Bryan Caplan  

At First Glance: Bias in the Media

PRINT
Skewness in Earnings... The Disease it Purports to Cur...

Economists have done some sophisticated work on media bias. For example, Tim Groseclose and Jeff Milyo have a neat paper, "A Measure of Media Bias," that compares the think tanks that politicians and the media cite. They find that the major media have citation patterns closer to that of the the typical Democrat than the typical member of Congress:

Although we expected to find that most media lean left, we were astounded by the degree. A norm among journalists is to present “both sides of the issue.” Consequently, while we expected members of Congress to cite primarily think tanks that are on the same side of the ideological spectrum as they are, we expected journalists to practice a much more balanced citation practice, even if the journalist’s own ideology opposed the think tanks that he or she is sometimes citing. This was not always the case. Most of the mainstream media outlets that we examined (ie all those besides Drudge Report and Fox News’ Special Report) were closer to the average Democrat in Congress than they were to the median member of the House.

While I love this paper, I want to propose a much more straightforward test of media bias: Simply by reading the title of the article, can you tell what the reader is supposed to think about the story? This is amusingly easy. For example, here are a few titles from the front page of the Washington Post:

Title: Debate on Climate Shifts to Issue of Irreparable Change; Some Experts on Global Warming Foresee 'Tipping Point' When It Is Too Late to Act

What the Reader is Supposed to Think About the Story: We should listen to these experts and act before it is too late.

Title: Some Palestinians See End of Secular Dream; Election Win by Islamic Group Hamas Clouds Prospects for Arab Nationalism

What the Reader is Supposed to Think About the Story: Palestinians' secular dream should not end.

This isn't always possible. For the article "ABC's Woodruff Injured in Iraq," it's not clear what deeper inference the reader is supposed to draw. "Woodruff shouldn't have gone to Iraq"? No. "The U.S. shouldn't be in Iraq?" That's reaching. Sometimes a cigar is just a cigar. What's surprising, however, is how often the media's cigars are much more.

If you're feeling puckish, you'll put my own titles to my test. Fair enough:

Title: At First Glance: Bias in the Media

What the Reader is Supposed to Think About the Story: The media is obviously biased.

You'll conclude that I, too, tell my readers what to think about the facts. Fair enough, but at least I admit it - and so would most bloggers. Old economy media like newspapers and tv should get off their high horse and do the same.


Comments and Sharing





TRACKBACKS (3 to date)
TrackBack URL: http://econlog.econlib.org/mt/mt-tb.cgi/443
The author at In Lehmann's Terms in a related article titled Caplan on Media Bias writes:
    Bryan Caplan links to a paper by Tim Groseclose and Jeff Milyo, A Measure of Media Bias, that proposes to measure media bias by comparing how often media stories cite ostensibly left wing or right wing think tanks with how [Tracked on January 30, 2006 1:13 AM]
COMMENTS (47 to date)
Matt McIntosh writes:

This is a pretty good "blink" test. I took a boring media studies class in high school, but I think I'd have learned more about media bias by spending an hour a day doing this sort of thing for several months instead.

Vorn writes:

To the extent that people make decisions concerning what to think based on mere headlines, we have much deeper problems than media bias.

The concern about media bias is silly. The real issue is to teach people to realize that ALL media is biased and MUST be biased. Why? Because given limited resources, some relevant facts must be excluded while others are included. A newspaper story should be, at most, the beginning of an intelligent inquiry; not the end of the analysis. Getting the media to admit it is biased is a waste of time; it is not in the media's interest to admit to the obvious -- they never will. Fox News is "Fair and Balanced." Yeah right. No they aren't. And neither is any other form of media. A more productive use of time would be trying to educate people to wonder about what facts are not being reported.

One last point, this academic article is SERIOUSLY biased. It excluded Fox News? Just ask Bill O'Reilly what the number 1 news program is. What is the point of excluding one media outlet? How about a study that just focuses on Fox News? Then maybe we would find their is a conservative media bias.

The solution is not to try to change the media; it can't be done, bias will always exist because it MUST exist. Some relevant facts must be excluded. Nor is it to get the media to admit to bias; that would be against financial self-interest. Why would any particular news outlet admit to bias when their competitors don't? The ONLY solution is to educate others to be intelligent consumers of information.

If I wanted to understand economics, I would be crazy to read this blog exclusively. Hey, there is no such thing as a free lunch. Understanding the truth is hard work.

Jane Galt writes:

I'm not sure that's a good test, for the simple reason that most journalists don't write their own headlines.

Or maybe it is a good test, and we should expand our worries about biased journalists to biased copy-editors . . .

Don writes:

Read it again, Vorn. Fox News wasn't excluded from the study, but was one of the only outlets that measured to the right of the median House member. Interestingly, Fox was closer to the median House member than CNN or any of the major networks.

Vorn writes:

I only had time to skim the study in question. I will note a few questions that it brings to mind that may or may not be answered in the study.

First, what is a median member of Congress? Since Congress has more Republicans than Democrats, isn't the median likely to be skewed in a conservative direction?

Second, do all citations count equally? If member x makes 500 citations to think tanks and member y makes 200 such citations, to the citations by member x outweigh the citations made by member y? If so, then that would skew the definition of a centrist think tank in the direction of those members who make the most such citations.

Third, we should be skeptical of the results of this study. According to this study, Fox News is more centrist than other networks. Further, the study asserts that it does not make any subjective assessments of ideology. That is too bad, because any reasonable person making a reasonable assessment of Fox News would come to the conclusion that it is just as biased as other news outlets.

Fourth, I think that using the citation patterns of members of Congress is NOT a reliable way to determine the ideological dispositions of think tanks. One really does need to make a subjective human judgment.

Regardless of the possible flaws in this study, the conclusion that the media is biased is a correct one. But then again, it MUST be biased. Further, it is biased in all sorts of ways; it is not necessarily the case the sorts of bias focused on by ideologues is the most significant.

For democracy to succeed, you need an educated populace that is not easily swayed by propoganda, one way or the other. If you don't have this, then you are going to have problems. Lets hope that very few people make up their mind based on mere headlines. To the extent they do, that shows that we must invest more in developing critical thinking in education.

With an educated and critical thinking populace, media bias is not much of a problem. Without that, you have much bigger problems on your hands.

David Thomson writes:

I am a “biased” neoconseravtive. So what? I never hide this fact. Pure objectivity is a myth. The best one can do is try to be fair and avoid misunderstanding those who are on the opposite side of the fence from themselves. How can anyone be objective about someone like Joseph Mengele? I can see it now: “We here at CNN are interviewing Dr. Mengele. Is he a disgusting Nazi monster, or merely a doctor doing his best to research the medical importance of twins?”

T.R. Elliott writes:

Here's how this study worked as far as I can tell. (1) Characterize ideology of congressman. (2) Look for correlation between approving congressional references to think tanks and media references to same think tanks; (3) label media with ideology of correlated congressman. The methods of this study are pretty suspect, certainly more suspect than the science of global warming.

From this study, we discover that the Wall Street Journal has more liberal media bias than any news organization they surveyed. Here's what the Dow Jones Company had to say: "The Wall Street Journal's news coverage is relentlessly neutral. Of that, we are confident. By contrast, the research technique used in this study hardly inspires confidence. In fact, it is logically suspect and simply baffling in some of its details." Granted, the Wall Street Journal's editorial and news organization are separated, and both are, of course, part of that "old economy media" (a phrase which is charged with bias, by the way).

This study tells us that the ACLU is a conservative organization, more conservative in fact than RAND.

Having studied the physics of both seismology and meteorology, I will be the first to admit that the science of global warming is complex. Yet the scientific community has reached a narrow consensus on that issue. And I'd say that in general, the physical science of meteorology is much further than the social science of economics. So the headlines you post are sensible in light of truth.

The most common form of media bias I see is that which takes the form of headlines such as "Economists tell us that.." or "Rising gas prices will.."

This is where we find the greatest amount of bias. Business, finance, and economics reporting.

As far as your blog telling people what to think about the facts: this is quite obvious. A blog is an opinion page. This blog, like many opinion pages, is often telling people what to think in spite of the facts. The "old economy media" admit as such as well. Their opinion pages are opinion. Since this blog is not a news organization, and doesn't pretend to be one, the comparison between this blog and "old economy media" is not relevant.

Media bias exists. And the primary bias, from my perspective, is one towards what sells. If anything, that's the problem. Truth suffers because popularity, or sales volume, is not a statement of truth. It's a statement of desirability in comparison to price. And most people desire, in news, to be told what they want to be told. Hence Fox News, for example, panders to the right because that is their audience. It's a nice niche. The "old economy media", in particular broadcast television and newspapers, pander to a broad swath of the American populace. I'd find them a lot harder to characterize. The ideology of the American population that is.

David Thomson writes:

“Hence Fox News, for example, panders to the right because that is their audience.”

There is no such thing as pure neutrality. The very concept is laughable. Still, Fox News is the most professional of all the TV networks. It truly does go out of its way to provide everyone with an “equal” voice. Fox has a number of left wing analysts. It is far more ideologically diverse than all of the other networks combined.

Brad Hutchings writes:

TR, the whole global warming topic is still open to debate. Frankly, I would put much more faith in historians to decide the political solution than the scientists. To me (an outsider), it looks like a replay of CFCs in the 1980s. I was fortunate to have Cicerone and Rowland as instructors for an Honors Physical Sciences course at UC Irvine. Fortunate, because they were in the eye of CFC storm and won Nobel prizes for their work.

Here's how I would characterize their lectures. Rowland was about the science. I may disagree with his politics, but his demeanor was appropriate for a scientist and a teacher. That is, he reviewed and explained the data, made his case, and let that stand for what it did. Cicerone, by contrast... It was all he could do to whip up the concern of the smart sorority girls about the horrific skin cancers they would have by the age of 30 and their future children would have at age 10 because McDonalds wouldn't abandon styrofoam Big Mac boxes. And of course, banning CFCs would show no positive results to the ozone layer for 50 years anyway, but we had to do it or life on earth would end before our own kids would get to go off to school and be in frats and sororities. Arguably, their crusade against CFCs (which had to and did become political, culminating in the Montreal Protocol which BANNED CFCs) made air conditioning more expensive and less available and may very well have resulted in destruction of Space Shuttle Columbia.

So if we have to put the well-being of our planet into academics' hands, I would much rather put it in the hands of historians, who might see the CFC story (and coincidentally, the participants) recycled in the global warming story. Think about how terrible life might be today if we acted on the scientific concensus of 1970, when Paul Ehrlich's Population Bomb dominated and shaped thinking about the future of earth and human beings. If science were devoid of politics, I'd think differently, but it's not. At least with historians, their discipline is all about politics and undeniably so. With scientists, we are often left to question what is fact and what is agenda. Very sad state of affairs.

T.R. Elliott writes:

David Thomson: My opinion is that Fox news panders to people who think Steven Spielberg is a good movie director. People who watch reality TV. People overexposed to manipulative flash advertising. Now please note, I'm not speaking about ideological slant. I'm speaking about delivery. Fox is primarily flash. The content is fairly low. I get my information from a broad range of sources on the internet. From that, it's clear that Fox panders to those who find yelling fests and insult throwing popular. Which, I will admit, I've been doing my share of on this blog, and will cease and desist from doing so. I'm not sure about the balance of Fox. I think they're pretty cozy with the Bush administration. You can call his bias whatever you want. I'd call it ad-hoc focused on short-term political gain (at the expense of conservative principles). The information content per second of Fox is very very low. Therefore it’s pretty darn worthless from my perspective.

Brad: Mentioning Erlich is a red herring. It just doesn't make sense to do it. Erlich did not represent the larger scientific community. So that part of your discussion can be completely discounted. People love to bring up the Simon/Ehrlich bet, for example. That is called drawing on a sample size of one to make a case. That bet was nonsense. Drawing on historians for science policy is laughable. The idea that historians only deal with politics, and therefore they are immune to politics, is like saying that sociologists study politics, and therefore are immune to politics. Doesn’t compute. Sorry. We should draw on a couple sources: (1) scientific consensus and (2) economics, e.g. cost benefit analysis. The fact that a science professor was histrionic about CFCs has no bearing on the need to gain answers, as best as possible, through the scientific process, ensuring as much as possible that the science is unaffected by politics. The fact that science is affected by politics is not a sad state of affairs, it's a fact of life and does not negate science. Otherwise you are just a relativist, a postmodernist who doesn't believe in truth. And in that case, you should be pursuing a degree in postmodernist critical theory. There is no question about the impact of CFCs. Except with the postmodernists. They probably don't even believe in CFCs.

Brad Hutchings writes:

TR, the sad part is that politicization of science, as most definitely happened with CFCs and is most definitely happening with global warming diminishes the rightful political influence of scientists and science. In calling me a relativist, you miss my point. I am skeptical about the so-called global warming "concensus". When I look at the political solutions proposed, such as a $3.6 Trillion transfer of US wealth to other countries under the guise of carbon credits, I can't help but think that global warming is a convenient excuse to suck wealth out of our great country. So when I say let the historians decide, you should read some irony and disdain for scientists who are grinding their post-modern Marxist axes. The solutions proposed in the name of the science always seem to be worse that underlying impending calamity. Go figure.

Assume for a moment that the Montreal Protocol hadn't banned CFCs for safety critical uses, such as booster foam. According to their own data, the ban wouldn't make a measurable difference for half a century. Might not scientists now have a better standing to bring the global warming crusade to us? And why does the global warming "point of no return" alarmism remind me of a really bad episode of Star Trek TNG that ran in the early 90s when all the current global warming alarmists were grad students and post-docs?

T.R. Elliott writes:

Brad: You can't be arguing the truth based upon the consequences. It's invalid. There are three issues to consider: (1) the contribution of humans to global warming; (2) methods to reduce that contribution; and (3) cost/ benefit analysis of those methods. Then a choice, that is based upon a whole host of criteria. Bjorn Lomborg, for example, does not negate the impact of human contributions to global warming. If my memory serves me right, he does not contest it at all. He contests it from a cost-benefit analysis perspective. And I think there is merit in discussing that part of his argument. Just as medical care should be considered from a cost benefit analysis. My point is that the scientific community has reached a fairly tight consensus that humans are directly causing global warming. It is not certain, and there are other factors to consider. But one's economic concerns should not influence science. That is what you and many on sites such as this do. You are just like the postmodernists. Truth is what fits the economic argument. Unfortunately, that is completely false. As far as CFCs, I don't know the specifics, and the fact that there were not exclusions or the likes is unfortunate. But that does not speak to the science of CFCs. I double majored in math and physics at UCSD as an undergraduate and received my Applied Physics/Engineering degree from the engineering school there, with work at SCRIPPS Institute for Geological and Planetary Physics. You may have been exposed to such Marxist professors. I was not. I was exposed to largely politically mainstream people. One advisor was a Reagan Republican. Many of them worked on the bomb, space based weapons (one played a primary role in inventing the free electron laser). The obsession with Caplan and Kling to work liberal/conservative dogma into their blog comments is indicative that they are obsessed with this to the detriment of truth.

And the real answer to why the shuttle blew up: bad engineering. They didn't work within the constraints. It was a cost benefit analysis issue. Did the shuttle blow up because of CFCs? Did the little kid who died when strangled by his seat belt the victim of seat belt laws. One can always find exceptions. Trial lawyers due all the time. I prefer to think much more along the mean, not the edges to try to win debating points.

Brad Hutchings writes:

TR, thank you so much for proving my point. With Montreal Protocol, there was no cost/benefit analysis done. Period. Cicerone and Rowland brag about how they went to see Margaret Thatcher and showed her the data, and she knew then and there that something had to be done, like ban CFCs worldwide, posthaste. No deliberation like, hey, maybe 90% of the CFCs being produced are being sprayed into people's armpits, how about if we just ban that to start. Or, the secondary usage is to cool our homes and cars. Will the public accept a black market in freon in order to save the planet from sure destruction? The questions were not even asked. As to Columbia... Prior to switching to non-CFC foams for the center tank, there mysteriously weren't any foam shedding problems. Yet a single, terse NASA memo in the aftermath of Columbia says that any suggestion that the move to CFC-free foam is a bad (read "politically unacceptable") suggestion. I have a feeling that if Richard Feynman were alive today, he might be the only brave authoritative voice to question that conclusion.

