This Could Be Big

October 13, 2010 3:06 p.m.

This might be a pretty big deal.

For years people have written in to TPM to ask: must it not skew the polls that most pollsters only use landlines and not cell phones? Especially since younger, more transient voters tend to skew more Democratic?

My answer has always been: not yet.

In principle, yes, it’s a big blind spot for current polling methodology. And pollsters have been watching it closely — and nervously. In practice though most studies have shown that the difference is either trivial or very small. Pew is one of the pollsters that’s been systematically sampling this question going back a few years. As recently as the 2008 cycle the difference was just over 2 percentage points. But according to a study just released by Pew, whether or not pollsters sample cell phones is producing “differences of four to six points in the margin” in favor of the GOP. Pew is basing this conclusion on four comparisons. Three of the four polls had differences in that range. One poll showed no difference.

I would strongly recommend that Democratic enthusiasts not start adjusting the polls they’re seeing by a 4-6 point margin. I will, shall we say, believe this one when I see it. It’s also important to note that a number of national pollsters are already incorporating cell phones. Where the real vulnerability comes is in state and district polls and robo-pollsters — like Rasmussen, PPP and SurveyUSA.

Again, in the past the differences usually seemed too small to figure significantly into the prediction equation. But this could be the cycle where that changes.

We’ll soon find out.

Latest Edblog
Masthead Masthead
Founder & Editor-in-Chief:
Executive Editor:
Managing Editor:
Senior Editor:
Special Projects Editor:
Editor at Large:
General Counsel:
Head of Product:
Director of Technology:
Associate Publisher:
Front-End Developer:
Senior Designer: