Pollster.com

Articles and Analysis

 

The Evidence on Cell Phone Only Voters

Topics: Cell Phones , National Journal

My NationalJournal.com column for the week looks at what we know about whether the rise in cell-phone only households is causing any skew in polling results. The short answer is that polls that cannot reach cell-phone-only voters may be slightly understating support for Obama and overstating support for McCain, although the difference is small (and likely within the margin of error of any individual poll). As the Pew Research Center's Scott Keeter recent told the Arizona Reporter (in a clip received too late to make the column), "For the first time, we’re actually seeing a difference between cell-only voters and land line voters when you take into account age."

I hope you read it all.

 

Comments
DTM:

Thanks for the roundup.

As an aside, I am not entirely sure why people are making the point that the cell phone effect appears to be comparable in magnitude to the MOE on most polls. Since the cell phone effect appears to be introducing a systematic bias in one direction, it really has nothing to do with the MOE, given that sampling error is assumed to have a symmetric distribution. And that is just for individual polls--as Mark points out, the MOE on individual polls isn't particularly relevant to those looking at polling averages, whereas of course systematic effects are.

____________________

wagthedog1001:

Well, it seems after reading the article that the 1-3 point difference should just be added (mentally) to Obama's score. No it isn't related to the MOE. I think the point being made is that the number is similar. Actually the MOE could be added to either side, but the cell-phone bias if you can call it that would simply be added to one side. I'm looking forward to research done AFTER the election is over to see if any of this is relevant or not.

____________________

JCK:

Some pollsters are calling cell phones, so it's worth noting that not all polls should be adjusted for this effect.

____________________

As a 41-year old cell-phone only user, it's great to be included in a group generally considered to be part of a youthful demographic. And yes, I'm voting for Obama.

On another note, Sarah Says What???

____________________

AngryAlaskan:

In my opinion, the lack of polling for cell phone only households plus the historic boost in voter registration and (hopefully) the greater youth voter turnout will give Obama an additional +2 in polls throughout many states.

This is completely an unfounded prediction :)

____________________

Mark Blumenthal:

@DTM:

Thanks for your comment. My point was, admittedly, something of a throw-away aside in it's own right, but the MOE issue depends entirely on the context: Are you a Pollster.com, RCP or 538 reader focused on multi-poll averages or trend estimates or, alternatively, a congressional candidate or campaign pondering the one 400 interview survey you just paid for?

If you're rolling up dozens of surveys, the absence of cell phones probably matters. If you're just looking at just one survey, odds are any cell phone affect will be smaller than the purely random variation.

____________________

CTPonix4BHObama:

I've said it time and time again in no uncertain terms.

The Youth Vote Will Turn Out.

so prepare yourself for it.

____________________

Stonecreek:

In years past, the lack of a telephone in the home cause underpolling of minorities and poor people. This didn't have much impact in skewing poll results since it was pretty well offset by the proportionally lower voter participation of these groups.

However, the cell phone-only households which are being missed by most polling organizations are generally not lower-participating groups. In fact, it can be argued that (adjusted for age) cell-phone only households are more activist and cutting-edge than land-line households, therefore potentially higher turnout voters.

Given that most observers believe that groups favoring Obama (minorities and the young) will significantly exceed their traditional turnout levels, and that another Obama-favoring group, cell phone-only households, are being undersampled, these two factors should more than offset the (perhaps mythical) Bradley effect. If this hypothesis is accurate, then the state where these combined factors would most likely make a difference would be North Carolina, where you have a combination of a large number of minority voters plus a significant population of younger "cell phone" techies. Will be interesting to see.

____________________

Loyal:

Very interesting piece. Thank you for the summary and analysis Mark.

____________________

THX1138:

Mark - great article.

Can you expand on WHY only a few pollsters are including cell phones? You mention 'legal and logistical hurdles' - what's the big deal? Why is polling cell numbers fundamentally different than the traditional land-line technique?

____________________

RS:

@THX1138:
Mark gave some examples - such as making sure not to double-count households which have both landlines and cellphones. So many of the calls could be a wasted effort - if we assume *everybody* has a cellphone, then on average 85% of calls to cellphones would be wasted. That's an extreme, of course.

@DTM:
Good point about bias vs uncertainty. One thing to keep in mind, though, is that the Pollster.com average uses all polls - whether they account for cellphones or not. Many don't, so maybe adding a point or two to Obama's margin (or subtracting from a lag) might be OK, but not entirely correct.

____________________

s.b.:

I'd give you that 2% for Obama if voting was online and instantaneous. My personal opinion is that cell phone only users are much more likely to be so ADHD they can't go to the toilet without a cell phone and are much less likely by at least that 2% to show up to the polls and wait in line to vote.

They also may have to use paper and pencils at polls, which can be a challenge for many of them. Yes I'm being sarcastic, but you catch my drift. Cell phone only users and consistent voters come from two very different cultures.

____________________

RS:

@s.b.:
Keep pushing that tired old trope. And I am glad to know that you find Obama leading by just 8% rather than by 10% a victory. The Kool-Aid is mighty strong in you!

____________________



Post a comment




Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.

MAP - US, AL, AK, AZ, AR, CA, CO, CT, DE, FL, GA, HI, ID, IL, IN, IA, KS, KY, LA, ME, MD, MA, MI, MN, MS, MO, MT, NE, NV, NH, NJ, NM, NY, NC, ND, OH, OK, OR, PA, RI, SC, SD, TN, TX, UT, VT, VA, WA, WV, WI, WY, PR