On the NSA and Media Bias: An Extended Analysis
By Albert Wong and Valerie Belair-Gagnon, Information Society Project at Yale Law School
In a recent article in the Columbia Journalism Review, we reported that major US newspapers exhibited a net pro-surveillance bias in their “post-Edward Snowden” coverage of the NSA. Our results ran counter to the general perception that major media outlets lean “traditionally liberal” on social issues. Given our findings, we decided to extend our analysis to see if the same bias was present in “traditionally conservative” and international newspapers.
Using the same methods described in our previous study, we examined total press coverage in the Washington Times, one of the top “traditionally conservative” newspapers in the US. We found that the Washington Times used pro-surveillance terms such as security or counterterrorism 45.5% more frequently than anti-surveillance terms like liberty or rights. This is comparable to USA Today‘s 36% bias and quantitatively greater than The New York Times‘ 14.1% or the Washington Post‘s 11.1%. The Washington Times, a “traditionally conservative” newspaper, had the same, if not stronger, pro-surveillance bias in its coverage as neutral/”traditionally liberal”-leaning newspapers.
In contrast, The Guardian, the major UK newspaper where Glenn Greenwald has reported most of Snowden’s disclosures, did not exhibit such a bias. Unlike any of the US newspapers we examined, The Guardian actually used anti-surveillance terms slightly (3.2%) more frequently than pro-surveillance terms. Despite the UK government’s pro-surveillance position (similar to and perhaps even more uncompromising than that of the US government), the Guardian‘s coverage has remained neutral overall. (Neutral as far as keyword frequency analysis goes, anyway; the use of other methods, such as qualitative analysis of article tone, may also be helpful in building a comprehensive picture.)
Our extended results provide additional context for our earlier report and demonstrate that our analysis is “capturing a meaningful divide.”
On a further note, as several commenters suggested in response to our original report, the US media’s pro-surveillance bias may be a manifestation of a broader “pro-state” bias. This theory may be correct, but it would be difficult to confirm conclusively. On many, even most, issues, the US government does not speak with one voice. Whose position should be taken as the “state” position? The opinion of the President? The Speaker of the House? The Chief Justice? Administration allies in Congress? In the context of the Affordable Care Act, is there no “pro-state” position at all, since the President, the Speaker, and the Chief Justice each have different, largely irreconcilable views?