Comments on Pew Polls Cell Phone Users

Even if the added support from cellphone users doesn't break out of the margin of error, that doesn't make it statistically insignificant, if it's always in the same direction.

http://www.fivethirtyeight.com/2008/09/estimating-cellphone-effect-22-points.html

538 figured it's worth 15 electoral votes. Sounds pretty fucking significant to me.

Have they bothered conducting a poll to show the influence of new voters? Those voters could make this a blow out for Obama.

ok. but does the new study isolate cell phones not owned by landline owners? in other words, is it possible that many of the newly polled cell users are also landline owners, and thus more likely to vote for mccain? arent the "obama" cell users the younger ones who don't also have landlines? without isolating cell-only from cell-and-landline, you tend to test the landline sample more than once.

right?