Cell Completes Needed

The Question(s)

How many people do you know who still use a landline phone?

How many do you know who would need a plastic surgeon to remove the cell phone from the palm of their hand?

The Answer

Despite the number of people in the second category (let’s call them cell-phoners) continually increasing relative to the number in the first category (let’s call them landliners), there are still politicos relying on polls and IDs completed entirely over landlines. Just this year alone, we’ve seen a 40%+ drop in the number of completes available through landlines compared to the source file.

 Aside: It need not be said, but I want to be sure we’re all on the same page. All 100% automated or IVR polls are completed by calling nothing but landlines. It is illegal to autodial cell phones.

The Conundrum

There are still reasons to complete polls calling only landliners. They are:

1. It’s cheaper

That’s it. End of list.

Cost is the only reason to conduct a poll exclusively using landlines if you care at all about accuracy.

In the past, you could argue that cell-phoners all fit a certain mold – younger, don’t vote, etc. That’s not the case anymore. The only potential commonality we’ve been able to discern in recent polling is that cell-phoners do tend to be lower-propensity voters. There could be many reasons for this, including moving more frequently while keeping their cell phone number, meaning they haven’t voted in the recent elections in the area you’re polling.

Put another way, they’re more…

wait for it…

mobile! (sorry)

The Evidence

But that’s really where the generalizations end. For example, in three recent polls in a single state – Tennessee – we found cell phone respondents to be disproportionally low-income in one district, disproportionally high-income in another, and almost identical to the overall sample in a third. And I don’t mean by a few percentage points either. There was as much as a 24-point spread between the percentage of high or low-income respondents by phone type within the districts.

Another area we’ve seen major differences is in the top issue question. Specifically, in a Florida state legislative district poll, “stopping illegal immigration” was the top issue for over 30% of landliners. Cell-phoners? 17%. “Better schools” was the top issue for 25.6% of cell-phoners. Landliners? 12%. For another example, a recent state legislative district poll in landliners Georgia found “fund education” to be the top issue for 34.6% of cell-phoners. Landliners? 15%.

We’ve even seen the differences in ballot tests. In a Wisconsin state assembly district poll, one candidate overperformed with landliners by almost 19 points, partially because cell-phoners were three times as likely to be undecided in that district. Doesn’t that seem like extremely valuable information?

Using the above examples, when anywhere from 30% to 60% of your electorate primarily use cell phones, you would have an extremely inaccurate view of your electorate by calling only landlines. As a result, if you’re running a small campaign, you’re going to waste precious resources focusing heavily on stopping illegal immigration. If you’re running a big campaign, you’re going to miss the opportunity to craft different messages and deliver them to cell-phoners.

And what about Voter ID? How many potential voters are you missing by not using texting/SMS as part of your effort? When it comes time to turn your voters out, a significant portion won’t hear from you because you didn’t include (extremely affordable) cell-phoner outreach.

The Conclusion

Yes, the above examples are anecdotal. (That’s kind of the purpose of examples, but I digress.) What they show, however, is not just the difference between landliners and cell-phoners, but the inconsistency, and therefore unpredictability, of those differences.

Even with the one aspect that is somewhat predictable – cell-phoners being lower-propensity voters – that’s the exact segment you need to be talking to in a year like this. Every sign points to high turnout on election day. If you ignore what 30%-60% of those voters are likely to do, good luck. And, if you get the results on election day that your landliners-only poll predicted, go buy a lottery ticket, because you are one lucky son of a gun!

Don’t get me wrong – no poll is perfect, and things can change between the time you run a poll and election day. What doesn’t change, however, is the core data in your crosstabs that helps you identify your voters and the messages that will resonate with them. How accurate do you think that data will be if you ignore cell-phoners? It’s like trying to color a rainbow with 1 crayon. (Wow – that’s cheesy. But I bet you remember it!).

The Summary

  • There are lots of cell-phoners now
  • They don’t all look, act, or think the same way
  • Ignore them at your own peril
  • Rely on certainty – let your opponent rely on luck