Capitol Fax.com - Your Illinois News Radar

* UPDATED x1 * Watch out for these bad polls

Posted in:

Illinois
Polls
Updated Posts

* From a July 6-10, 2016 Anzalone Liszt Grove Research poll of 606 likely Illinois voters when asked about their cell phone usage…

* