Tricky: Gallup Mixes Polling Methods, Shows False Dem Resurgence

Posted: Jul 22, 2010 8:02 AM
Like every other person who looks at polling data on a daily basis, I was pretty shocked when Gallup published these results on July 19, purportedly showing Democrats with a recent surge in the polls.  Political analysts on TV were baffled as to why Republicans had seemingly squandered a six-point lead on the generic congressional ballot--a full 12-point swing in favor of Democrats in just two weeks. 

An examination of Gallup's polling methods is now shedding light on the unwarranted swing in public opinion.  The earlier poll which showed a Republican advantage polled registered voters, while Democrats lead among a wider sample of all adults

However, Gallup doesn't point out this subtle difference in polling methods, leading us all to believe that Democrats had actually made quite a resurgence in the polls. 

It seems a bit dishonest for Gallup not to differentiate between the two polls to and present the results as ongoing trends when the different data sets inevitably lead to different statistical outcomes. 

RedState suggests that Gallup is lying, purposefully misleading with their polling methods.  I'm not sure if I'd go that far--we'll see what their next poll looks like.  But I'll readily admit this latest episode seems like trickery and agree with RedState's final assessment:
You cannot pretend that a poll of all adults and a poll filtered by registered voters are part of the same series, even if the same questions are asked. That’s Polling 101, and whoever’s responsible for the Gallup release should have known this, and certainly whoever’s responsible for oversight of the Gallup releases would know this.