The American Association for Public Opinion Research held its annual conference a few weeks ago in Hollywood, FL. Unfortunately, I could not attend, but several of the presentations from the conference and interviews are posted on the web.
One of the presentations focused on the cell-phone-only voters from data collected in the 2008 exit poll. Michael Mokrzycki, the polling director for the Associated Press, which is member of the National Election Pool (NEP), along with Scott Keeter of the Pew Research Center, Courtney Kennedy of the University of Michigan, and The Everett Group, presented a report focused on the following three topics:
1. Cell-phone-only (CPO) incidence and rate of growth
2. How do CPO voters differ from landline-reachable?
3.Bias implications for future surveys that don’t attempt to cover the wireless-only population
The NEP question in the 2004 and 2008 exit polls was:
What type of telephone service is there in your home that you could use or be reached on?
Both regular land-line and cell phone service
Only regular land-line phone service
Only cell phone service
No telephone service at home
In 2004, just 385 of the exit poll interviews out of 5,619 were with CPO-voters, or around 7%. But in 2008, 20% of voters in the exit poll were CPO-voters (1,496 out of 7,341). Note: these interviews are based only on election day voters.
Cell-phone-only voters backed Obama by large margins (38% McCain/61% Obama). The Presidential ballot among CPO-voters in 2004 was 44% Bush/54% Kerry. CPO-voters in 2008 were twice as likely as landline voters to be voting in their first election. This is largely due to the fact that CPO-voters are younger than landline voters and younger voters are more likely to be first time voters.
Much has been written about the 2008 voter contact programs of the Presidential campaigns, but was there any difference in their reach by phone status? The results from the exit poll data show cell-only-voters were much less likely to have been contacted by either campaign. 74% of CPO said they were NOT contacted by either campaign, versus 56% of land-line voters.
The report noted, “Even Obama’s campaign – renowned for its technological innovation – was more likely to contact landline reachable than CPO-voters – 34% of landline-reachable reported contact from Obama campaign, versus 23% of CPO.”
Reaching cell-phone-only voters is not only an issue for pollsters, but also an issue for campaigns when they are deciding ways to communicate their messages.