper, or a food critic etc...It didn't really include guys like Brian Williams cause those guys know to cover their rears better...
One only need look at the 2000 election to know which way media bias really flows...
I feel like adding that thier argument is only an argument for one reason why the media may lean left. However, their argument doesn't say that the media, as a whole, leans left. The 'confirmation bias' theory is overwhelmed by the fact that the execs of the media conglomerates pursue a right-wing agenda, and that the media is generally lazy and relies on personality traits (like Bush's swagger) to examine candidates.
I see bias as more a matter of ignorance than politics. If reporters don't know anything about a given subject, their work will tend to show it, and it will also be more easily manipulated by interest groups and interviewees.
I'm hesitant to say the media is in the tank for a particular person or party--Clinton's scandals certainly got reported--but there are certainly issues where cluelessness and groupthink show through pretty well.
To pick on of my personal bugbears, how many reporters know anything meaningful about guns, for instance? How many have friends who hunt? Lack of knowledge comes off as anti-gun bias when reporters are misled by anti-gun sources or their own instinct to make things seem more sensational than they are. Similar problems probably plague most any story where reporters are short on knowledge.
So if most reporters are urban and liberal, and know the things that urban liberals know, then they are more likely to make mistakes on things which rural conservatives know. If those mistakes are interpreted as deliberate falsification, rural conservatives are going to see bias where the problem is ignorance.
McArdle writes:
No doubt there are those who cannot be disabused of their beliefs in media neutrality, whether they're in the media (whatever that is) or not. The more important point is not the perception of bias, but whether bias actually exists. Many academics and journalists may indeed be biased, but the evidence for this claim has mainly amounted to gestures towards voter registration and voting records. All you're allowed to say about that, though, is that many of these people voted one way or the other, which is obviously uninteresting.
And while I agree that Westen et al. (2006) is an interesting read, it's beside the point. The sample consisted of 30 "committed partisans" recruited by
Unless you're willing to claim that all or even most academics and journalists are "committed partisans," the Westen results simply don't apply.
One other, Freud-friendly conjuring trick to notice here: There's a temptation to say that because we can somehow read bias right of the brain scan, denial of bias (particularly strong denial) is in fact a confirmation of it. McArdle comes really close to making this move at the end of the post, but don't buy it. Sometimes a cigar is just a cigar.
I'm hesitant to say the media is in the tank for a particular person or party--Clinton's scandals certainly got reported--but there are certainly issues where cluelessness and groupthink show through pretty well.
Except...
The news media tried not to cover it, and it was the internet that forced their hand.
I also note the way the MSM (didn't) cover John Edward's affair until the National Enquirer forced the issue.
Now compare and contrast that with the coverage that Bristol Palin has received.
These are not all directly analogous, but there seems to be a trend, no?
You seriously are blind to the other side -- that the media outlets are overwhelmingly owned by conservatives. Out of the (almost exclusively liberal) applicant pool, conservative owners will be more likely to hire journalists who at least appear to agree with them.
Rob, that's a good point about bias in regards to urban liberals/rural conservatives. But, it may be somewhat a matter of perception - is everyone in the media really an urban liberal?
I'd love to see an actual comprehensive study, but according to wikipedia, here's where the current network news anchors and the previous generation grew up:
Couric - Virginia
Williams - upstate NY, then NJ
Gibson - Illinois, then DC
Brokaw - South Dakota
Rather - Texas
Jennings - Toronto
Walter Cronkite grew up in Missouri and Texas.
I was with the post until the assertion "journalists are overwhelmingly Democrats". How do we know this? Wouldn't we have to define journalist first?
If most "journalists" were Democrats (by which we mean what? they vote Democratic? they agree with the party platform? they self-identify as Democrats but have voted Republican as many times as Democratic? They hate Republicans?), which is at least unverified and probably unverifiable, then we could draw the conclusion relatively safely that they are in some proportion liberal partisans, which would in turn support the reasoning that "the media" would be biased in favor of liberals... in the sense that a bigger share of journalists would probably be liberal partisans than conservative partisans. Still, we couldn't say it's a much bigger slice of the pie than non-partisans and conservative partisans because we wouldn't know the exact proportions of those groups.
Plus, even if "the media" were completely without bias, the viewership might not be. What if a plurality of news consumers are Republican partisans, this same study would lead us to believe they would read liberal bias into fair coverage, no?
There are too many unknowns to draw Megan's conclusion.
Just look at all the lefty's at the Atlantic!
The determination may be better where columists and reporters got their education. Most universities have a distinct liberal slant in journalism. Business and egineering has a conservative slant. Young adults are often at the stage they are best influenced with their political opinions.
Also associations are very strong. If a conservative is among mostly liberals he/she will mute their opinion and will get swayed.
Most major newspapers have liberal editorial and managment and that does infulence the articles.
Ownership often is very distant. The Washington Times is owned by a Mooney and that bias has never been detectable. GE owns NBC and is connected to MSNBC and the GE bias if known is not present.
So the argument that the corporate ownership influences the newsroom is unlikely. A hands on owner will have a stronger influence like Pinch of the NY times or Ted Turner when he had CNN.
RAH-
You forgot to explain why Phil Donahue was fired from MSNBC.
The neurological/confirmation bias points might help explain the general point "journalists have biases" (human thought processes work certain ways that reinforce biases, journalists are human, ergo...). But it says nothing about the claim that "professional mainstream US journalists consistently show systematic bias in favor of Democratic party policies and against Republican party policies". I'd acknowledge this might never be fully addressed by empirical analysis, but you are ignoring an awful lot of simple logic. What (at any point in time are the Dem/Rep policies that would engeneder these kinds of hugely emotional (logic overriding) responses? There's huge diversity among registered voters of both parties in terms of policy preferences, both general and extreme "hot-button". You can't make concusions about political bias without defining exactly what those biased preferences include/don't include. Why wouldn't journalists' "hot buttons" be much more skewed towards professional issues (whatever storylines are easier to write, or will more easily resonate with audience/bosses, etc) than with personal policy judgements on the economy/immigration/national defence etc?. I think you see journalist bias overriding logic and evidence much, much more in terms of overstating the horserace aspects of elections versus issues, and overstating politics in general versus economic/social issues (favoring stuff that is easy/fun/popular for journalists versus stuff that is complicated and hard to sell). Jumping to the conclusion that Republicans are fundamentally correct that the news media works systematically against their policy objectives seems silly and unjustified. (cross-posted at Economics of Contempt)
A more convincing study would have been to have the subjects objectively relay the information they were given (in other words, paraphrase the two statements from each politician).
If a perceived change in tone, inflection or even meaning came out of the paraphrase, then a "bias" could be inferred.
Even then, I'd like to see actual data for the claim that most journalists are democrats; not because I don't believe it, but because we should be able to trace validity of any basis for an argument.
"I heard that Al Gore said he invented The Internet, and that George Bush, Jr. is a regular guy who likes peanut butter sandwiches. But the Medias is Liberal!!! Yeesh."
I can understand why liberals don’t see the media bias … it’s because they are looking at the wrong thing. The bias is in the subtle; it’s in the nuance. Characterizing a reduction in the growth of a program as a cut; framing a tax cut as “giving money to the rich.” Suggesting that deregulation will “hurt consumers,” or that a minimum wage increase is “welcome relief.” These small characterizations happen all of the time, but if you’re a liberal, you don’t see them, because you really believe that the minimum wage is “welcome relief,” and that deregulation will “hurt consumers.” If you’re a liberal, you think that these characterizations are just facts.
But people in the media are self-conscious … they don’t want to be accused of bias. And this sometimes results in awkward overcorrections. So maybe they overreport Al Gore’s exaggerations. And if they do, it’s because they understand that he has exaggerated … they can see it; it’s tangible to them. Because it’s hard to know how much to report these kind of things, they err on the side of too much.
If you’re Al Gore, this is probably pretty annoying. And if you care about having an informed electorate, this overcorrection isn’t very helpful, because things like minimum wage, deregulation, trade, and tax policy are FAR more important than whether Al Gore is a serial exaggerator.
As soon as you've rendered bias in simple, left vs. right terms, your message is devoid of any meaning. These things are simply too complicated and disjointed to be able to be rendered meaningfully into partisan camps.
Look, watch the convention coverage by CNN and MSNBC again. Again and again and again and again, the liberal wing of the Democratic party is regarded as some freak, fringe ideology; conservatives are naturally assumed to be the base of the Republican party. Moving towards the left makes the Democratic party weaker; moving towards the right means the conservatives are "energizing the base". Being anti-war is being idealist; being pro-war is being pragmatic. The liberals are the weak minority which drags down the Democratic party, the conservatives are the lifeblood of the Republican party. Again and again and again, the cable news networks and the network news organizations portray liberalism as something to be ashamed of and conservatism as the default stance of "real Americans". To say unambiguously that there is a liberal media is just contrary to real life, I"m sorry.
Of course, as long as you say "nothing could convince liberals this is true", you're excluding anyone who might disagree with you from the argument. Which is a neat trick, if your purpose is to win, but it's kind of dirty pool if you're actually interested ni discussion. I could just as easily say that nothing could convince conservatives that there is no liberal media, and you and I would have equal access to the truth. Both are just dueling assertions.
Al Gore didn't exaggerate! He said he was "instrumental to the creation of the Internet". which is absolutely 100% true. He said he discovered Love Canal in the way that you say "I discovered this great coffee shop", which was eminently clear to everyone who was there. And the person who wrote Love Story said that in fact Gore was perfectly honest in the way that he recounted his inspiration for the story. All of those things were absolutely true; people like you still don't know that. Score another one for the "liberal media".
I don't see how your conclusion follows from the premises. I guess you asked a question, but you pretty much came to a conclusion that there must be liberal bias.
I disagree completely. You're comparing random individuals to professional reporters who are made constantly aware of the bias issue.
One can also credibly argue that because there are more liberals in the media, they're desire to appear unbiased actually moves them into a "pro-conservative" bias because they are overcompensating for their own personal feelings.
Zic's comment here might be anecdotal evidence of that.
I'm not arguing that there is a conservative bias. I'm simply saying that this study offers no evidence to support ANY conclusions regarding media bias one way or the other. It only offers evidence of how information is perceived by non media people with certain biases. That's a big difference.
By you logic, the magnitude of the bias would vary by how partisan the reporters in each media organization is. This suggests that conservative media is also biased. I think you have just demonstrated that Fox News is neither fair or balanced.
Freddie,
Actually, I think Al Gore said that he "took the initiative in creating the Internet." He probably should have said that he advocated funding for expansion of networks used by the (already invented) internet, or something like that. But the whole issue was trivial, and as I said, probably over-reported.
-Jeff
Freddie:
While I'm quite liberal, I'm fairly sure the current self-identified ideology numbers are something like 35% conservative, 40% moderate, 15% liberal. I mean, *look* at the Republican party these days. There are very few moderates in it (and almost all of them didn't go to the convention), while there are a ton of moderate Democrats. It's why there's a partisan advantage.
RAH, you have done nothing to disprove the notion that corporate ownership affects the newsroom. The reason I'm not satisfied with your argument is that it is based on specific examples. In Megan's post, she is attempting to argue that there is an inherent, systemic bias that MUST exist. In essence, rather than attempting to demonstrate an actual bias, she's moving one level above the argument.
In speaking of a systemic, neurological basis for this approach, we cannot ignore the influence of ownership across the media spectrum based on some specific examples. The question at hand is whether the media would be more "liberal" if controlled by neutral owners rather than the current right-slanted group. This is difficult to prove without going back to the subjective work of attempting to determine who is biased, and how biased they are. Once you head in that direction, as your rebuttal did, we have essentially destroyed the argument that the bias "must" exist because neurology says so.
Wow, that's some, uh, really airtight reasoning. To take just one of the many holes in that argument, it assumes that the only pressure operating on media output is the personal bias (actually the party affiliation, which is not necessarily the same thing) of individual journalists. I think that's pretty obviously not the case. "News" is a product manufactured for a market; it is shaped by the pressures of consumption as well as production. And the increasing consolidation of the media in a few corporate hands is bound to (has in fact been documented to) put additional pressures on the production end as well.
As soon as you've rendered bias in simple, left vs. right terms, your message is devoid of any meaning. These things are simply too complicated and disjointed to be able to be rendered meaningfully into partisan camps.
That is what it comes down to, in every election. The candidate is always either Republican or Democrat.
Until recently, most major newspapers and television stations were locally owned and operated. Biases were pretty easy to find. For example, one of the dailies in Houston was owned by a family that was active in Democratic politics, including a governor and lt. governor. The other was owned by the local Democratic kingmaker until he died, passing control to a charitable trust. When the chains took over, they left editorial control local, so hiring and firing was based on the already in place biases. I know some former writers and editors of the papers, and would describe their politics as left-of-center.
The NYT and WaPo, though corporate, still bear the stamp of their controlling families in editorial policies. (The WaPo has moved somewhat to the center over the last few years, though.) And certainly the Hearst chain is far from conservative. TV networks still have their independence on the news side from their corporate masters - if they didn't, the staff would quit.
Bottom line is the purpose of a newspaper (and journalism) has been said to be to "comfort the afflicted and afflict the comfortable", a task which leans left.
Oh, and the owners of news outlets have better things to do than interview the reporters and columnists.
These are not all directly analogous, but there seems to be a trend, no?
For those two examples, yes, but then others will come up with other pairs which show the opposite. Confirmation bias is a bitch.
Al Gore didn't exaggerate! He said he was "instrumental to the creation of the Internet". which is absolutely 100% true.
Here's my question about the Gore thing: where was it reported by "neutral" journalists that he said he "invented the internet"? I've see that in conservative punditry and I've seen it in all manner of jokes, but I don't recall the original flap well enough to recall if it was actually reported that way in the NYT or nightly news.
Part of the problem here is that we are conflating different things; talking heads on Crossfire may say one thing, but Robin Meade (tail wags violently) says something different. Just as when discussing what "liberals" or "conservatives" think, it's helpful to discuss specific people and specific quotes, so we know what we're talking about.
is everyone in the media really an urban liberal?
Probably not, but everyone in a position of influence in a major media organization is by definition someone who values something other than a quiet rural existence. That is, the South Dakotans who go on to be major media figures are those who chose to leave their parents' life behind (assuming their parents didn't live in an urban part of South Dakota...)
Well, fundamental error #1 in this is the assumption that Democrats are liberals, or that liberals are Democrats. Eric Alterman points out in his book What Liberal Media that journalists are, for instance, considerably more likely to support free trade or NAFTA than the general population. My point here is that there isn't much that is obviously liberal with either journalists or for that matter the Democratic Party. There is little doubt that those who agitate for things like a 'living wage' or fully nationalized medical care tend to support the Democratic Party, but when was the last time you saw any significant number of Democratic Party leaders come out in support for programs like these?
The second problem with this line of reasoning is a belief that people, including people who make their living collecting and analyzing facts and narratives, cannot or do not reach conclusions that are contrary to what we might expect if we were to look at their political affiliation.
And it is with regard to this last one that you made a conclusion which is demontrably falsifiable, at least with regards to the existence of alleged liberal bias in academia. Most university economics professors are, it is true, registered Democrats. But if you ask them about the effects of a minimum wage, the effects of rent control, or the benefits of free trade, you still get a remarkably broad consensus among economists, regardless of party affiliation (That is: minimum wage increases unemployment, rent control has the long-term effect of decreasing housing stock, and free trade is very, very good).
When it comes to liberal bias in academia proponents of this theoory usuually trot out this or that crazy statement by some professor in some field like Women's Studies or Chicano Studies. But of course the overwhelming majority of both professors and students are in the more traditional fields, and for some strange reason proponents of liberal bias don't look to those for examples of bias. Can someone enlighten me as to what the liberal bias is in evolutionary theory? Or the causes and consequences of the Spanish Civil War? Or the causes of the Cambrian Explosion? What's the 'liberal' perspective on Shakespeare's Romeo and Juliet, and what's the conservative take? Can you show me the liberal bias in the debate over the authorship of the Pauline epsitles in the New Testament? Or the liberal bias in the historiography of the Second World War? With all the registered Democrats in electrical engineering, there's just GOT to be a bias in there somewhere, right? Or the liberal bias in the debates as to whether or not Japanese can be identified as belonging to the Altaic language family?
No- I didn't think so.
THERE WERE A TOTAL OF 30 PEOPLE IN THE STUDY!
30 people!
Why not just have an anecdote about something a friend of a friend told you really happened to someone he knows?
30 people! 15 from each group!
My Lord! Is there any "study" McAardle won't fall for if it supports her pre-existing notion and the libertarian line?
She's not stupid -- well, not when compared to, say, Will Wilkinson (check out his recent comments on Sarah Palin to see a real rube in action) -- but she's incredible gullible and naive as only a libertarian can be.
A study with a fifteen person group (and enough hints of bias in the researchers to set off alarms in any adult) is to science what, oh say, 'The Fountainhead' is to literature. An utter joke. And an absolute embarrassment.
Without passing judgment one may or the other, this is only true if you prove that all or most journalists are democrats. This cannot be taken as a given. Change this assumption to journalists are republicans and the entire conclusion changes.
Here is what you are asking:
P(liberal bias)
But, the answer you give is:
P(liberal bias | journalists are democrats)
These are very different questions!
Eg:
P(rain)
vs.
P(rain|clouds are in the sky)
Obviously the probability of rain is much higher if you take it as a given that there are clouds in the sky!
Look up Baye's Rule in any probability book and you'll see what I'm getting at.
But, to answer this, one should look at all the top shows, and weight the political leanings by viewership, listeners, readership.
For example, Limbaugh should have a much stronger weight than Sam Seder or some other liberal radio host.
My guess is that, once weighted, conservatives would dominate cable news and talk radio. With Murdoch controlling so many papers now (and widely read ones), conservatives may have an advantage there. Network news would be harder, and I don't know who gets more readers in the drudge, malkin, glenn reynolds vs. Kos, Atrios, etc.
I know there have been studies on the political leanings of the Sunday talking head guests, and if I am not mistaken, they lean conservative.
There is also a compensation effect. A journalist who is liberal in private life may be overly friendly to conservatives to give the appearance of being "fair" (the opposite of that, a conservative host being nicer to liberals to fight off any accusations of bias seem less common).
Point being, any anecdotal evidence is meaningless. Compile a list of the most popular shows across all formats, with political leanings, and weight them for popularity. Once that is done, THEN maybe someone can draw a conclusion.
Megan,
How about an experiment on the blog?
Have readers rate your posts in terms of their perceived biases of your posts.
They can also add their political leanings, and we can account for yours, if you state them clearly.
Could a widget be added to your posts for this?
I'd be interested in any other readers perceived biases in Megan's posts.
I think Megan has made it clear she votes for Obama, but with posts like this, "This proves the MSM is liberal" (a very popular conservative meme), this post would probably get an overall rating of a right-bias.
"Bias" can be produced in a number of ways. The one I assume you refer to is journalists selectively filtering information or communicating information in a manner that favors one side over another.
Far more prevalent to me seems to be bias caused by the sheer ignorance and incompetence of journalists in the media - and I am referring particularly to the talking heads on our news programs. The inability or unwillingness to ask tough questions or follow-up questions, or challenge guests who recite unwarranted, unsubstantiated, and clearly fallacious talking points from either side is what produces bias. The lesson to both sides of the political spectrum is to tell ever bigger lies because they have realized that they hardly ever get challenged or called on it. He who tells the bigger lie wins. It is just a pity that Republicans seem so much better at playing this game.
I think that the role of a processing-of-information bias is rather trivial when compared to this effect.
A few thoughts on thi
1) This assumes journalists area allowed to write what they would like to and editors edit as they please without corporate involvement. Not sure what the journos at the WSJ might say about that.
2) Could the media bias be driven (in part) by public consupmtion? If one supposes that individuals seek vindication for their affective response (which the post argues) then would it be reasonable for them to consume more media that "speaks to them". That would be why Fox was so successful in that there was an untapped market. This could also explain why most of the mass print and online MSM may be liberally biased, as stats indicate consumers of those media tend to be more educated (and with increasing education comes more liberal tendencies). I recognize this is a bit of a chicken/egg question, but I tend to believe more in people driving markets than vice-versa
What Niels says.
Most Journalists, methinks, tend to be paid by organizations that are either funded directly by large corporations or indirectly by them (through ad revenue).
