I read Bernard Goldberg's Bias on a recent business trip to Washington DC.  Overall, the book was a disappointment.   It's the personal rant / memoirs of a single ex-reporter and was consequently often lacking in sophistication of analysis.  I was looking for interesting structural theories about what generates press coverage rather than a few anecdotes here and there.   He does have a few, but overall, Bias didn't satisfy.

The book starts with the following central thesis [p 1]

...the media divide Americans into two groups -- moderates and right-wing nuts.

According to Goldberg, we've got a media elite at the top who's generally far more liberal then the American public at large.   They're more liberal than self identified American "conservatives" -- that shouldn't be a surprise -- BUT they're also more liberal than many American "liberals".   Perhaps more importantly, media elite peers tend to be uniformly liberal and hence consider this "normal" and have a hard time seeing otherwise.   Bernard argues that this "star power" goes to the media's head [p 8]

Journalists, you see, think they're better than "regular people."   Smarter than them, more sophisticated, fairer, more sensitive -- things like that.

Goldberg is using a rather traditional definition/mainstream definition of the political spectrum.   But, nevertheless, he argues this liberal bias is pervasive [p 9]

...that bias in the news is not mainly about politics -- about going easy on liberal politicans and hard on conseraties -- but about how the media elites report on all the big social and cultural issues of our time, everything from feminism and abortion to race and affirmative action, to gay rights, to the homeless.

and is reflected in party affiliations [p 130]

what party do journalists identify with?

• 50 percent said they were Democrats
• 4 percent said they were Republicans

Goldber presents examples from

• AIDS coverage (going out of their way to portray it as a suburban, heterosexual problem which is only a tiny % of cases)
• Homelessness (coverage of which varies based on which party is in office)
• Commentator "warning labels" (conservative guest speakers are prominently identified as such while liberal guest speakers are merely, well, experts)
• Coverage of Afro-American crime (going out of the way to portray victims and commiters of crime as anything but Afro-American when possible)

I generally agree with many his narrow observations but there are several avenues I wish Goldberg had explored:

What kind of people tend to be attracted to Journalism as a profession?  For ex., of the folks in Journalism grad school, what sorts of undergrad degrees did they hold?   How well represented are "conservative" bedrock skills like economics and business within the journalist corp?  How does the process of rising from cub reporter to star anchor / producer manipulate this individual?   I would imagine that you rise farther faster by reporting on market failures than on market successes.  

I'd argue that the process of looking for "hits" in the news often leads to an outsize focus on narrow, identifiable beneficiaries (e.g. Microsoft) or narrow, identifiable cost-bearers (e.g. homeless) without proper weighting / discussion of the distributed counter parties (e.g. all the folks who get cheap, usuable PC's, or all the tax payers who provide $$$ via the govt). 

Soundbite reporting leads to simple/fast args vs. deeper args.   The headline would read "Bush wants to lower taxes for the rich" as opposed to "nearly every economist agrees that dividend taxes are double taxes and most countries don't have them for that reason"   (this argument, of course is often used in the "conservative" / religious right direction as well)

For example, Arnold Kling's "Economic Idiotarians" article does help us generate a structural theory for why so many journalists might hold a "liberal" bias.   *If* they lack formal economics training, then they could be intuitively prone to see allocations of resources as the result of a Command or Authority based ranking rather than a price based scheme.   In a price based scheme, the Journalist's role is integrated into the market as one of many voices -- he/she provides more data to the consumer to address an individual choice.  

By contrast, the assumption of an Authority or Command scheme places the role of  journalist on a higher pedestal.   The Journalist thinks he/she is simply providing a factual, non-biased article about the outcome of an authority / command allocation to feedback into future command allocations.  In actuality, the bias being displayed is the original assumption that Authority was responsible for the allocation in the first place!  (or, in a Hayek-esque way, the creeping demand that Authority -- which is more accountable to journalists -- be brought *in* to affect resource decisions displacing price mechanisms)

Shades of this can be seen in Goldberg's description of homelessness coverage in the major media [p 79]:

... a column about the homeless and about a village voice study that showed that in 1988 the New York Times ran fifty stories on the homeless , including five on page one.   But a decade later, in 1998, the Times ran only 10 homeless stories, and none on page one.   ...   in 1990, when George Bush was president, there were seventy-one homeless stories on the ABC, CBS, NBC, and CNN evening newscasts.   But in 1995, when Bill Clinton wa sin the White House, the number had gone down to just nine!

The media apparently *wants* to run the story connecting the President with a rise in homelessness (rather than underlying economics).   But here, Goldberg falls short.   I intuitively believe this is the case (perhaps I *want* to believe it).   But Goldberg never presents the key data to back up the story:  Did the rate of homeless really stay stagnant across these administrations?   Over a 10 year period, intellectual honesty forces me to recognize that there could have been some large, underlying swings in the rate of homelessness (or at least the derivative -- the rate of of growth of homelessness).   Goldberg's case would have been truly bolstered if he provided this data.  But he doesn't.

---

UPDATE: Great example from Instapundit -- the type of stuff that Goldberg needed to put more of in his book.

UPDATE2:  Another great quote - this time from Smallest Minority who's quoting an OpEd in Fox:

The majority of young men and women who enter journalism do so not because they want to report the news but because they want to make a difference in society. In other words, they want to report certain kinds of news. They do not want to convey facts or explain processes; they want to shine spotlights on abuse. In some cases they are motivated by idealism; in others, by the hope that some of the light will reflect back on them.

---