Latest Entry: The politics of Ebola     Latest Comments: Talk Back Here

« The Only Job You Can Trust Charlie Crist To Fight For ... | Main | USAID report: WH Spent $23M of Taxpayer Money to Back Kenyan Constitution That Legalizes Abortion »

July 21, 2010

Uh oh, Gallup caught lying?

Topics: Political News and commentaries

Yesterday Gallup came out with their somewhat stunning report that "Democrats had jump into a six-point lead on the Generic Ballot" causing Steny Hoyer to almost wet himself with excitement.

But Steny and his fellow Democrats should have read the fine print.

As Neil Stevens aptly points out over at Redstate, it appears that Gallup packaged its results in such a way as to deliberately mislead readers into seeing trends that just aren't there. - making it appear that Democrats have the trending advantage in the the Gallup generic ballot:

How? Gallup is combining two different sampling methods into one so-called trend of Registered Voters (complete with captioned graph), getting different results, but pretending they show one trend all the same.
Read more ...

While saying that Gallup purposefully mislead readers sounds at first to be a bit harsh, there's simply no other way to describe the reasoning behind the Gallup report. Gallup is far too experienced in polling to ever make such an error in sampling as mixing a report with "registered voters" and "adults." As Stevens goes on to note in his piece, no matter how you cut it - even if Gallup mis-labled the most recent report, the error suggests a serious lapse of quality control at Gallup leading to misinformation being published, and still calls into question every poll release Gallup makes. And yet, given that Rasussen's latest Congressional Ballot poll shows that Republican candidates now hold a nine-point lead over Democrats on the Generic Congressional Ballot for the week ending Sunday, July 18, the widest gap between the two parties in several weeks, for Gallop to come back and claim that their report was mis-labeled, Gallop would be saying that their polling results differ from Rasmussen's (poll of likely voters) by a total of 15 points (from a Dem advantage of +6 to a Republican advantage of +9), and the probability of this event is highly unlikely.

Posted by Richard at July 21, 2010 12:33 PM



Articles Related to Political News and commentaries: