cell phones

The Pew Internet Project has a new report out about the usage of smart phones by demographics and there are a few interesting things to point out.

According to the data about possession of a smartphone - 83% have some kind of cell phone. From those with cell phoes, 42% reported owning smartphones. Total that's 35% of all adults surveyed. The highest rate of smartphone ownership among demographics is ... as you would guess - those under 45 years old.

Interesting factoid: 28% of US smartphone users use their phone as their primary Internet connection. I wonder if this has anything to do with an increase in availability for mobile but not for high speed broadband access in rural areas. This could also have something to do with users who are less economically able to afford both mobile and broadband and would prefer to use both just on their phones.

"Smartphone owners under the age of 30, non-white smartphone users, and smartphone owners with relatively low income and education levels are particularly likely to say that they mostly go online using their phone."

Users also prefer Android phones over iPhone though iPhone users are beginning to slowly move up.

From the "you don't say" file:

Republicans received a boost over Democrats in landline-only telephone polls, according to a new study by the nonpartisan Pew Research Center.

Polls this fall that reached only landline phone numbers ended up giving Republicans an extra five percentage points compared with polls that dialed both landline and mobile phones, the study concluded. Pew said that’s because almost a quarter of U.S. homes don’t have a landline telephone number, and cell phone users tend to be younger and more Democratic.

This serves as a reminder to be especially critical of polls purporting to measure any kind of reality while failing to allow for representation of cell phone-only households in its results. Given the tendency of youth to forego landlines, we'd be doing ourselves a disservice to continue to peddle the polling garbage that is landline-only surveys.

One of the bigger points of contention this year in the (horridly geeky) world of polling was whether or not excluding cellphone-only users skewed the poll results. Throughout the cycle, PEW slowly but surely crept closer to saying "yes" to that question. Now that the results are in, they are finally making that judgment official.

According to a new report, polling that excluded cell-only showed a bias towards John McCain of 2.3 percentage points:

An analysis of six Pew surveys conducted from September through the weekend before the election shows that estimates based only on landline interviews were likely to have a pro-McCain tilt compared with estimates that included cell phone interviews. But the difference, while statistically significant, was small in absolute terms – smaller than the margin of sampling error in most polls. Obama’s average lead across the six surveys was 9.9 points among registered voters when cell phone and landline interviews were combined. If estimates had been based only on the landline samples, Obama’s average lead would have been 7.6 points, indicating an average bias of 2.3 percentage points. Limiting the analysis to likely voters rather than all voters produces similar results. Obama’s average lead among likely voters was 8.2 points across all six surveys versus 5.8 points in the landline sample.

When it came to young voters, the differences were much starker, though sample sizes make the numbers a little more fuzzy. Ultimately, polls that accounted for cell-only young people hit much closer to the final numbers among 18 - 29 year-olds on election night:

In Pew’s polling this fall, there was a gap of similar size in Obama’s advantage between cell-only young voters and those reached by landline, though this difference was not statistically significant given the relatively small sample of young cell-only voters. Among cell-only voters under 30, Obama led by 38 points (66%-28%); among those in the landline sample, Obama’s lead was 29 points (61%-32%).

Perhaps as a result of this pattern, Obama ran slightly better in Pew’s dual frame samples of young voters than in the weighted landline samples alone. As with overall voter estimates, the differences are small but statistically significant. Obama led McCain by 33 points (63%-30%) in the full dual-frame sample,
compared with his 29-point advantage in the landline sample.

The sample difference among likely voters under 30 was even larger. Obama led in the full dual frame sample by about 33 percentage points; in the landline sample his lead was 26 points. According to the national exit poll, Obama won this age group by 34 points, 66%-32%.

Lots of other juicy information out there for the data geeks about the demographics of cell-only, dual-use, and landline-only voters.

With all the polls in, Nate Silver at FiveThirtyEight takes another look at how those which included cell-phone samples fared against those that excluded cell-only users. Nate confirms what PEW found last month - that pollsters who excluded cell phone users consistently underestimated Sen. Obama's support:

The polls in the Cingular-y orange color include cellphones in their samples; the polls in gray do not. The cellphone polls have Obama ahead by an average of 9.4 points; the landline-only polls, 5.1 points.

We've been talking about this for at least a year here on Future Majority, but now Pew Research is ready to come out and say it. The number of cell phone-only voters are now numerous enough that their exclusion from traditional polling is skewing the data. A new report from Pew shows that in three straight surveys, lack of cell-only data skewed the survey results 2 - 3% in favor of John McCain.

From the report (emphasis mine):

The Pew Research Center for the People & the Press has conducted three major election surveys with both cell phone and landline samples since the conclusion of the primaries. In each of the surveys, there were only small, and not statistically significant, differences between presidential horserace estimates based on the combined interviews and estimates based on the landline surveys only. Yet a virtually identical pattern is seen across all three surveys: In each case, including cell phone interviews resulted in slightly more support for Obama and slightly less for McCain, a consistent difference of two-to-three points in the margin. [...]

As implied by these results, in each of the three polls, the cell-only respondents were significantly more supportive of Obama (by 10-to-15 percentage points) than respondents in the landline sample. For example, in the September survey Obama led McCain by a 55%-to-36% margin among cell only voters, but the candidates were tied at 45% in the landline sample.

In large part, this reflects the fact that a substantial minority of the cell-only sample is younger than 30 - a demographic group that has consistently backed Obama this year. 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.

As the chart on the right shows, cell-phone only voters under the age of 30 are substantially different in their identification with the Democrats and their support for Obama than are their peers with landline access. Young voters who rely solely on their cell-phones, and thus are often excluded from polls, are far more supportive of Obama and the Democratic Party than are their landline counterparts.

The implications are clear: Obama's youth support, already underrepresented in polls that screen for "likely voters," are further underrepresented due to their phone preferences. As a result, it's not unreasonable to look at the polls that exclude cell-phone samples and compensate for that bias by adding a point to Obama's total and removing a point from McCain's.

For those of us supporting Obama (and biting our nails in recent weeks) that's an encouraging thought, but it also begs a question. This problem isn't going to go away. In fact, it is only going to get worse. As the Pew report points out, cell-phone only voters are growing at a rate of 2% a year, and could be 17% of the electorate in 2008. As that population grows, will pollsters rethink their methodologies to accommodate that shift? And if young voters are being underrepresented due to their cell-phone habits and likely voter screens, what will that mean for the accuracy of the polls leading up to election day? Could the pollsters be as wrong in 2008 as they were in 2004?

First some youthy news:

• Two articles are out today on polling. The Michigan Messenger does a great job analyzing a PEW study on the demographics of cell-only and "cell-mostly" users and how pollsters are dealing with under-represtentation of these demographics in their polling. If that's a little too data-geek for you, Campus Politico has a good "polling 101" story that might be a little more your speed.
• At Tapped, Tim Fernholz questions the effectiveness of a voter registration drive launched this week by the Hip Hop Caucus.
• Matt Zeitlin at PushBack follows up on that post, noting that celebrities are not an effective way to move young voters to the polls.
• Teaming up with SPIN, CMJ, and others, Head Count has launched a 90 day voter registration challenge. They hope to register 100,000 voters by the end of the 90 days. You can watch a video of SPIN at the Warped Tour interviewing an artist about the program here.
• NDN notes Connecticut Democrats are out-registering Republicans among young voters by 4.3 - 1. That registration and turnout advantage might help Democrats defeat Republican Chris Shays in the fall. Shays is the lone Republican congressman in New England.
• South Carolina young Republicans are not feeling the love from John McCain, and Young Republicans nationally are having a tough time drawing young people to the convention.
• Meanwhile, Young Democrats are thriving in delegate-rich Florida.
• Tech President argues that 2008 is a boring year in tech/politics. While 2004 and '06 saw huge paradigm shifts in how we organize, 2008 is all about refining those practices.

Some less-youthy, but very interesting news:

• Democrats are getting ready to spend $20 million to register and GOTV latinos.
• The Nation writes about MoveOn's 10th birthday and how far the organization has come.
• A new study by the RAND Corporation tells us the obvious - to win the war on terror, we need to stop fighting the war on terror.
• Finally, the Washington Post has a front-page story basically calling "liar liar pants on fire" on the McCain campaign for their latest ad attacking Sen. Obama.

I've written in the past about the problems associated with polling young voters and how these are exacerbated by our habits of not owning a land-line. This is a growing problem for pollsters as more and more people abandon their landline and go cell-phone only.

This week, Gallup took a huge step forward in addressing these problems when they announced plans to begin including cell phones numbers in their surveys.

Still, Gallup has been studying and investigating the implications of cell phone only households for well over a year now. And, as of Jan. 1, 2008, Gallup has made the decision to include cell phone interviewing as part of the sample used for its general population studies.

This is a complex and costly modification in methodology. Our statisticians and methodologists have spent a great deal of time reviewing the procedures and implications of the change. Essentially, in addition to sampling from the traditional database of all landline telephone exchanges, Gallup now also adds in sampling from a new database of all cell phone telephone exchanges in the country. We screen for those individuals using cell phones who report not having a landline, and then interview a random sample thereof. We then weigh into the sample a proportionate percentage of these interviews conducted via cell phone.

We’re continually monitoring the methodology of our interviewing, and revise on a regular basis as appropriate. We'll be analyzing the implications of this shift in methods particularly carefully.

Mark Blumenthal has more on the consequences and significance of this change in methodology:

For now, at least, this change is not likely to produce dramatic differences in the results. The ongoing cell phone surveys conducted by the Pew Research Center have shown that the missing cell-phone-only population rarely makes a difference of more than a point or two. But that point or two may sometimes make a difference, especially in a close race. Consider last week's Gallup poll in New Hampshire. USA Today polling editor Jim Norman let us know, via email, that they included a cell-phone sample on that survey:

[I]t added a point to Obama's total and took one away from Clinton. In other words, without the cell-phone-only respondents, Obama's lead among likely voters was 11, not 13.

The bigger significance in this change is symbolic. Gallup is the granddaddy of all polling firms. Their polling "time series" goes back to the 1930s. As such, they are typically the most cautious about changes in methodology, so their move to regular cell-phone sampling is likely to have a big ripple effect on the polling industry. At very least, this most closely watched poll will provide a regular source of data on the potential impact of the cell-phone-only households that will be missing from other surveys.

If Gallup continues to adopt this as their standard, other polling firms will follow, and the "added costs" of collecting a large enough sample of younger voters will instead become the normal costs of doing business. Maybe then we'll see more polling that can reliably break out the opinions of young voters, which currently are subject to wide margins of error or ignored altogether.

If Gallup begins to show statistically different findings from other polls, you can be sure that this will happen.