And so TR, this is why I recommend that historians ought to be the arbitors of global warming. We are going down the same path of manufactured scientific concensus and drastic global "solutions", that of course screw the United States because that's the fun thing to do these days. Perhaps if you can't trust the historians, we the people ought to just be able to vote on the issue. If the Palestinians can elect Hamas, I am sure that Americans could be convinced to vote in favor of continued global warming. Especially, if the election were held after a cold winter storm or a night like tonight in SoCal where the temperature dips below 55 degrees.

P.S. Your academic pedigree is most impressive. Observe and study history and try to figure out your context. Then, you'll go far.

Don writes:

Wow, talk about a target-rich environment:

"I only had time to skim the study in question. I will note a few questions that it brings to mind that may or may not be answered in the study."

You didn't read it, but you'll comment on it anyway. Good start!

"First, what is a median member of Congress? Since Congress has more Republicans than Democrats, isn't the median likely to be skewed in a conservative direction?"

Skewed away from what?

"Second, do all citations count equally? If member x makes 500 citations to think tanks and member y makes 200 such citations, to the citations by member x outweigh the citations made by member y? If so, then that would skew the definition of a centrist think tank in the direction of those members who make the most such citations."

Yes, I think that's how Groseclose and Milyo constructed their index.

"Third, we should be skeptical of the results of this study. According to this study, Fox News is more centrist than other networks. Further, the study asserts that it does not make any subjective assessments of ideology. That is too bad, because any reasonable person making a reasonable assessment of Fox News would come to the conclusion that it is just as biased as other news outlets."

"Let's ignore the data, because we know the truth!" There's a couple of interesting posts on confirmation bias over on Marginal Revolution, check 'em out.

"Fourth, I think that using the citation patterns of members of Congress is NOT a reliable way to determine the ideological dispositions of think tanks. One really does need to make a subjective human judgment."

This was discussed in the paper. You would, of course, know this if you had bothered to read it.

liberty writes:

>First, what is a median member of Congress? Since Congress has more Republicans than Democrats, isn't the median likely to be skewed in a conservative direction?

Look more carefully, it wasn't based on just a few years of congress nor the most recent few. It was based on the 1990s.

"For the congressional data, we coded all citations that occurred during the period Jan. 1, 1993 to December 31, 2002. This covered the 103rd thru 107th Congresses. We calculated the average adjusted ADA score for each member of Congress over the period 1993 to 1999."

R.J. Lehmann writes:

"For the congressional data, we coded all citations that occurred during the period Jan. 1, 1993 to December 31, 2002. This covered the 103rd thru 107th Congresses. We calculated the average adjusted ADA score for each member of Congress over the period 1993 to 1999."

Republicans controlled the House for eight of those 10 years, and the Senate for six of those ten years, with a dead draw in the Senate for two other years. How does this avoid the bottom line that the "median" member of Congress over that span is likely to be conservative?

In any case, the point journalistic objectivity is not to replicate the biases of a median member of Congress, or even a median member of the public, but to strive for the abstract goal of "balance." Perhaps that is something not possible to achieve, and perhaps it not worthwhile to even try, but the assumptions of this study don't even seem to recognize what the word MEANS.

liberty writes:

>How does this avoid the bottom line that the "median" member of Congress over that span is likely to be conservative?

Well it may be slightly conservative, but the papers come out far far left of it.

>In any case, the point journalistic objectivity is not to replicate the biases of a median member of Congress, or even a median member of the public, but to strive for the abstract goal of "balance."

And clearly they don't. They tend to have the bias of someone far to the left of the median congressperson. If they were unbiased, they would be likely to come out somewhere toward the median of the congress, simply by accident. A computer program that had all the available data and written to be unbiased would presumably use the sources evenly and therefor land between those that rely heavily on leftist sources and those that rely heavily on rightist ones, hence coming out at the median.

Josh writes:

> the bottom line that the "median" member of Congress over that span is likely to be conservative

... as conservative as the median member of the public that voted from them (well, roughly)... And the last I checked, the median member of any group is defined as that group's "center". Hence the study's claim that the median member of congress represents a "centrist" position.

Boonton writes:

T.R. Elliot does a nice job summarizing what's wrong with this survey. If you're measuring device tells you that the ACLU is more conservative than Rand, the Wall Street Journal is the 'most liberal' of old school mediea then the problem isn't with the media but your measuring device.

I see a problem with Arnold's method as well. What are the headlines supposed to say? For example:

Title: Debate on Climate Shifts to Issue of Irreparable Change; Some Experts on Global Warming Foresee 'Tipping Point' When It Is Too Late to Act

What the Reader is Supposed to Think About the Story: We should listen to these experts and act before it is too late.

This tells us that:

1. The debate is shifting to an issue that was either unaddressed or given less attention before, irreparable change.

OK, now this is a technical debate and I dont' know if anyone has done any surveys of climate experts but it is probably an accurate description of what people that talk about global warming from a technical POV are talking about.

2. Some experts are seeing a 'tipping point' approach.

Some experts probably are seeing that. Back when Kyoto was a big point I recall some critics saying stuff like its only going to delay global warming by '5 years' or something like that. I've seen stories from people saying that we have to accept global warming and concentrate efforts on how to manage it rather than trying to reverse it. Now a tipping point is newsworthy because:

A. It's current.
B. it's important.


Now what about the people who think global warming isn't happening? How about the people who think its a good thing? How about the people who think avian flu will kill us all first? How about the people who think that hydrogen will have some miracle breakthrus first making the issue moot?

All those people exist but there's no need for the unbiased reporter to cover them in a single story. Here's an example, if a newspaper is covering a local Synagogue that is choosing a new Rabbi it does not have to quote Jews for Jesus, does not have to quote evangelical Christians who think the Jews should convert, does not have to quote athiests who think religion is bunk, does not have to quote Muslims who think they should convert to Islam, does not have to quote anti-Semites who want to blow the place up and so on!

What I would expect an unbiased article on the subject to tell me is what are the issues involved in the decision and where do the major players stand on it.

If, for example, after reading the article on global warming you are aware that some scientists are skeptical it even exists then that is all that is sufficient in my book. You do not need the article to quote skeptics equally, to figure out what % of the scientific population consist of skeptics and quote them exactly that often and so on. In fact for some articles it makes sense to have no coverage of 'the other side'. For example, shouldn't a newspaper be able to cover what the evangelical community thought of Bush's original nomination of Meyers? But if you insisted that such an article quote equally to be unbiased it would have to double in size as every evangelical position is coupled with a quote from NOW or Democratic groups!

liberty writes:

>If you're measuring device tells you that the ACLU is more conservative than Rand, the Wall Street Journal is the 'most liberal' of old school mediea then the problem isn't with the media but your measuring device.

Because you assume that you know better than the report? That isn't science at all, thats hogwash.

The Wall Street Journal is actually considered by many to be left leaning in its editorials and more and more in its articles. Just because you think you know better doesn't make the measurement by a scientific study wrong.

A scientist might wonder whether there was a mistake if a result seems to be the opposite of what he predicted, but he doesn't simply throw out the results; he checks his own assumptions, perhaps he retries the experiment, he studies the inputs and outputs, and then perhaps tries an additional measure. Your response is one of a partisan, not a scientist.

R.J. Lehmann writes:

"... as conservative as the median member of the public that voted from them (well, roughly)... And the last I checked, the median member of any group is defined as that group's 'center'. Hence the study's claim that the median member of congress represents a 'centrist' position."

But to be "centrist" is most definitely not the same thing as to be balanced or unbiased. A centrist's bias is against opinions that swing toward either extreme. The goal of journalistic objectivity is to be able to provide a clear and factual account of news events. When matters of opinion or interpretation are vital to those events, the goal should be to present (over time, if not all in one sitting) a reasonable balance of those views that are truly relevant to the matter at hand from across the spectrum of thought.

liberty writes:

>When matters of opinion or interpretation are vital to those events, the goal should be to present (over time, if not all in one sitting) a reasonable balance of those views that are truly relevant to the matter at hand from across the spectrum of thought.