Since we know that most corporate leaders are republicans--and that they have influence over their sponsorship--it follows that the MSM actually holds a conservative bias.
At least, such an argument seems about as valid as assuming that since most journalists are democrats, they must write in a way that presents a liberal bias..
"Regardless, I think EofC makes an excellent point: based on what we know about journalists and political cognition, it's very unlikely that there isn't substantial bias in both academia and the media. I also think there's no way to develop any direct test of media bias that will satisfy people who want to disbelieve in it."
I can't prove it, but I know it must exist. It works on a small scale in the tightly controlled, reductive environs of the laboratory, so it must be true in the chaotic, complex environment of the wider world. No place for professionalization, self-restraint, careerism, over-exposure, etc.
Next assignment, prove it. We biased folks in the academy have a certain standard: start with a theory, then gather evidence and see if the latter supports the former. All you've got here is a theory about journalists based on a small laboratory study of non-journalist volunteers in a lab study.
As an example of how your theory might go wrong, take the work of Josh Correll, a pyschology professor at the University of Chicago. He designed a study wherein subjects were shown an image depicting either a bystander or an assailant, and then would decide whether to fire or holster a weapon. Data gathered included whether the subject fired/holstered appropriately. Patterns emerged whereby both white and black subjects showed a bias against blacks in terms of the mistaken shootings (i.e., they all tended to be more likely to mistake a black bystander for a black assailant). However, this tendency largely disappeared when the subjects were trained police officers.
The point here is not to show that the police don't treat blacks unjustly, but to demonstrate that training and experience can serve to distinguish professionals from the general population. Thus, journalists and academics might very well buck the trend demonstrated by the participants in the study you've cited.
Now, we have two lab experiments that argue for different conclusions with respect to the question. One says that people are biased in their information processing. The other says that professionalization can help overcome biases. Empirical evidence of journalistic output could help answer the question, but EofC waves off on that prospect. Better just to keep the unproven biases firmly in place.
Really? I mean, I know there was a survey done a couple of years ago that looked at the campaign contributions of journalists. If I recall, it looked at something like 100 journalists, and 90 of them gave to Dems. However, it basically looked at people like a sports columnist for a local paper, or a food critic etc...
I can see all those neurons in the "contradiction dismissal" part of all the "what liberal media?" brains firing right now ... give 'em a moment ... Ah, that feels goooood.
Megan, thank you for addressing my question from a couple posts back! This study is interesting and certainly relevant to the existence and direction of media bias. At the same time, I can't help agreeing with the emerging consensus in comments that saying "committed partisans interpret events to the advantage of their candidate, therefore the media has a liberal bias" begs at least three questions:
A) Does what applies to partisans also apply to trained journalists?
B) If so, do journalists' skewed perceptions of reality actually translate to a skewed presentation of the news?
C) If so, is that presentation systematically liberal? (That is, are most trained journalists really Democrats?) And if so, what accounts for clear instances of conservative bias in the media?
I'm afraid this may be a deeper topic than a single study of ordinary people (not trained journalists) can really penetrate. :( Nevertheless, thank you for the insight!
pickabone said "All you've got here is a theory about journalists based on a small laboratory study of non-journalist volunteers in a lab study."
Pickabone, don't give this that much credit. It's just a hypothesis, not a theory. ;)
I just re-read your comments at the bottom of the posts, Meghan. "Regardless, I think EofC makes an excellent point: based on what we know about journalists and political cognition, it's very unlikely that there isn't substantial bias in both academia and the media. I also think there's no way to develop any direct test of media bias that will satisfy people who want to disbelieve in it."
Uh, wha? I alluded to this in my first post, but wow. You really made a leap. I didn't realize how far.
"it's very unlikely that there isn't media bias."
You're going to need a heck of a lot more info than this survey to make such a claim.
The responsible thing to say would be "based on what we know about journalists and political cognition, there may be a significant risk of liberal bias entering into journalism if not managed appropriately." I'm sure you have some level of access to research what editors due to combat bias, so you could look into this.
I'll let it slide because blogging is a little more off the cuff, but I think a correction is in order.
How much "training" to journalists really get? I mean that seriously. Cops spend a bunch of time in a quasi-military academy, and do shoot/no-shoot drills before they're ever issued a badge. Then they get to play shoot/no-shoot pretty much every day, with their lives and freedom at stake if they screw up.
Do journalists have to sit around spotting bias accurately before they get their MSM decoder rings? Are their serious, immediate consequences to failing to adequately remove bias?
Color me skeptical that journalism is really a trained profession like, say, medicine, or that journalist's motivation to be fair comes from anything other than a personal sense of fairness.
Obama's campaign has live streaming video of him in Michigan right now. Check it out undecided voters, it is so nice to listen to someone speak without the interruptions, the framing of bizarre hypothetical questions that don't have a reasonable answer. I am off to check out if McCain's site has the same thing. I am sure Obama will be on live streaming the whole election.
Decideds, if your Dems, scrape up your more aloof friends and get em to the polling station.
noseeum said "don't give this that much credit. It's just a hypothesis, not a theory"
Agreed, to a point. In my field a theory is a general statement of informed belief. A hypothesis is a precise and testable statement that operationalizes (ugh, what an awful word, but I can't fight city hall) the theory.
In other words,
"the media is biased to the left" is a theory
"the media give more air-time to liberal advocates and Democratic office-holders" is an hypothesis.
I realize that in other kinds of science, you start with hypothesis, test them with data, and then state the confirmed set of hypotheses as a theory. In mine, you start with a theory, decompose it into hypotheses, and these them. Then you have a proven (ie tested) theory.
Oh, and I completely agree with your criticism of the ridiculous formation Meghan uses to interpret the study: "it's unlikely that there isn't" is simply an untestable version of "it's likely that there is."
Now compare and contrast that with the coverage that Bristol Palin has received.
The McCain campaign issued a press release, for goodness sake. You're comparing a decision to report on shady rumors with the decision to report on a formal announcement by a campaign?
al feingold:
I'd love for someone to try and make the case for me that "Tweety" Matthews is liberal. He's admitted, on air, voting for Dumbya twice. He's talked about the thrill up his leg when Fred "Foghorn Leghorn" Thompson and Rudy "Noun, verb & 9/11" Giuliani are in his presence.
Jim Glass:
Do you have names that go with those numbers? Do any of those journalists work at the major papers or TV outlets? They all could work at small town papers, or could all write for Rolling Stone.
It would seem obvious that the question that should be asked is:
Why are most reporters Democrats? Or, to be a little more selective, why are most reporters liberals? (Assuming that they are liberals, and not moderates.)
I can think up multiple possible reasons, from the fact that more reports work in cities vs. suburbs or small towns, to selection bias in hiring, to social pressure to conform once hired. But it might be enlightening to find out what the actual reason is. Especially since, without knowing that, all anyone who finds that unfortunate can do is complain without effect. Gotta understand the cause before you can come up with a cure.
Rob Lyman,
That's a good question for the development of the theory.
