Cell phone bias

As Nate Silver points out today, Obama is really winning the race quite handily in the polls:

Here's the long and short of it for John McCain: Barack Obama has as
large a lead in the election as he's held all year. But there is much
less time left on the clock than there was during other Obama periods
of strength, such as in February, mid-June or immediately following the
Democratic convention. This is a very difficult combination of
circumstances for him.

On the strength of a set of national tracking polls
that each show Obama at or near his high-water mark all year, our model
projects that he would win an election hold today by 4.2 points. It
discounts this lead slightly to a projected margin of 3.3 points on
November 4, as most races tend to tighten as we approach election day.

lead might not sound like that much, but it's fairly significant: we've
been through two conventions and one debate, voters have dug their
heels in, and Obama's position in the Electoral College is extremely
robust. Trimming away a 4-5 point lead isn't that difficult over the
summer months — in fact, McCain accomplished exactly that in July and
August — but it's a steeper hill to climb after Labor Day.

Here's the thing… as well as Obama is doing in recent polling, it may actually be understating his strength.  Generally speaking, pollsters only poll those with landlines.  People who rely solely on cell phones (predominantly younger Americans), do not get polled.  Pollsters are not stupid– they adjust by statistically weighting the sample to account for the fact that they are reaching fewer younger voters than they should.  Such adjustments should work fine, if one assumes that the young people with landlines are not significantly different from those with cell only.  It increasingly appears that when it comes to Obama, cell-only users are disproportionately strong supporters and that simply weighting the sample does not solve the problem.  Mark Blumenthal at Pollster.com nicely summarizes a recent Pew report on the matter:

Pollsters have long understood
that the cell phone only population — those who have cell phone but no
landline telephone service — tend to be younger, and that the growth
of that population has made it more difficult to reach 18-29-year olds.
However, the conventional wisdom among pollsters has held that
weighting by age could mostly alleviate any potential bias, as they did
they did in 2004.

The new Pew report shows why weighting by age may not have the same effect now:

Traditional landline surveys are typically weighted
to compensate for age and other demographic differences, but the
process depends on the assumption that the people reached over
landlines are similar politically to their cell-only counterparts.
These surveys suggest that this assumption is increasingly
questionable, particularly among younger people. […]

the pooled [August-September] data, cell-only young people are
considerably less likely than young people reached by landline to
identify with or lean to the Republican Party, and even less likely to
say they support John McCain. Among landline respondents under age 30,
there is an 18-point gap in party identification – 54% identify or lean
Democratic while 36% are Republican. Among the cell-only respondents
under age 30, there is a 34-point gap – 62% are Democrats, 28%
Republican. The difference among registered voters on the horserace is
similar: 39% of registered voters under 30 reached by landline favor
McCain, compared with just 27% of cell-only respondents. Obama is
backed by 52% of landline respondents under 30, compared with 62% of
the cell-only.

Long and short of it, by not polling cell phones, typical polls may actually be underestimating Obama's support by about 2% on average.  Not a gigantic number, but it certainly makes his current lead seem all the more imposing.

%d bloggers like this: