Of the Kansas candidates scored by Americans for Prosperity, eleven were incumbents targeted by the Kansas Chamber of Commerce. The highest AFP “budget, spending and taxes” rating in the group, according to Project VoteSmart, was Deena Horst’s 60. Horst was defeated by four percentage points and Jill Quigley lost by six, but the other nine held their offices. Beyond that, the race for the highest office available went to Jerry Moran, who campaigned for Sam Brownback’s senate seat effectively by moving as far to the right as Todd Tiahrt dared drag the race, but is nevertheless widely considered the more moderate of the two Republicans. #
Yet Bob Weeks’ blog, the voice of the hard right in Wichita, focused on the Wichita Eagle’s endorsements. The Eagle favored Moran for the Senate seat being vacated by Brownback, who will become the state’s governor this fall, but also picked against primary winners Tim Huelskamp and Mike Pompeo. Weeks notes a mixed bag of accurate and inaccurate picks among the Eagle’s endorsements (neglecting the logical point that the Eagle’s endorsements are a form of favor rather than prognostication) and attempts to draw a conclusion that does not appear warranted by the data he cites: #
Voters ignored many of its endorsements, indicating that the newspaper — its editorial side, at least — is increasingly out of touch with its readers. #But given that the Eagle’s endorsement coincided with Moran’s success in the state’s most important race, by the same token, is it not appropriate for the reasonable reader to consider that the conservative hardliners may be as “increasingly out of touch with voters” as any other group of policy advocates? #
you wrote “neglecting the logical point that the Eagle’s endorsements are a form of favor rather than prognostication”
But when I read Bob’s article, in the second paragraph it read “An endorsement is a recommendation to voters, and not intended to be a prediction of the outcome.”
That is true, but I think I phrased my point fairly and clearly.
There is a logical gap between the premise “Yesterday’s primary election in Kansas provided a measure of the influence of the Wichita Eagle editorial board” and the conclusion that “the newspaper — its editorial side, at least — is increasingly out of touch with its readers.”
Mr. Weeks attempts to fill that gap with “voters ignored many of its endorsements” — a point he then attempts to prove by examining outcomes of various elections. Although he attempts to suggest otherwise, Mr. Weeks *is* measuring endorsements with the predictions yard stick in his post.
He acknowledges the flaw but doesn’t take care of it, and I think “neglecting the logical point…” is an apt description.
But even if we assume that the method he applies is a reasonable way to evaluate whether “voters ignored many of its endorsements,” the data he produces does not actually indicate that “many voters” ignored the Eagle’s editorial board. Qualitatively, his analysis is a mixed bag, and it is anything but quantitative.
But even if we grant the point that “many voters” ignored the Eagle editorial board, we’re left with a conclusion that does not speak to the Eagle’s ability to influence elections. [EDIT: Of course, whatever influence it exerts would be with and through voters—if even minimally significant numbers of voters are influenced by the Eagle, that would indicate the newspaper is in touch with an important part of the public, would it not?.]
Consider the race for Brownback’s seat: if the Eagle editors held sway over even 7,000 votes (~2.2 percent of the total) then the newspaper’s view might be considered decisive. If we then take the same 7,000 votes away from Jean Schodorf, she holds ~14-15 percent of the total in her race. This coincides with poll results from the days and weeks before the Eagle published its endorsements.
Of course, there are factors besides the Eagle’s picks holding influence over voters (making a formal attribution difficult or impossible). But I’d argue that meaningful data can be gathered easily (from among Mr. Weeks’ own posts!) by comparing how the final totals compared to poll results from before the Eagle published its endorsements.