After I read the Groseclose/ Milyo study, I wondered if there was a way that we Googlers could assess the existence of bias in the news output. I came up with the following vastly less precise method. My hypothesis was this: most journalists think they are moderate, thus institutions to the right of their perspective are "conservative". If this is true, then the oft-cited Brookings Institution will rarely be flagged as "a liberal think-tank" by reporters - because they don’t think it’s liberal, though generally conservatives do. Similarly, AEI and Heritage should be frequently flagged with a cautionary label like "conservative".
One can crudely test this by using Google News to search their news-only database for the specific phrases, limiting the sources to USA to more closely approximate the Groseclose/Milyo study sample. This is a much bigger sample of news outlets, which necessarily will include mostly local or regional papers. I would expect these "not mainstream" outlets to have a less liberal bias than the twenty outlets discussed above (I’ve not seen a study that confirms that point).
Here are my search results, which you can easily repeat by running each search string in Google News Search (NOT Google, that won’t work). You will certainly get different results because the scope of the news search is constantly changing, with only 30 days archives available - but I’ll bet the ratios you get are similar:
e.g., <"liberal brookings institution" location:usa>, or <<"conservative Heritage Foundation" location:usa>
"Brookings Institution" 1590
"liberal Brookings Institution" 2 = 0.125%"Heritage Foundation" 1500
"conservative Heritage Foundation" 312 = 21%
Dang! Does that mean whenever a reporter cites a conservative think tank she feels about 150 times more compelled to warn the reader? I had better write this all up for the New York Times! Unless you think they won’t want to publish it?
0 Responses to “Seeker Blog Proves Media Bias!”