And that is what the study aims to record. If you consistantly quote left or right leaning think tanks - as judged by how politicians quote them - then you are slanted. If you quote both sides, both opinions, left and right think tanks, then you are objective or centrist.

"A feature of our method is that it does not require us to make a subjective assessment of how liberal or conservative a think tank is. That is, for instance, we do we need to read policy reports of the think tank or analyze its position on various issues to determine its ideology. Instead, we simply observe the ADA scores of the members of Congress who cite the think tank. This feature is important, since an active controversy exists whether, e.g., the Brookings Institution or the RAND Corporation is moderate, left-wing, or right-wing. "

That is why the study uses congress.

R.J. Lehmann writes:

"And that is what the study aims to record. If you consistantly quote left or right leaning think tanks - as judged by how politicians quote them - then you are slanted. If you quote both sides, both opinions, left and right think tanks, then you are objective or centrist."

But there are very rarely just two sides to any given controversy, and sources don't really line up as neatly as of that.

I am a journalist and I cover the insurance industry for a trade publication. My audience is undoubtedly, by most standards, right wing. I would likely also be considered so by the median member of the American public. But while I have no problem admitting, for instance, that my bias is in favor of markets, I don't think you could tell very much about that from the sources I cite. To give you some examples, here are a few recent or ongoing issues before Congress that I've been following, and a few examples of those I've cited for and against pieces of legislation:

-- Extension of the federal Terrorism Insurance backstop:

Mike Oxley
Richard Baker
Insurance companies
Real estate Developers
Commercial banks
RAND Corp.
Barney Frank
Chuck Schumer

AGAINST -- The Consumer Federation of America
Public Citizen
Richard Shelby
The Bush Administration
Several Bermuda reinsurers

-- The Asbestos Trust Fund

FOR:
Manufacturers
Bill Frist
President Bush
Veterans of Foreign War
The United Auto Workers
Dianne Feinstein
Arlen Specter
Pat Leahy

AGAINST:
Harry Reid
the Trial Bar
Insurance companies
National Taxpayers Union
FreedomWorks
AFL-CIO
Jon Cornyn

-- Federal regulation of insurance companies

FOR:
Life insurers
Large brokers
Large commercial insurers
John Sununu
Barney Frank
Paul Kanjorski

AGAINST:
Insurance agents
Small personal lines insurers
Mike Oxley
Elliot Spitzer

Clearly, I could pick and choose "right" and "left" sources to craft completely one-sided articles on any of the subjects above, or choose exclusively right or left sources and still present balanced sources. And none of it would have very much to do with what I think, personally, on any of those topics.

Boonton writes:
Because you assume that you know better than the report? That isn't science at all, thats hogwash.

The Wall Street Journal is actually considered by many to be left leaning in its editorials and more and more in its articles. Just because you think you know better doesn't make the measurement by a scientific study wrong.

Sounds like you've constructed a neat little argument from authority. Because the report tells us so it must be so.

True, if a thermometer told me a block of super cooled ice was -75 degrees but it feels burning when I touch it I would trust the thermometer because it has a long track record of measuring temperature in a way that is much more reliable than skin sensation. However if a new measuring device just off the drawing table tells me a block of ice is hotter than a pot of boiling water the most reasonable assumption for me is that the device is broken.

Now I read the Journal quite often and find most of its articles to be about as neutral as possible while its editorials run very much to the right with the exception of occassional guest writers. Perhaps the NY Times is really more conservative but I doubt it. Now how do you deal with the ACLU? Well one obvious issue is did the report determine if the Congressmen were citing the think tanks in positive or negative contexts? If Republicans spend a lot of time bashing the ACLU then an article with lots of ACLU cites will look similar to a conservative Congressman's profile using this system. Yet the opposite is really the case if the citations are presented in a neutral or positive manner.

Let's imagine another scenero. Let's imagine two think tanks; conservative and liberal. The tanks, though, do not spend equal time issuing reports on all subjects. Let's say on liberal causes like the min. wage, pro-union laws, environmental issues etc. the liberal think tank issues numerous reports covering many angles while the conservative one issues just simple blanket policy statements. Likewise let's say the conservative tank is issuing lots of reports on issues like English-only, illegal immigration, etc.

Now a detailed article on one type of issue may yield many think tank citations from one side while the other side only gets short statements. Is that biased? Would it be fair to ignore 100 reports from one side just because the other side only bothered to do 1 report? If you job is to summarize the body of knowledge on a particular topic are you telling me it would be unbiased to ignore significant amounts of knowledge in order to give the 'other side' time to catch up? That sounds more like some strange sort of intellectual affirmative action.

Finally let's imagine another complication. Suppose Congress spends a lot of time on some issues but not others. For example, suppose Republicans spend a lot of time disputing the ACLU for challenging the Bush admin. on its terror policies. That exagerrates the ACLU's rank in this system as a 'conservative' think tank. Likewise suppose liberals spend a lot of time bashing Bush's deficits. They might cite a lot of think tanks that concentrate heavily on the deficit making them appear more liberal.

However if a newspaper does an article on the deficit don't they have to quote the leading think tanks that concentrate on the issue? If Democrats ahve been citing those very think tanks because Bush is truely bad on that topic the article is going to score liberal no matter what.

And that is what the study aims to record. If you consistantly quote left or right leaning think tanks - as judged by how politicians quote them - then you are slanted. If you quote both sides, both opinions, left and right think tanks, then you are objective or centrist.

No you'll just happen to have a quoting pattern that matches the pattern of politicians described as centrist. I think objectivity can be judged on a case by case basis but what people here want is a simple system that feeds, say, all last years articles into a program and spits out a score. Of course one can create such a program but the question is does the score have any real meaning?

Boonton writes:

To bring the point home let's suppose Arnold is teaching a class on the history of economic thought. A lazy student feeds in the texts of Keynes and Milton Friedman into a program like this. He notes that Keynes cites classical economist Marshall quite often, he also notes that Miltion mentions Keynes quite often.

As a result he answers on a test that all three are classical economists. What's his grade?

liberty writes:

>Well one obvious issue is did the report determine if the Congressmen were citing the think tanks in positive or negative contexts?

If you read the study a little more closely you'll see that this is addressed. The think tanks are being cited, they are not being argued with. If they are described in a such a context as not to be seen as an unbiased source (eg the liberal think tank the EPI said such and such, but I see it differently) then they weren't counted.

>Let's imagine two think tanks; conservative and liberal. The tanks, though, do not spend equal time issuing reports on all subjects.

It is based on the most popular 200 think tanks. Because of freedom and democracy in this country, there tends to be two sides represented for each of the issues. If you want to talk about unions there are pro-union and anti-union think tanks. Both sides will have many reports on the subject, generally. In addition, the paper you write for will not only cover one subject - trade journals on a certain topic are not included. So if you cite one think tank for union issues, what do you cite when you discuss the economics of marriage vs. single mothers? What think tank do you cite when you discuss the deficit? etc.

>For example, suppose Republicans spend a lot of time disputing the ACLU for challenging the Bush admin. on its terror policies. That exagerrates the ACLU's rank in this system as a 'conservative' think tank.

Again, disputing a think tank won't count as citing it.

>However if a newspaper does an article on the deficit don't they have to quote the leading think tanks that concentrate on the issue?

And which one would that be, pray tell?

There are think tanks that focus on the deficit on both sides - liberal and conservative.

>As a result he answers on a test that all three are classical economists. What's his grade?

Neither congresspeople nor journalists are citing as part of an academic study and only citations that are deemed to be using the citation as a claim for a positive factual support are considered.

asg writes:

Wow, it's clear that those professors have no idea what they're talking about, and their study is shoddy, because look at how easy it was for a bunch of blog commenters to tear it apart without even needing to read it! I bet that if the study's authors read this comment thread, they would be speechless, having failed to anticipate a single objection that was raised here. Someone contact the journal it's being published in and let them know about this thread so they can publish it as an addendum. Boonton's comment comparing academic citations to congresscritters' citing of think tanks was particularly incisive and ingenious, because as we all know congresscritters are JUST LIKE academics, and congressional press conferences and soundbites are JUST LIKE academic treatises in their attention to detail, concern for opposing arguments, and attempts to honestly engage issues! In fact, look at how many congresscritters were academics before they got elected! Almost all of them! Right?

Boonton writes:

indeed it's also pretty stunning that liberty has proven that all think tanks on both sides cover all issues with equal attention because of 'freedom and democracy'.

Here's an idea, it sounds like this concept could be reformulated into a neat little web tool. Just paste in an article and out pops a 'bias score'. Would it be a good idea for a newspaper to replace some of their editors with just data crunchers who will spit articals back at reporters until their bias score falls?

Boonton writes:

Ohh BTW, I'd be happy to contact the journal its being published in but the only link provided seems to lead to the author's academic homepage where the paper resides.

Chris Bolts writes:
But one's economic concerns should not influence science.

This is akin to saying that one's concerns should not wory about the laws of physics when he/she runs smack into a brick wall. Scientific matters should ALWAYS consider the economic benefits and costs imposed on society, otherwise what use is it to undertake any venture improves mankind or try to come up with solutions to existing problems as a result of scientific breakthroughs?

To bring the point home let's suppose Arnold is teaching a class on the history of economic thought. A lazy student feeds in the texts of Keynes and Milton Friedman into a program like this. He notes that Keynes cites classical economist Marshall quite often, he also notes that Miltion mentions Keynes quite often.

As a result he answers on a test that all three are classical economists. What's his grade?

It would all depend upon the definition of "classic". If by classic you mean that someone whose principles are still being used today, whether academically or practically, then all would be classic. If by classic you mean all those who are dead, then Milton Friedman could not be considered a classical economist. I could go on, but you get my drift. An argument can be made and supported in which all three could or could not be labeled as classical economists just as an argument can be made about what is or isn't media bias, but it isn't necessarily the content of the argument that determines a person political predisposition, but also by other mitigating factors. For example, after the recent behavior of the ACLU in recent months I would not classify them as "conservative" in a political sense. But if we go back to the original meaning of the word "conservative" then perhaps the ACLU could be considered a conservative organization in that they refuse to forfeit any of their civil liberties as ordained by the Constitution. Because of this I tend to hate reading reports of media bias because the waters that constitute what is or is not "bias" is murky and is best treaded carefully.

liberty writes:

>But if we go back to the original meaning of the word "conservative" then perhaps the ACLU could be considered a conservative organization in that they refuse to forfeit any of their civil liberties as ordained by the Constitution.

Except the 2nd amendment, the 10th, the right of property in the 5th, and also the rights they do protect are protected unevely, usually protecting terrorists, young women who want to abort their babies without informing their parents, etc. and not any groups or institutions that disagree with their interpretations of their favorite constitutional rights or spend their time protecting those rights that they ignore.

>liberty has proven that all think tanks on both sides cover all issues with equal attention because of 'freedom and democracy'.

Why don't you pick a few issues and go through the 200 and see whether any of those issues are only represented by think tanks that you see as right or left leaning. Come back and prove your point because in my experience there tend to be two sides that pay attention to every issue in this country - that is why we effectively have a two party system. Abortion? You'll find pro and con; eminent domain? pro and con; anti-trust? pro and con; McCain-Feingold? pro and con; taxes? pro and con; any issue discussed in congress or in the media has at least two sides and think tanks that have written papers on them.

Boonton writes:

It would all depend upon the definition of "classic". If by classic you mean that someone whose principles are still being used today, whether academically or practically, then all would be classic. If by classic you mean all those who are dead, then Milton Friedman could not be considered a classical economist. I could go on, but you get my drift.

I think you're just being cute here. Classical (and neo-classical) are well recognized schools of economic thought. Yes we can argue over whether or not someone belongs to this or that school but the point is that there are real limitations to using this citation system as a short cut to evaluating stuff on a one by one basis. To use a more extreme illustration, Friedman wrote a lot about Marx yet no one would think he was a Marxist. Even if you adjust the citation system to incorporate critical citations I think the problem remains.

liberty:

Except the 2nd amendment, the 10th, the right of property in the 5th, and also the rights they do protect are protected unevely, usually protecting terrorists, young women who want to abort their babies without informing their parents, etc. and not any groups or institutions that disagree with their interpretations of their favorite constitutional rights or spend their time protecting those rights that they ignore.

Really? Why don't you back that up. The ACLU has a pretty good record of defending the rights of those who are very politically incorrect.

Why don't you pick a few issues and go through the 200 and see whether any of those issues are only represented by think tanks that you see as right or left leaning. Come back and prove your point because in my experience there tend to be two sides that pay attention to every issue in this country...

You're missing the point. It isn't that you can find a position paper on just about any major issue from any major POV. The point is that there is no reason to assume that liberal and conservative groups spend an equal amount of time issuing relevant research on all topics evenly.

A good example of this happened right here. When Maryland passed the 'Wal-Mart' law (any employer of 10,000 or more had to spend at least 8% of payroll on healthcare or put the difference into the state Medicaid fund) the authors here analyzed it using the old simplistic argument against the min. wage. Missing from the analysis was any attempt to look at the specifics of the law & the situation. For example, comparing the burden on Wal-Mart to tax breaks and subsidies that Wal-Mart is able to win based on its size and brand appeal. A more liberal orientated think tank, though, is more likely to do that.

So a reporter writing an unbiased story on the law finds a rightist source that attacks the law from a theory POV. He finds several sources, though, that defend the law from a liberal POV and he will probably want to find some examples of ancedotes. Recently the NY Times Magazine had a long article on 'living wage' laws. From my memory (which isn't perfect), the main features included:

1. A review of how the 'living wage' movement got started & some of its key players.

2. The classical economic argument against it.

3. The well known NJ study that indicated small increases in the min. wage did not hurt employment. Also Robert Reich, liberal economist, who stated that the min. wage was inferior to the Earned Income Tax Credit as a way to boost income for poor workers but he would accept it as better than nothing.

4. Several profiles that personalize some of the arguments. They included the worker who got a raise due to the law, the diner owner who felt he was unfairly depicted by advocates of the law as some type of greedy sweatshop type but was in fact very happy he provides quite a few jobs for the community that does not require advanced skills or education and some surveys/studies of businesses a year after the law passed that did not show strong results either way.

So in sum the impression I got was that this was an unbiased article. It provided me with both sides and tried to explore the potential problems with the reasoning of both sides as deeply as possible. Also, very importantly, it tried to deflate superficial sloganistic reasoning by providing concrete examples such as the business owner that illustrated the world is nowhere nearly as black and white as the more ideological among us often depict it as.

Now using the citation method from above I count:

'Liberal'
- The review of the start of the living wage movement
- The NJ study showing min. wages do not hurt employment at least in some cases.
- Robert Reich but a computer program might miss the nuanced argument he presented that the EBIT was better...in fact Reich should actually be under the conservative column since he was critical of the law & movement but the citation system would almost certainly peg him as liberal.
- Profile of the worker who works for low wages

'Conservative'
- Classical economic argument
- Profile of the business owner who opposed the law

Neutral
- The survey/study of the effects of the law a year later.

From that method the story would score more liberal than conservative...in fact the above undestates the liberal citations since the review of the history of the living wage movement would include much more than a single citation since the movement involved several groups in different cities.

Now the primary purpose of the article was the explore the living wage movement which primarily exists in very liberal cities such as San Francisco. Will there be sources for 'both sides'? Of course but the fact remains that there will be more relevant sources for the liberal side. Likewise if you did a story profiling the evangelical movement in the Republican party almost all of your relevant sources will score as conservative even though there will be those critical.

My test for bias is much simplier but more reliable. I ask myself if I only had this story or article to go on would I be able to summarize the major arguments of the main sides (often at least two but certainly not always) in the issue?

