Manipulating polls?

We’ve all seen the media reports trumpeting the supposed sudden lead for McCain/Palin in opinion polls. An increase in polling would be expected, if there was a “post convention bounce”. But the polls are not quite what they seem.

It turns out that all the major polling organisations have decided to increase the percentage of poll respondents who are Republican, despite the fact that there are more registered Democrats than Republicans. Polling companies also don’t poll new voters, since they don’t turn up on voter lists, and Democrats have a huge advantage in new registrations. They also don’t poll people who have only mobile phones (and no landline), many of whom are younger and tech-savvy—which describes a good chunk of the people who were attracted to Barack Obama’s campaign in the primaries.

All of which means the poll results are highly suspect. According to the Huffington Post, “Emory University political scientist Alan Abramowitz is highly skeptical of the new Gallup, USA Today and CBS polls.” He points specifically to problems with the CBS poll:

"One reason for the dramatic difference between the two recent CBS polls is that the two samples differed fairly dramatically in terms of partisan composition. The first sample was 35.2% Democratic, 26.2 percent Republicans, and 38.6 percent independent. The second sample was 34.9% Democratic, 31.1% Republican, and 34.0% independent. That's a change from a 9 point Democratic advantage to a 3.8 point Democratic advantage. That alone would probably explain about half of the difference in candidate preferences between the two [CBS] polls."

So: Are the polling companies deliberately manipulating poll gathering? Or are they just making huge errors? Either case isn’t very encouraging, though the first raises serious questions.

I have no idea why opinion polls are being done so badly now, but it would help if the media obsessed about them less. That, or if journalists took statistics classes so they’d learn how to assess and evaluate polls then reporting could improve. You don’t need to take a poll to find out how unpopular both options would be in newsrooms across America.