New PolitIQs: Voter Behavior Research Newsletter
Reaching RAE Voters with Cell Phones: Developing the Best Phone Voter Outreach Program
Pundits across the political spectrum agree that Democrats’ success in the midterms hinges on our ability to reach and mobilize RAE voters at higher rates than in 2010. We read each day about exciting new ways to find these targets online, but the body of research on ROI for digital advertising is still too small to draw sweeping conclusions. We know from more than a decade of research that quality field efforts, perhaps not quite as sexy as the digital communications, are essential to victory.
As campaigns begin to run the numbers on the cost-benefit ratios on door-to-door canvassing this fall, we offer our own findings about outreach to voters who can only be reached by cell phones.
We took the results from a range of calling programs in 2013 and early 2014 where there was a field/phone component. We measured the differences in list quality, list penetration and overall costs between land line and cell phone voters. The study included statewide ballot measures, local ballot measures, a congressional special election and local candidate elections in 6 different states. Note that about 27% of these phone records in the study were mobile numbers. We know that the cell phone percentages will only continue to increase.
Here are some highlights from our recent study.
List Quality: Mobile phone lists are slightly cleaner than land line lists.
We compared results of several different phone record dispositions.
- Bad phone numbers. Mobile phones records were a bit fresher.
- Deceased voters, rarely more than 2% of a file, were not an issue for land lines or cell phones.
- Non-english speaking voters were more likely with mobile phones, a result consistent with Pew Studies on cell phone penetration among Hispanic voters and others where English is a second language.
- Wrong numbers. Mobile phone respondents were slightly more likely to tell us we had the wrong number. Still the differences were minimal.
List Penetration: The rates for actual conversations with voters were about the same with land lines and cell phones.
- Early Hang-ups were almost twice as frequent with land lines but still a very small portion of the total. We can only speculate at this point that mobile phone users are less burdened by telemarketing calls and more likely to answer a ringing phone without first checking caller ID.
- Do not call. There was not much difference between land line voters and cell voters who said “don’t call me back” or “I don’t have time to talk to you now.” On the latter point, we expected more cell phone voters to terminate the call because they were driving, walking the dog, at a restaurant, etc. It didn’t happen.
- Voicemail. There was not much difference between land lines and cell voters who let the call go to voicemail. We expected higher rates with cell phones— as voicemail is a standard feature on all cell phones but not a given with land lines. We suspect that a sizable portion of cell phone users simply haven’t set up their voice mail features or don’t empty them regularly.
So with list quality and connection rates comparable between land lines and cell phones, it follows that any quality midterm calling program to Democratic voters must include a cell phone component. Campaigns and consultants would be well served to insist upon the dialing of cell phones unless the targets are exclusively older, white rural voters—and even then the program would be a significant number of potential voters.
We will report in the next edition about the costs differentials between land lines and cell phones. In the meantime, if you’d like the details of this study, just click here.
And as always, we invite you to join our voter research conversation on Twitter @ChismStrat.
SIGN UP here to receive PolitIQs in your inbox
Chism Strategies is America’s leading direct voter contact firm. We are innovators, grounded in the best social science research, with decades of experience in the field to mobilize voters and change public opinion.