The job of a pollster is to find out lots of different things. How popular politicians are, for example, or what Alaskans think of certain issues, and which products we like to buy. During the course of my long career in the polling business, I've even asked people which way they wind their toilet paper.
Surveys can be conducted by mail, in person or on the phone. In this day and age, they can also be done over the Intertubes, though this method isn't yet suited well for surveys of the general population. Telephone is the most common method still ... it's cost effective, it's quick, easily manageable and usually produces reliable results.
At least it used to.
Back in the old days, and we're really only talking about ten years ago, all but 1-2 percent of households in the U.S. had a land-line telephone. The same was true here in Alaska. Phone numbers could be randomly generated on specific geographic areas, and accurate random samples drawn where very few people were unreachable.
Then along came the cell phone. I still have very specific memories of a survey I did back in 1997 where I asked people in Anchorage if they ever used a cell phone (50 percent said yes) and also asked if the sight of other people using a cell phone ever annoyed them (50 percent said yes to that too, including half the cellphone users ... Go figure). Times have changed. Cell phones are now so commonplace, that they're changing our lives.
One of the changes they are causing is making life difficult for us pollsters, because Alaskans like their cells ... and they're giving up their land lines in droves. What's the point anymore when you've got your cell phone in your pocket?
Just last week I finished up a statewide survey, the first in a regular quarterly series that includes a subsample of what are called "cell-only respondents," i.e. people who have ditched their land lines. Including cell phones is not a decision we took lightly ... we ONLY call them on the weekend, to avoid the possibility of using up someone's minutes ... just about everyone has free weekend minutes these days. And we check before we start the survey to make sure that respondents can safely and conveniently talk, i.e. are not driving somewhere or standing in line at the supermarket.
We have learned two things about polling cell phones. First, cell phone respondents are in many ways more accepting and willing respondents than land line respondents ... which is, I suppose, not surprising since they take their phone with them everywhere they go! These folks like talking on the phone. My conclusion is that as long as we're upfront and conscientious about what we're asking them to do and make sure it's a good time for them, they are as likely to participate as anyone.
The other thing we found is this: Preliminary estimates from data in our cell-phone- inclusive surveys indicate that the percentage of adults (aged 18 or over) in Alaska who have a cell phone but don't have a land line in their home is just under 30 percent.
Holy moley... that's higher than every estimate I've seen for the U.S. as a whole. And the conclusion that it leads me to draw is inescapable. Cell phones simply must be included in Alaska surveys from now on. It is no longer good enough to fix the problem by weighting the data, or worse, by pretending the problem doesn't exist. You simply can't measure public opinion accurately when 30 percent of the public you're measuring the opinion of can't be reached by your methods.
So, to the 90 percent of you out there who use a cell phone ... you might be getting a call from me sometime soon with a survey. I'll do my best to make the surveys interesting, conduct them professionally and be concerned about your convenience, if you will do me a favor and understand that the decision to call you on your cell was made because it's a changing world out there.
Oh, and if you wind your toilet paper over the top, you're part of the 75 percent majority.
Ivan Moore is longtime Alaska pollster and researcher.
By IVAN MOORE