I suspect that there is a much higher variance in the training of journalists than in the training of police officers. Still, I tend to believe that journalists spend a lot of time gathering experience before given any real clout. A front-page political journalist may have started out as a reporter on a high school or college paper, spent four years as an undergrad and possibly and extra year or two in grad school studying the trade, spent a few years at a copy-editing job or some lower-profile desk, etc. All the while they're socialized into the norms and practices of the trade. It's a long road to travel before they're given a front-page political assignment, where they finally get to wield the journalistic equivalent of a fire-arm. Cops, on the other hand, go through a highly regimented but brief training program, and are promptly given a patrol assignment and a 9mm side-arm, which many (if not most) of them never fire in the course of their careers.
That said, this all simply informs our theory-building. Without empirical evidence about the biased output of journalists, it's just an informed but untested set of beliefs and assumptions.
Whoa! That was kind of jarring. Was Josef Mengele the senior author? Must be an old study; they closed his lab down in the forties. Might have been hard to concentrate on those questions if you suspected they were going to take your brain. I think I'd have stayed 'Independent' myself; Althouse's 'cruel neutrality' is looking better.
It's usually described as brain MRI or PET scans of the brains. I think Andrew Sullivan requires explanation as do all of the commenters on this blog who introduced the Bush=Nazi meme. Rob, it was Freud's brother who had the trains running on time in Austria (before Hitler); wonder what happened to him. This study would get you close to a hypothesis (re: Sullivan and the commenters). Similarly, if you see the water boarding of a few senior Al Quaida as equivalent to Hitler, then Obama being a savior, bringing light to the darkness of America (I'm such a nice guy. I left out the "k" out of it) is a pretty appropriate euphoria.
A much better way of getting an answer to this question would be to look at information that might actually relate to the area of interest.
http://www.cmpa.com/Studies/Election08/election%20news%207_29_08.htm
If you do that, you will find that there is a bias
in the media. AGAINST liberals.
Journos may be democrats but that tells us nothing about their editors or about how much the right likes to work the refs.
I assume you will now update your opinion and not suffer from confirmation bias of any kind.
All the while they're socialized into the norms and practices of the trade.
There's a key point right there; socialization like that will probably tend to reinforce the biases of the majority and suppress those of the minority; that's precisely what makes the dominance of D's in the newsroom troubling. Even if most cops are racist, if their training teaches them to, say, watch the suspect's hands, they it can tend to overcome their bias. But if their "training" consisted of listening to what the old-timers said about people of various skin colors, I doubt they'd do very well on the racial shoot/no-shoot.
And--to go with my ignorance-not-bias theory--socialization and apprenticeship will do nothing to overcome ignorance, because the old-timers can't pass along knowledge they don't have. They're likely to pass along the misinformation and cognitive mistakes they've been making their whole careers.
But I don't see much empirical evidence coming to light here, because it's pretty close to impossible to construct objective standards for journalistic bias.
"socialization like that will probably tend to reinforce the biases of the majority and suppress those of the minority; that's precisely what makes the dominance of D's in the newsroom troubling."
You're still assuming that ideology trumps professional practice, which is what you're trying to prove.
As Correll has shown, cops do perform better in the racial shoot/holster experiments. Does this prove that police departments don't have a history of racism? Hardly. It doesn't even prove that contemporary police departments are better at discerning the intentions of people on the street in practice. All it shows is that professionalization alters the psychological tendencies of police subjects in a lab.
Journalists have professional norms that require them to back up their reporting with factual evidence. Thus, they have a similar mechanism for overcoming the biases of information processing discovered in the lab experiment Meghan provides.
Again, all I've done is argue that professionalization can work to overcome psychological tendencies in controlled environments. Journalists have professionalization practices. Thus, they might be able to overcome their biases.
"But I don't see much empirical evidence coming to light here, because it's pretty close to impossible to construct objective standards for journalistic bias."
That's a cop-out. I have two off-the-top-of-the-head suggestions, starting with a lab experiment.
- Re-run the above mentioned experiment with 15 Democratic political journalists and 15 Republican political journalists. I hypothesize that the journalists would be better at indicating co-partisan contradictions than other subjects.
That would give us some insight into whether journalists were as affected by biased information processing as other subjects. But we're still in the lab. Let's move on to the real world.
- Analyze reporting to see whether there is a partisan difference with respect to whether partisan politicians are the primary or exclusive source of the stories. Do reporters simply quote politicians and their surrogates, or do they provide relevant factual information to support or contradict the politicians' positions? Is there readily available contradictory or supporting information that the reporter ignores? And so on...
It's not necessarily an easy task, but it's far from impossible.
Usually, when someone says it is "obvious ... common sense" that something "almost has to exist," it means they don't really have much evidence. Take an empirical attitude toward the actual output of the media and you will find a very complex picture that defies the "liberal bias" article of faith. The conservative case for bias seems to depend on intense reading between the lines, dissecting every gesture for hidden motives--a level of hermeneutic suspicion that would make Derrida blush. By contrast, straightforward empirical studies have shown that conservative commentators get a good deal more air time than liberal ones, and it is no secret that major media outlets (particularly broadcast) have been courting the large and, let's say, easily excitable conservative audiences unearthed by Fox News.
I recently met a guy who works for the NY Post. He's your typical young, urban, quasi-bohemian type. Very left-leaning. According to him, a good 80% or more of the staff at the NY Post are clearly liberal/progressive. According to the pop-psychology theory embraced by Megan et al., it should be riddled with liberal bias. Yet anyone who has gotten within 10 feet of the Post and its 80-pt screaming headlines knows it is a hysterically right-leaning rag.
The puzzle is solved once you realize that journalists (whoever they are) are first and foremost doing a job, and that job (like any job) mostly consists of doing what your bosses want you to do. They would quickly be shown the door if they did otherwise (as some do). And the "bosses" of media by and large want one thing--to get more eyeballs on their pages/broadcasts/web site. The idea that the media agenda is set by the ideological biases of a bunch of rank and file stringers or teleprompter readers is wildly detached from reality. It is a bit like saying that GM must be a "liberal car company" because most of the line workers vote democratic.
Has anyone read "The Intellectuals and Socialism" by Hayek?
In my view, it is the definitive answer to the question.
So you found a study that proves confirmation bias and a study that "journalists" donate more to "Democrats" and you've proved that the media is liberal?
What if you redid that study, but every time the subject displayed confirmation bias in the direction of a Democrat, you gave them an electric shock? And when they gave it in favor of a Republican, you gave them a cookie?
Think you'd see slightly different results? That encapsulation sums of the actual environment worked in by actual media members.
Do you know, Megan, that there is an entire universe of psychological function out there beyond confirmation bias? What if... all those other psychological forces ... ***were to somehow intersect with journalists????*** You might see something *entirely different*!!!
The closest thing I've seen to an objective analysis of media bias is this commentary by Geoffrey Nunberg. It's admittedly very narrowly focused.