This test has some shortfalls. You have to read or view the article you are judging. It's very time consuming if you want to grade all the news. Most importantly there's a fair amount of room for disagreement as well (but not as much as one would think at first glance). However is this any different than a teacher grading student essays? I doubt Arnold or anyone here would agree that any computer software available today could do the job automatically. They would also agree that there's a degree of subjectivity involved (one Professor's A is another's B-) but I don't think they would throw up their hands and declare the whole enterprise impossibly subjective either!

liberty writes:
liberty:

Except the 2nd amendment, the 10th, the right of property in the 5th, and also the rights they do protect are protected unevely, usually protecting terrorists, young women who want to abort their babies without informing their parents, etc. and not any groups or institutions that disagree with their interpretations of their favorite constitutional rights or spend their time protecting those rights that they ignore.

Really? Why don't you back that up. The ACLU has a pretty good record of defending the rights of those who are very politically incorrect.

This is a good dissection of the way the ACLU treats the 2nd amendment differently than it treats the 1st.
http://rightreason.ektopos.com/archives/gun_rights/

This article describes the ACLU's protecting of pedophiles in the name of "free speech" - though it is not speech when they rape young boys - and at the same silences people who wish to pray in school.
http://www.thebatt.com/media/paper657/news/2003/01/14/Opinion/Stripping.Pedophiles.Of.Their.Civil.Liberties-515174.shtml?norewrite&sourcedomain=www.thebatt.com

There may be reason to protect NAMBLA, disgusting as it is, but is is disgusting: http://crookedtimber.org/2004/03/05/how-far

And they protect child nudist camps: http://www.traditionalvalues.org/modules.php?sid=1718

And they do not protect the rights of the religious at all:
http://www.worldnetdaily.com/news/article.asp?ARTICLE_ID=41915

causing this kind of mess:
http://www.christianlaw.org/christmas2004.html

-- I am not religious nor a social conservative, but it is obvious to me that the ACLU has an agenda and picks and chooses what constitutional rights to over-do (separation of church and state, protection of free speech even when it is in the name of violence, rape, etc and when the groups speaking mean to violate laws and rights of others) and which to completely ignore (gun rights, religious freedom, etc)


>A good example of this happened right here. When Maryland passed the 'Wal-Mart' law

Once again you've proven your liberal bias. There are many "well known" studies that show that the minimum wage does hurt the poor: causing unemployment, pushing businesses away, etc. Your article *was* liberally biased as it ignored those studies and cited only the background of classical economics and one single study that purportefd to find the opposite, letting readers beleive that all factual evidence conflicts with classical theory - which is completely false.

There are at least as many hard science studies to back up the problems with minimum wage, as liberal ones that support that kind of law.

In addition, review of the start of the movement could include people who opposed it for theoretical reasons from the start and should not land in the liberal category if it was truly neutral. And once again you assume this system (which, by the way if you read the article on it is not a computer program) is faulty by saying that "Reich should actually be under the conservative column since he was critical of the law" but assume that the system could not notice that, but as I said it can and does.

Boonton writes:

-- I am not religious nor a social conservative, but it is obvious to me that the ACLU has an agenda and picks and chooses what constitutional rights to over-do (separation of church and state, protection of free speech even when it is in the name of violence, rape, etc and when the groups speaking mean to violate laws and rights of others) and which to completely ignore (gun rights, religious freedom, etc)

You haven't provided anything of substance other than to say that the ACLU has taken legal stances that you would not take. You provide no legal arguments but you seem to imply that the unpopular or even criminal should not have their civil liberties protected. Needless to say this stands at odds with the American system of gov't and over 200 years of history here.

There are numerous cases where the ACLU has sided with those who are certainly not liberals. Defending the right of groups like the KKK and neo-nazi's to march is an easy example. There are also numerous cases where the ACLU has stepped in and defended religious people in public schools who were told, for example, they could not do a book report on the Bible when the class was allowed to pick any book they wanted.

Once again you've proven your liberal bias. There are many "well known" studies that show that the minimum wage does hurt the poor: causing unemployment, pushing businesses away, etc. Your article *was* liberally biased as it ignored those studies and cited only the background of classical economics and one single study that purportefd to find the opposite, letting readers beleive that all factual evidence conflicts with classical theory - which is completely false.

Your imagination has invented some fantasy world where I said that the only study ever done was the one that showed no harm. Since your perception of bias is based on something that objectively is not there how much respect should it be given?


In addition, review of the start of the movement could include people who opposed it for theoretical reasons from the start and should not land in the liberal category if it was truly neutral. And once again you assume this system (which, by the way if you read the article on it is not a computer program) is faulty by saying that "Reich should actually be under the conservative column since he was critical of the law" but assume that the system could not notice that, but as I said it can and does.

I'm not convinced that the system could detect such a nuanced view as Reich's unless it has a human beign read through each article evaluating whether cites are critical enough to make them critical. If you need a human beign to read each article is this system any better than just having someone 'grade' every article manually?

You miss the larger point. Suppose the article was just focusing on the change in the debate from the pro-min wage side. In other words the shift away from a national movement to increase the min. wage towards local groups passing city laws requiring various types of living wages. Such a narrowly focused article would probably use many sources that would read liberal by the computer but could be unbiased.

IMO a better standard is simply to ask after reading the article can you articulate the major arguments of the major sides in the article's focus whether it be as broad as the min. wage or as narrow as who will win the Democratic primary in Vermont?

Boonton writes:

Also you are right that the ACLU does not seem to take much of an absolutist position on the 2nd amendment. But so what? You obviously don't take much of an absolutist position on the 1st!

No one ever said here that either you or the ACLU has a monopoly on correct Constitutional interpretations.

More important, though, is the limiting of bias to just two camps; liberal.vs.conservative. This ignores the fact that there's a considerable degree of overlap between both when dealing with perfectly valid alternative POV.

For example, take US support of Israel. You'll find plenty of pro-Israel liberals. If you recall the first Bush administration, it was heavily criticized by many for 'Arabists' in the state dept. who advocated treating Israel coldly to advance relations with other Arab countries. Likewise you'll find anti-Israel sentiment in such obvious conservatives as Pat Buchannan and Robert Novak.

How would one write an unbiased article on Israel under this system? One could construct an article that easily passes the test but is actually biased for or against Israel by citing only people in the same camp on both the liberal and conservative side....say Noam Chomsky and Pat Buchannan!

Libertarians are another example. If you know anything about the history of ideas and different schools of thought you know that libertarians are not just conservatives with more sex, drugs and booze. They have different fundamental assumptions that while they often land in the Republican camp makes them different intellectual creatures.

liberty writes:

>You provide no legal arguments but you seem to imply that the unpopular or even criminal should not have their civil liberties protected. Needless to say this stands at odds with the American system of gov't and over 200 years of history here.

No I didn't. I simply pointed out that the ACLU chooses to protect NAMBLA and *not* those who have their religious liberties in jeapordy - for example private organizations that display Christian symbols.

There were several dozen lawsuits against groups that wanted to to keep religion in public or in a private institution this past few years, and the boyscouts, etc. Sometimes Christians were actually discriminated against while Jews and Pagans were allowed to have religious symbols, and rather than defend anyones religious freedom, the ACLU took the side of government mandated atheism. They are picking and choosing rights to protect and always pick the leftist point of view. Its also not coincidence that they were started by communists. I am not saying that they don't have a right to an agenda, just pointing out that they are non unbiased.

>Your imagination has invented some fantasy world where I said that the only study ever done was the one that showed no harm.

No I didn't. I simply pointed out that the article is biased if it chooses as the only study using empirical evidence one that is in favor of the minimum wage.

>I'm not convinced that the system could detect such a nuanced view as Reich's unless it has a human beign read through each article evaluating whether cites are critical enough to make them critical. If you need a human beign to read each article is this system any better than just having someone 'grade' every article manually?

It does use a human being and yes it is better because it is a human being (many in fact) using a systematic non-biased approach. Its called science.

>IMO a better standard is simply to ask after reading the article can you articulate the major arguments of the major sides in the article's focus whether it be as broad as the min. wage or as narrow as who will win the Democratic primary in Vermont?

