Adding cell phones to a poll gives you a five point swing toward the Democratic candidate.
A large question going into this year's midterm elections was whether polling agencies who only called landlines were missing a key group of voters: people who only used cell phones. What made this more important was the second hypothesis, that people who only have a cell phone tend to be more liberal.
Now that election results are in, we can check the math. Anecdotally, we saw a few Democrats (most prominently Harry Reid in Nevada) who seemed to outperform their polling numbers. Now Pew—which published a pre-election report on this subject—has found definitively that pollsters were missing a lot of Democrats by not calling cell phones:
In Pew Research's final pre-election poll in 2010, the landline sample of likely voters found Republican candidates ahead 51%-39%, a 12-point lead. In the sample that combined landline and cell phone interviews, the Republican lead was 48%-42%, a six-point advantage. The national vote for House candidates is not yet final; currently, Republicans lead by approximately a seven-point margin.\n
Here is the hard data. As you can see, adding cell phones gives you a five point swing in the Democratic direction.
Next election, as this trend gets stronger, be very suspicious of polls that don't include cell phones.