We don't need it, do we

Heard about an interesting study by two economists, Jesse Shapiro and Matthew Gentzkow, on NPR this morning. I've read other papers Gentzkow has worked on, so I was predisposed to have a favorable view of this one. They compared the use of politically slanted phrases in newspapers with the political views of the populations served by the paper. Their conclusion was that papers slant their political language to maximize newspaper sales.

While I think it's a great study and I think they did a thorough job demonstrating a real correlation between the two it is a logical fallacy to state causation. I've always believed that newspapers, and most other news media, are driven primarily not by a political viewpoint but rather by the desire to make a profit. And I've long argued that any liberal bias the media* may have is motivated by the desire to sell more papers.

I don't think this study demonstrates whether the chicken or the egg came first. It is, however, a great step towards making that case and I expect to see more strong work from the authors in confirming it.

With exceptions for cases as in the early 90's when Rupert Murdoch freely admitted that he regularly interfered with the political slant of his properties.

*The liberal media bias claim is most often used in reference to newspapers, as there is demonstrably a conservative media bias to tv and radio news.

Listening to: Who's side are you on? - Decompopsure

Del.icio.us | Digg | Technorati | Blinklist | Furl | reddit


comments

rev says: Got a note from Gentzkow expanding on their causation data in the Robustness section. I am much more convinced, though the math is beyond me. I didn't read the robustness section thoroughly enough the first time through, it is definitely there. So in liberal newspaper markets there is a liberal slant, in conservative ones there is a conservative slant.
Posted at: 2006-12-13 10:58:35



add a comment

< /p>

(will not be displayed)