You only have to read the coverage of Palin to see that this "study" and McArdle's comments are all hooey.
Again, all I've done is argue that professionalization can work to overcome psychological tendencies in controlled environments. Journalists have professionalization practices. Thus, they might be able to overcome their biases.
My point is that this "professionalization" lacks the gatekeeping and disciplinary practices of most of the things we call "professions," such as law or medicine, and the penalty for deviation from norms may in fact be increased readership and popularity instead of, disbarrment or (in the case of extreme police behavior) imprisonment. If Ann Coulter did the medical equivalent of what she does "journalistically," can you imagine the reaction?
So, theoretically speaking, I doubt the validity of the comparison between journalism and other professions.
That said, I like your first idea. The second one I find more dubious as a measure of bias because the validity of contradictory/supporting evidence is subject to a large dose of subjectivity, and including invalid evidence could be just as great a sources of bias as neglicting to include valid evidence.
Professionalism is leaving the TV/Cable channels. That was so evident with the squabbling on MSNBC.
When Walter C. was an anchor it was almost undetectable his personal opinion. It was only it what they choose to report and whether it was factual or on an emotional basis. It was not until he retired did he reveal his own slant.
Not so today.
Kondrake was a definite liberal when he started and has been drifting right in the last 20 years. Barns has always been a conservative as well as Kristol. But the analysis is often more even in that they criticize both sides.
PBS is liberal but more evenhanded in their commentary.
Fox is conservative but will book opposing viewpoint guest for entertainment. The rest is tabloid sensationalism.
Megan is a liberal with a more open mind that likes to try to see the other side. She is young and will probably get more conservative as she ages. Often the illogic slowly changes their viewpoints.
Maddow is a liberal but has a more honest analysis than the others.
Matthews reacts totally emotionally.
Most political shows are opinion commentary and flare-ups create interest and ratings.
This has led to obvious bias slants in commentary and has bled into reporting.
Self interest trumps these unconscious cognitive responses. The experiment is meaningless in relation to those having or seeking employment in the national media, and the way the speak and write. While nobody but the experimenter would judge the subjects responses and quite probably the subjects didn't even consider how their responses would be judged the situation would be much different for someone whose livelyhood depended upon how their response was perceived by others.
Come back to us with a study of journalists so we are not comparing apples and oranges.
My point is that this "professionalization" lacks the gatekeeping and disciplinary practices of most of the things we call "professions," such as law or medicine, and the penalty for deviation from norms may in fact be increased readership and popularity instead of, disbarment or (in the case of extreme police behavior) imprisonment.
I agree that the rigors of medical and legal professionalism (and possibly that of law enforcement) exceed that of journalism. But we're not trying to determine whether journalism rivals the professionalism of medicine or law. I made a broad analogy between the experience of journalists and police officers, but the purpose was not to measure precisely the relative professionalization of the two. Such measures are probably incommensurable. The point was to recognize the potential for social and technical learning to train individuals to overcome their prior biases. The real question is not whether journalism rivals the professionalism of other vocations, but whether the professionalism of journalists is sufficient to overcome these biases.
the validity of contradictory/supporting evidence is subject to a large dose of subjectivity
...as is the evidence presented in the study by EofC that Megan used to spark this discussion. We have "undeniably inconsistent statements" as the experimental treatment. And we have (presumably) fMRI scans as an observed response. Don't be over-awed by the objectivity of fMRIs; they're a fairly subjective technology when used in the social sciences. The scan patterns they identify are based on reported or induced emotional states (or, in some cases, clinically diagnosed pathologies), which have the nasty habit of showing similar patterns for states we normally consider antithetical to one another. Anxiety matches arousal. Love matches hatred. Etc.
including invalid evidence could be just as great a sources of bias as neglicting to include valid evidence
Great idea. Let's add that to our data collection plan.
It appears that noone in this thread has provided actual examples of the media being biased in favor of liberals. Would anyone care to do so? The closest I've seen is the John Edwards vs Bristol Palin example, but that's flawed because the media reported fairly extensively on the Edwards story once it became public. With Bristol, the McCain campaign announced the pregnancy, after which it was reported on. Claiming that the MSM is biased because they didn't break the Edwards story while they did report on the Palin story once it did break is comparing apples to oranges.
I'm not claiming that the media is perfect and/or completely unbiased/impartial. (In fact, I have several problems with their coverage that are more structural than related to bias; for instance, it annoys me greatly that they seem to prize evenhandedness over objectivity.) But I would be interested to see the most egregious examples of liberal bias that people can list.
Your first line hooked me:
But, eventually, the let down:
Now, if you can come up with the explanation for the given, you'd really be on to something interesting.
"I also think there's no way to develop any direct test of media bias that will satisfy people who want to disbelieve in it." -MM
Maybe so, but there positively is absolutely no conceivable test that could disprove media bias to someone who is determined under any circumstances to see it and perhaps complain of it in search of redress.
(Note also: cThat someone is heard on air complaining of bias does not prove that the complainer honestly sees it. People in the political communication field do occasionally employ strategery.)
I'm reminded of psychological studies that demonstrate how prefacing an argument with irrelevant neuroscience verbage makes people more receptive to patantly false arguments.
"I was with the post until the assertion "journalists are overwhelmingly Democrats". How do we know this? Wouldn't we have to define journalist first?"
Pew Research Center's studies of journalists shows them as consistently less conservative and more liberal than the public. Although this is less true of local news.
http://people-press.org/report/?pageid=829
"Most national and local journalists, as well as a plurality of Americans (41%), describe themselves as political moderates. But news people especially national journalists are more liberal, and far less conservative, than the general public." (That's from 2004)
However it does not say partisanship. Also what is defined as "conservative" or "liberal" is open to interpretation. If you think Obama is a bit more conservative than what you want than journalists probably are conservative compared to you. I might even say that if you think Hillary Clinton is "Right-wing" than the media is conservative from your perspective.
As for the "positive" image the media gave of Bush this is again depending on perception.
"It's a good thing Al Gore has an unappealing demeanor, or George W. Bush would be in real trouble." Tucker Carlson in 2000 (granted not liberal, but this is part of my point about them not going as googly over him as the Left thinks)
"he could not answer about the names of foreign leaders. The doubts about Bush since then have been a low-grade fever on his leadership ability, pushed along by the wisecracks of late-night TV comedians and by Democrats' claims that he has risky ideas or doesn't know much history or geography...Much of the examination of Bush has been bouncing between self-assuredness and arrogance. " USA Today 2000
So "not unappealing, but stupid and a walking smirk" was around in 2000 too.
"It appears that noone in this thread has provided actual examples of the media being biased in favor of liberals. Would anyone care to do so?"
Palin to Edwards 2004. In 2004 he was basically a new Senator who made good speeches and was considered "pretty." Was he treated as negatively as Palin?
That's not ideal as you can say "but he ran for President so they'd already covered him."
