tag:blogger.com,1999:blog-2569772432953120875.post7483317370963210773..comments2024-02-22T19:45:29.872-05:00Comments on Sources And Methods: How Accurate Is Your Pundit? (Hamilton.edu)Kristan J. Wheatonhttp://www.blogger.com/profile/02566135545863154089noreply@blogger.comBlogger3125tag:blogger.com,1999:blog-2569772432953120875.post-48159920314275807872011-05-06T19:55:28.976-04:002011-05-06T19:55:28.976-04:00Nancy Pelosi a "Good" prognosticator?
...Nancy Pelosi a "Good" prognosticator? <br /><br />Nancy Pelosi, Oct, 2010: "I have every anticipation that we will come together...with me as speaker of the house," after the election. <br /><br />The inherent biases of this study were clearly evident when they coded Sam Donaldson as "conservative." <br /><br />Sam Donaldson: <br />"The problem Republicans have, so many of them are sanctimonious."<br /><br />But the bias of timing the data gathering period as the anti-Bush election cycle of 2008 is almost beyond comprehension. <br /><br />Comparing Republican "predictions" during the run-up to the 2008 election, of course, resulted in Republicans being mostly wrong. And vice versa for Democrats in that election. <br /><br />If the data had been collected during the run-up to the 2010 election, the results would have been exactly opposite. Nancy Pelosi's partisan posing would have been scored negative, and George Will's more conservative predictions would have been more correct. <br /><br />The first thing an analyst has to do is to examine their own biases. <br /><br />This "study" scores "fail" for biased selection of time period to collect data. <br /><br />The fact that the study is a product of Hamilton College is even worse for its bias score. Hamilton College is notorious for its actions taken against a couple of non-liberal professors: <br />http://www.nas.org/polArticles.cfm?doctype_code=Article&doc_id=1268<br /><br />Maybe a better title for their study would be: "How Biased is your Professor?"Kent Clizbehttp://www.kentclizbe.comnoreply@blogger.comtag:blogger.com,1999:blog-2569772432953120875.post-84617516694885437292011-05-06T10:33:06.861-04:002011-05-06T10:33:06.861-04:00Andrew,
We did not do this study at Mercyhurst. ...Andrew,<br /><br />We did not do this study at Mercyhurst. It was done by students at Hamilton College.<br /><br />The students indicated that their aggregate findings were generally consistent with Tetlock's. <br /><br />They, however, looked at only a slice of the estimative conclusions (those circulating around in advance of the 2008 elections) of the various pundits. <br /><br />it is entirely possible that Krugman was only good/lucky in that specific circumstance or for those specific questions. <br /><br />KrisKristan J. Wheatonhttps://www.blogger.com/profile/02566135545863154089noreply@blogger.comtag:blogger.com,1999:blog-2569772432953120875.post-34020381734158856522011-05-06T10:27:07.876-04:002011-05-06T10:27:07.876-04:00Interesting. I interpreted this as counter-Tetlock...Interesting. I interpreted this as counter-Tetlock (of course I haven't seen your entire study). I consider Krugman a pretty intense expert on the subject of economcs (PhD, Nobel, etc.)...so why does his expertise rank so much higher than other "generalists"? <br /><br />I could be wrong, but I think he was even mentioned specifically in Tetlock's book as an example on an expert who wasn't necessarily better at predicting than others...Andrewnoreply@blogger.com