That is far too subjective. You need a standardized approach or it isn't science.

Even a very narrow subject such as a primary election or a change in focus of a group should be able to withstand this systematic approach. I suggest that you go back and actually read the report before continuing on your cruisade. When you are done reading the report and understand that it isn't a computer, maybe you'll see why your objections are baseless.

>Also you are right that the ACLU does not seem to take much of an absolutist position on the 2nd amendment. But so what? You obviously don't take much of an absolutist position on the 1st!

Actually I do. I would never ban NAMBLA pamphlets, disgusting as they are. I simply see it as a left-wing group because they only protect left-wing causes. That's fine. They have the right to exist. But they should not be mistaken for an unbiased institution.

>How would one write an unbiased article on Israel under this system? One could construct an article that easily passes the test but is actually biased for or against Israel by citing only people in the same camp on both the liberal and conservative side....say Noam Chomsky and Pat Buchannan!

Finally, an objection that has merit.

I agree on this count. I think the libertarian argument is also valid - if there are any conservative think tanks that are protectionist, for example, then I would not want an article that cites them to be considered conservative.

Still, there are relatively few cases where this kind of thing will come up and the think tanks that go to the "other" side or a third side as a deviation from normal two-party system can either be excluded (Pat Buchanan's anti-Isreal think tank can be ignored, along with anything coming from Chomsky) or a third category can encompass the viewpoint (add libertarian to left and right, the congresspeople can be given a libertarian scoring by a libertarian group so people know where they stand and then newspapers can be compared with this scoring as they are to the ordering of democrat/republican - the only problem is the different brands of libertarianism...) or something like that.

In any case, it doesn't make the study useless, just limited.

Boonton writes:

No I didn't. I simply pointed out that the ACLU chooses to protect NAMBLA and *not* those who have their religious liberties in jeapordy - for example private organizations that display Christian symbols.

That's an odd thing to say. More than a few times I've heard about the ACLU intervening when private religious displays or speech were threatened. Just off their web site, hardly burried or hidden (http://www.aclu.org/religion/)

PROVIDENCE, RI -- The American Civil Liberties Union of Rhode Island announced today that it has filed an appeal in federal court on behalf of a Christian prisoner who was barred from preaching during religious services at the state prison.
TAMPA, FL - In a victory for free speech, Polk County officials today agreed to temporarily eliminate a burdensome rule that had been challenged by the American Civil Liberties Union requiring that individuals and groups purchase expensive insurance policies before being allowed to set up displays in front of the county administration building.
ACLU of New Jersey Joins Lawsuit Supporting Second-Grader's Right to Sing "Awesome God" at Talent Show (09/20/2005) NEWARK, NJ -- The American Civil Liberties Union of New Jersey today announced that it is seeking to participate as a friend-of-the-court in the case of a second-grade student who was barred from singing a religious song in a voluntary, after-school talent show.


PHILADELPHIA -- In Pennsylvania's first court interpretation of the state's Religious Freedom Protection Act, a city judge has ruled that a devout Muslim firefighter who refuses to shave his beard on religious grounds cannot be fired while his legal case continues.

ACLU of New Jersey Successfully Defends Right of Religious Expression by Jurors (12/22/2004) NEWARK, NJ-- The State Supreme Court ruled today that a prosecutor violated the New Jersey Constitution when he removed two jurors from a jury pool, one for wearing Muslim religious clothing and another for having engaged in missionary activity.

Hardly sounds like good stands to take if your goal is to establish "government mandated atheism"


It does use a human being and yes it is better because it is a human being (many in fact) using a systematic non-biased approach. Its called science.

Ahhh yes, they blinded us with science! GIGO is a scientific expression as well.


Still, there are relatively few cases where this kind of thing will come up and the think tanks that go to the "other" side or a third side as a deviation from normal two-party system can either be excluded (Pat Buchanan's anti-Isreal think tank can be ignored, along with anything coming from Chomsky) or a third category can encompass the viewpoint (....

It's stunning that your solution to the problem of writing an unbiased article begins by excluding points of view. Weren't you just raking me over the coals for not mentioning studies that did show a cost to the min. wage? If you can exlude libertarian orientated position research why can't you, say, exclude conservatives and just present the left and hard left?

No I didn't. I simply pointed out that the article is biased if it chooses as the only study using empirical evidence one that is in favor of the minimum wage.

Every article I read about that study began by stating the study challenged economic orthodoxy and the results of previous studies, IMO that is sufficient. The study was unique since it illustrated a rare 'controlled experiment' where employment could be compared in to geographically close and economically similiar areas whose only difference was the change in policy.

Actually I do. I would never ban NAMBLA pamphlets, disgusting as they are. I simply see it as a left-wing group because they only protect left-wing causes. That's fine. They have the right to exist. But they should not be mistaken for an unbiased institution.

Wait a second! Let's leave aside the odd assertion that NAMBLA is a left wing group. The fact is you can easily find dozens of cases where the ACLU is defending the free speech rights of right wingers as well as those who are not on the left-right spectrum.

In any case, it doesn't make the study useless, just limited.

I think we can come to some common ground there. There have been interesting things done studying rates of citation in various fields. There's probably more interesting patterns going on there than simple bias studies though.

liberty writes:

>Hardly sounds like good stands to take if your goal is to establish "government mandated atheism"

Ok, you win on that one. I was under the impression that they had not been supporting those causes, but rather in their fight for separation of church and state had caused some of that trouble. Still they ignore the 2nd amendment.

>Ahhh yes, they blinded us with science!

No, they didn't blind us at all. We can see exactly the rules the humans use to make the choice and it is much better than a subjective reading, as you suggest.

>exlude libertarian orientated position research why can't you, say, exclude conservatives and just present the left and hard left?

Because we live in a two-party country at present. If newspapers slant any which way they will primarily slant toward one or the other of the two parties - if you stack the congress with Libertarians then the study will automatically start comparing things to Libertarians! Imagine that! Anyway, that is where all the money is so newspapers will tend to pander to them and with a few execptions the split is pretty easy.

If some newspapers actually slant "Green" they will look like they slant democrat as they will cite similarly to democratic congressmen: but thats ok, most Green's vote democrat because they are ideologically similar. If the newspaper actually slants economically-libertarian, they will look conservative, as more conservative congress than democratic congress will be citing the same newspaper. Thats the beauty of the system, it just compares to sitting congresspeople, so it is only splitting dem/conservative because those are the sitting congresspeople.

If we were to choose only left/hard-left we would have to ignore half the congress, thats hardly the same thing.

>Every article I read about that study began by stating the study challenged economic orthodoxy and the results of previous studies, IMO that is sufficient.

So, if an article quoted it they should explain that this was the *one* study that challenged orthodoxy? Is that what newspaper articles are supposed to do? "The classical thinking is such and such, but one group from Berkelely was able to prove everyone else wrong... hence it makes sense that these poor poor folk would want to pass this law."

>The study was unique since it illustrated a rare 'controlled experiment' where employment could be compared in to geographically close and economically similiar areas whose only difference was the change in policy.

Really. Because I know of many such studies.

This one has a good control group of cities, it compares cities where the law passed and ones where it didn't pass and found higher unemployment after passages:
http://econpapers.repec.org/paper/izaizadps/dp1566.htm

This is a summary paper looking at many studies:
http://econpapers.repec.org/paper/nbrnberwo/10562.htm

>Wait a second! Let's leave aside the odd assertion that NAMBLA is a left wing group.

I remember seeing them at lefty protests. They were hip-to-hip with the socialists. Its true, not every lefty liked them, but they found a lot more friends among the left wing groups and the gay marches then they ever could find on the right.

>There's probably more interesting patterns going on there than simple bias studies though.

Sure there are other ways to use citations, but I do think its important and interesting to see the ways that papers can be biased - how else can we begin to see how our worldview is shaped? Did we know in the fifties what our assumptions were and where our media was biased? The sixties? The seventies? How has it changed, when was it most biased? How might our media have shaped our thinking versus our shaping of what we see and what gets written in the media? Is a small section of thinking writing their viewpoint into the mass media, swaying public opinion based on assumptions most of us don't hold?