Okay then. Barbara Lee vs Tom Tancredo. Both of them are equally on the far-end of their party and came in 1998. Even before Tancredo ran for Prez who do you think was criticized more?
Simple question. What were the statements used. There are degrees of inconsistency obviously otherwise it wouldn't be worth asking them to rate them on a scale of 1-5 instead of t/f. It might well have been possible for the statements to be asymmetrically inconsistent.
Anon Y. Mous says, "Now, if you can come up with the explanation for [why journalists are overwhelmingly Democrats], you'd really be on to something interesting."
How about this observation: the profession of journalism has very high status in a few small US social classes. Loyalty to the Democratic party is also a very important class marker to these same classes. Hiring in journalism reflects (like most other industries) significant class bias. Hence you end up with an industry dominated by a single (or at least a very few) social classes.
Class bias is not unique to journalism. The oil industry, chemical engineering industry, building and construction trades all exhibit the same characteristic. But it is more of a problem when your product is "the facts."
I'm surprised no one has yet mentioned the Groseclose study, from 2004, which found a strong liberal bias throughout our national media. Their conclusion, from their summary paragraph:
Our results show a strong liberal bias. All of the news outlets except Fox News’ Special Report and the Washington Times received a score to the left of the average member of Congress. Consistent with many conservative critics, CBS Evening News and the New York Times received a score far left of center. Outlets such as the Washington Post, USA Today, NPR’s Morning Edition, NBC’s Nightly News and ABC’s World News Tonight were moderately left. The most centrist outlets (but still left-leaning) by our measure were the Newshour with Jim Lehrer, CNN’s NewsNight with Aaron Brown, and ABC’s Good Morning America. Fox News’ Special Report, while right of center, was closer to the center than any of the three major networks’ evening news broadcasts. All of our findings refer strictly to the news stories of the outlets. That is, we omitted editorials, book reviews, and letters to the editor from our sample.
It was far from a close call.
You can read the study here:http://www.sscnet.ucla.edu/polisci/faculty/groseclose/Media.Bias.8.htm
The results, which look at news content alone, regardless of the personal political views of reporters, confirm what the Pew studies suggest. Pew found that reporters overwhelmingly self-report as Democrats/liberals, with conservatives FAR underrepresented. These, in turn, confirm the findings of Lichter, Lichter and Rothman (1986, cf. their book "The Media Elite), and again by Jim Kuypers (2002).
And Groseclose demonstrates that these biases do indeed pass through the editorial process into the reporting. This is significant because it weighs in against the idea that reporters are able to correct for their own personal biases, absent a more politically and culturally diverse newsroom.
I think the bias is more in some areas than others. Journalists want to be safe and they work for corporations, so I think they're a bit more conservative on business or crime issues. Business papers are generally more conservative.
I think journalists political biases might be similar to the average for those who graduated from elite "Blue State" Universities.
It seems likely Megan had made her mind up before she read the "study", which simply "confirmed" HER biases.
It's really all just circular, flawed logic, yes?
"I have confirmed media bias" - and no, my own bias plays no role whatsoever in this conclusion.
Others are biased, but, not me!
Uh-huh.
First, the lefty bias in major news media is part of our social fabric to such an extent that most Americans are convinced it's true. And, since they believe it's true, they discount the lefty babble. That's the beauty of an informed readership -- they don't buy all the horsehockey put in print. Score one point for democracy: not all the people can be fooled all the time.
Second, in this small sample study, the conclusion is that political decisions are made on an emotional basis, and facts then ordered to support the pre-ordained conclusion. If that's true, and it rings true to me, why all this pretense at deliberate political discourse? Take away one point for democracy: it's the feelings, stupid. That's what wins campaigns, and all people can be fooled all the time if you just stroke them right. Don't reason with them, just stroke them, kindly.
And if politics is all about emotion, so are probably all our decisions. I feel, therefore I think I think, and I feel I am.
The study you are reporting on is just a specfic example of a more fundamental problem in information science-information theory which can be paraphrased in laymans terms the following way: "in order to make information useful, we make it wrong"... The idea was at first so bizarre to me when I first learned about it years ago, I thought I had mis read it. But it is true, and it is a FUNDAMENTAL problem of information
All this particular study did was just illustrate how evolution solved this problem in the human hominid brain. Computer scientists, AI scientist and animal behavioral specialist have shown other ways this can be solved.
Bias is fundamental to any intelligence
I looked at the Groseclose study, and the research design is pretty wretched. They compute ADA scores (a commonly used but outdated measurement of ideology, most political scientists use DW-NOMINATE) by comparing citations of think-tanks by news outlets with citations by politicians. The idea is if a politician and news outlet cites the same think-tank, they have an ideological similarity. Over the course of many citations, they get an indexed score for the news outlet based on a known score for the politician.
The first problem is that they assume that think-tank citations are superior to politician citations. If the first makes sense, so does the second, and ADA scores can be computed just as easily. The authors should be looking as well at how much media ink and air-time the outlets give to politicians to craft their own messages. Additionally, it's naive for the authors to assume that citing a source indicates support for the message. No journalist worth her salt would fail to recognize that citing the ACLU would likely undermine an argument among a certain segment of her audience.
The second problem has to do with the profusion of think-tank and policy organizations on the conservative side of the spectrum. It's pretty simple: if there are two organizations doing similar work, it's less likely that the study will pick up a "hit" when comparing news outlets to politicians. If the politician mentions Americans for Tax Reform and the news outlet mentions National Taxpayers Union or Citizens Against Government Waste, they're probably conveying a similar message but it fails to register as such. Basically, it attenuates the relationships on which ADA calculation depends. If those relationships are attenuated more often for conservatives, it appears that the media are biased toward liberals.
The last problem is in the plausibility of their rankings of news outlets. They list the Wall Street Journal as the most liberal, nearly as far to the left as Ted Kennedy. This result looks even worse when you consider, as you point out, "that these biases do indeed pass through the editorial process into the reporting." The WSJ Editorial page is among the most conservative of all major outlets. Also, they list Drudge as substantially left of center. It's laughable. Now, I know we're supposed to let our empirical results speak for themselves, but this is a constructed index, which makes it especially vulnerable to conceptualization and design flaws. When you're creating an index, you have to keep an eye on plausibility, especially when we have other ways of determining the value you're looking for. In this case, we can just ask Drudge: is his "Report" more liberal than the average voter? I can't imagine that he'd be shy about admitting his conservative biases and his intention to influence political coverage in a conservative direction. In short, the authors' instrument came up a bust.
You could argue that if the media is made of of mostly liberals, and they're aware of a natural bias, they overcompensate by being tougher on liberals. One can think of many situations in daily life where this holds true. Any kid who's been coached by a parent could assert that they were treated much harder than the other kids on the team. In an effort to be fair, the media may in fact wind up being too tough on liberals, and/or easy on conservatives.
Onr of the problems with the cause-and-effect relationship between brain function and bias the study posits is that if it were accurate, we would expect to see a similar type of media bias operating in all times and places where there were media similar to our own. For instance, did mainstream media bias exist in American media from, say, the end of World War II to 1968 or so? Journalists presumably had opinions then and human neurobiology hasn't changed. Yet I have never heard claims by the right that the media in this period was biased against them. Near as I can tell, according to the right, media bias took off some time around the time Sergeant Pepper's came out and journalists lost the sense of noblesse oblige elite opinion-makers should have, gave in to their base libertine urges, consumed themselves with neurotic guilt over their putative complicity in America's racist history and took it upon themselves to foist these feelings on an uncooperative public, a project they pursue to this day. If media bias is hardwired, it doesn't come from social causes, which is the drum the right has been beating for a damn long time.
It's only within the last couple of years that the American right has begun shreiking at the BBC for U.S.-style liberal bias. Before that, presumably, it was a paragon of soberness and objectivity. If the study is right, how could that be? If media bias is a consequence of brain anatomy, where was it all those decades and centuries before Andy Warhol got famous?
Pickabone, the Wall Street Journal's news page is totally separate from the editorial page, and known for being liberal in the business.
Your other argument doesn't make much sense. The point of the study is to determine what information reporters are letting through. That's where the bias is going to operate 99% of the time--not because they write outrageous things about conservatives, but because they decide that liberal data sources are more intrinsically interesting/reliable than conservative ones. Its certainly not perfect, but it's consistent with other known facts, and there's nothing obviously wrong with the methodology. Whatever the problems with ADA scores, they're not wildly off--they don't list Barbara Boxer as a strong conservative. The Groseclose results are large enough that it's hard to see how a coding change would have significantly altered them.
There's an assumption underlying this post's conclusion:
That the left and the right have equally accurate views of reality.
Consistently inaccurate reporting would be actual "bias."
Reporting accurate views that happen to be more liberal would be...accurate reporting.
Is it possible that reality skews liberal?
Since even Fox is now reporting that McCain and Palin are repeatedly lying about the candidates' tax plans (to give only one example), it's not a crazy idea.
http://bourbonroom.blogs.foxnews.com/2008/08/31/does-john-mccain-have-a-tax-problem-answer-probably/
Megan
I was responding, in part to this from Jason Van Steenwyk:
these biases do indeed pass through the editorial process into the reporting
but I guess I read that a little too quickly. I read it as "pass from the editorial page into the reporting." the points stands just the same. Do you imagine the WSJ (reporting side) to be the most liberal major news outlet in the US?
The point of the study is to determine what information reporters are letting through.
You're mistaken. I quote from the abstract:
The point is to locate the ideology of news outlets in a common space with politicians. The method is to see which information sources are being let through. That method is flawed, due to the reasons I discussed above.
As for ADA scores: when applied to elected legislators, it works fine (though the DW-NOMINATE scores seem to work better). Voting with another legislator puts you closer to her on the scale. It works because it uses roll-call votes, which are relatively unambiguous dichotomous choices (not counting "present" and "not voting" of course). News reporting is monumentally more complex in its construction. In short, the think-tank citation is a poor bridging mechanism between journalists and legislators.
Moreover, the authors neglect to mention how much these source citations contribute to the basis of the stories and to the overall reporting of the issues. Do reporters cite think-tanks more frequently than politicians? If not, what is the pattern of citing politicians? Where in the story do they place these think-tank citations? You can put one NRA citation in the first paragraph and 10 ACLU citations after the jump. Now which information is getting through more prominently? Are these front-page stories or buried in back? Economists do a lot of things well, but textual analysis tends not to be one of them.
In the end, it's the plausibility of the results that make the study irredeemable. Specifically, any instrument that finds the Drudge Report to be liberal has fatal flaws.
Megan:
Regarding the Groseclose study, and the title of this post, can it be possible that you didn't even know about the extensive discussion of that study at...Media Matters?
http://mediamatters.org/items/200512220003
Pretty damning. Well and extensively argued and evidenced. After reading it, it would require some damned impressive refutation for a reasonable (read: unbiased) soul to give the Groseclose study any credence.
Pickabone, I'm surprised you didn't find this.
None of which has anything to do with my previous post, on reality's unfortunate tilt to the left.
Steve,
Hat-tip to you. I wasn't really looking. I was just getting annoyed with people copping out and saying, "it's too hard to prove, but we just know it's true." Then this Groseclose study came along, I gave it a good read, and I had problems with it. Only so much I can do over a morning cup of coffee and an early dinner.
BTW, here's the nail-in-the-coffin quote from the Groseclose study:
You can pretty much guarantee that when authors claim to present an exhaustive list of prior research, they're probably hiding something they don't want you to see.
"For instance, did mainstream media bias exist in American media from, say, the end of World War II to 1968 or so?" Nick
TR: I have no survey on the self-described political ideology of journalists from back then so I don't know. I do think the composition could've changed in the intervening decades due to changes in academic structure and regional cultures. Many states that are now "liberal" were moderate Republican in that period. For example in the following states Nixon beat JFK by 5% or more and Dewey defeated Truman by 3% or more.
Maine
New Hampshire
Oregon
Vermont
http://uselectionatlas.org/RESULTS/
From 1948 to 1956 New Jersey and New York voted Republican in the Presidential election.
Do I think the media from the 1940s to the early 1960s preferred the Northeast and Pacific Coast? Yeah probably. Although at that time it mattered less on many of these "culture" issues as abortion was illegal everywhere, school prayer was common, and even gay men's magazines were against same-sex marriage. There were some "culture wars" issues then (Anti-Communism, birth control, miscegenation, rock-and-roll) but on those it seems to me the media of that time probably was to the Left of the people if only moderately so. (Granted the Left then was largely right, but it was not necessarily Right about Communism)
"Granted the Left then was largely rigght"
I should've added it was totally right if it rejected racism more than the population. I'm not entirely certain it did, on reflection, but if so good for them.
Fascinating to see how people's biases show up in this article about bias. Just curious: Of those who think the media is biased left, how many of you voted for Bush 2004? Of those who disagree, how many of you voted for Kerry?
Final definitive study on the subject: When Hugh Hewitt has a journalist on, he always asks him about his politics. The conservative ones always tell him, the liberal ones always deny that they have any; I don't recall an exception. Then he asks them the percentage of liberal-leaning journalists in their newsrooms. Whoever they are, they always say a number like 90%.
So it's fun for us on the outside to argue about our perceptions of whether there's a big liberal majority. But, as Megan said, everyone in the business seems to take it for granted.
"Fascinating to see how people's biases show up in this article about bias." MikeR
TR: I think the article means all people have a bias not just journalists.
"Of those who think the media is biased left, how many of you voted for Bush 2004?" MR
TR: I did not vote for Bush in 2004. I considered voting for Obama if he won the Democratic nomination and Giuliani had won the Republican. (Obama won his side, but Giuliani didn't win his so I'm voting Republican)
Here's a test:
Google "Arkansas Blood Scandal" and see who reported on it, and observe who did not.