Boonton writes:

Because we live in a two-party country at present. If newspapers slant any which way they will primarily slant toward one or the other of the two parties - if you stack the congress with Libertarians then the study will automatically start comparing things to Libertarians! Imagine that! Anyway, that is where all the money is so newspapers will tend to pander to them and with a few execptions the split is pretty easy.

This doesn't seem right at all. Suppose libertarians have a lot to say about a particular issue? They don't get representation in a story until their citation ranking increases in Congressional speeches? This might be objective but what you're doing is eliminating media bias by establishing a quota system to citation.

I remember seeing them at lefty protests. They were hip-to-hip with the socialists. Its true, not every lefty liked them, but they found a lot more friends among the left wing groups and the gay marches then they ever could find on the right.


I learned long ago that the types that typically go to protests tend to be the more wacko sort. After 9/11 a friend and I made it a point to go to anti-war protests to heckle them (in a nice way of course).

liberty writes:

>This doesn't seem right at all. Suppose libertarians have a lot to say about a particular issue? They don't get representation in a story until their citation ranking increases in Congressional speeches? This might be objective but what you're doing is eliminating media bias by establishing a quota system to citation.

Put more libertarians in congress if you want them to be on our scale, okay?

No. What its doing is simply saying that given that most likely any citations that indicate bias are going to be found as well in congress, then bias can be measured by comparing the newspaper citations with congress citations. In general any major issue will have several think tanks from various sides writing reports on it and both congress and media will have to choose among them. If a popular media outlet consistantly used un-popular, out of the mainstream think tanks, they would tend to be seen as biased already - hence the top 200 think tanks should do. If a paper consistantly uses the same ones that the republicans in congress use then they will be considered biased to the right. If they consistantly use the ones the democrats use, they will be considered biased-left. Again, citations that argue against the think tank statement or which preface it by saying that its a really conservative think tank so take it with a grain of salt, are ignored. Only citations that use it without preface to back a point are considered, both for congress and for the papers. So, basically, if by this measure you sound like Dennis Kucinich, then you're a red. If you sound like Jeb Bush, you're right wing.

liberty writes:

>I learned long ago that the types that typically go to protests tend to be the more wacko sort. After 9/11 a friend and I made it a point to go to anti-war protests to heckle them (in a nice way of course).


Oh, I did not mean to imply that most on the left like them, but 100,000 students and hippies of the left isn't a fringe to be completely ignored either. I mean, ignore them, make fun of them, more power to you but it explains why NAMLA is on the left-fringe not the right-fringe.

Boonton writes:

Put more libertarians in congress if you want them to be on our scale, okay?

Quite frankly no. If the scale is objective then I shouldn't have to campaign for anyone to make it work better.

Israel is a good issue but an even better one is drug policy. For the most part both Republicans and Democrats seem to favor keeping drugs illegal and fall over one another in coming up with ways to make the drug laws harsher. An article that is perfectly unbiased from this scale could leave out entirely the position of many libertarians that drug policy has been a total failure.

This leads to another interesting problems. Objectively the libertarians are unbiased in that statement. US drug policy has failed to achieve the goals it has stated for itself. Therefore it is a failure. That statement, believe it or not, is an unbiased fact. It doesn't mean that drugs should be legalized or even that the policy should be changed. It is just a statement of fact.

The problem here is that you are assuming Congress is perfectly balanced netted together. If you don't sound like a Republican or a Democrat then, hurray!, you're unbiased. But are both parties always unbiased? Isn't it possible for, at times, for Dems or Republicans to be less biased than normal on a particular issue?

liberty writes:

>Quite frankly no. If the scale is objective then I shouldn't have to campaign for anyone to make it work better.

The point is that it finds bias based on the two major parties! Nothing is perfect. Most major newspapers are not going to bias libertarian. If you only catch 99% of newspaper bias, is it a complete failure?

>US drug policy has failed to achieve the goals it has stated for itself. Therefore it is a failure. That statement, believe it or not, is an unbiased fact. It doesn't mean that drugs should be legalized or even that the policy should be changed. It is just a statement of fact.

And as a statement of fact it will come across as unbiased - if you use statistics found in the government agency or some other institution that collects facts, not a think tank. However, if you go to the CATO institute for that particular fact, you'll be seen as right wing, if you go to EPI, you'll been seen as left wing.

>The problem here is that you are assuming Congress is perfectly balanced netted together.

No I'm not. I simply use tem as a scale to see which way you may lean. If you use no think tanks you are also counted as unbiased. I would usually tend to sound unbiased as I quote the census, bls, oecd, academic papers, etc. Then sometimes I will be caught sounding right wing if I quote Heritage or something. Stick to the facts and you'll sound unbiased. Quote one of the leading top 200 think tanks from right or left without giving the balancing opinion or stating clearly that they are partisan and you'll be lavelled as biased. Makes sense.

> If you don't sound like a Republican or a Democrat then, hurray!, you're unbiased.

Well, it is a tool to look at bias in media and congress, if used on pure editorial it would miss the obvious bias that doesn't bother to cite think tanks but just spews muck and opinion. But then each tool has a particular job.

Boonton writes:

And as a statement of fact it will come across as unbiased - if you use statistics found in the government agency or some other institution that collects facts, not a think tank. However, if you go to the CATO institute for that particular fact, you'll be seen as right wing, if you go to EPI, you'll been seen as left wing.

Now I don't doubt that this is true using the method the study tried but is this true using the common sense approach? For one thing gov't agencies might bury the statistics or make them almost impossible to find unless you have the time and energy to do it manually. So assume that the reporter does not have the skills or the time to compile the statistics from raw gov't data but both think tanks do. So he uses one of their reports (not both because space is usually at a premium).

The system is going to buzz 'bias!' but in reality common sense tells us there is no bias there.

liberty writes:

> So he uses one of their reports (not both because space is usually at a premium).

>The system is going to buzz 'bias!' but in reality common sense tells us there is no bias there.

Does it? If I'm so lazy that instead of quoting the census or the bls, I go immediately to Heritage and quote them and don't consider saying that "the right wing think tank, Heritage" said it, and don't balance it with the EPI or anything, just quote Heritage as if they are an unbiased source, that shouldn't count as biased? What if I do it every day?

No, that's bias.

Boonton writes:

So if the Heritage foundation says 2+2=4 I have to write "the right-wing Heritage Foundation claims 2+2=4" or else I'm writing a biased article?

Also, doing statistics yourself is quite dangerous. Often it is better to use the work of people who do it for a living rather than doing it yourself, I think this applies to many reporters who math education is usually not that heavy. Additionally I note you implicitly seem to assume stats issued by gov't agencies are automatically unbiased.

liberty writes:

What you are missing is that think tanks are opinion makers and by definition have a viewpoint - otherwise they aren't called think tanks. State agencies like the census bureau are not affiliated with a party and don't write opinion articles, and if you cite them you cite their factual statistics and factual reports, so yes this is unbiased. I don't suggest that to be unbiased it must be a government agency - and some by their nature (like the EPA) may be biased because to stay in existance they must report certain things (we wouldn't need an EPA if there is not threat to the environment, we would still need a census and bureau of labor statistics reagrdless of what they report). You could cite any other statistic gathering agency. And you need not be a statistician to use their reports, they are very straightforwad. If your story is on the minimum wage and you want to be unbiased, you can do some of the following:

1. Cite a major economics textbook regarding agreed upon history of the contraversy / theories / economic orthodoxy.

2. Cite two opposing think tank opinions

3. Cite two opposing recent econometric papers if you come across two, none if you don't find two unless there is a clear consensus that the one is well agreed upon by both opposing think tanks

4. Cite BLS data on who tends to make use of the minimum wage (http://stats.bls.gov/cps/minwage2004.htm)

Both any reasonable person and this approach to catching bias would see some combination of those four as unbiased. A good article would do #1 and #4 and either #2 or #3 in my opinion. Your article above cited a paper that was not agreed by both sides as being the "new truth", but was just one side, and there are many papers that should have been seen as an equally good "other side", so in my opinion that article was un balanced and should be seen as biased.

Comments for this entry have been closed
Return to top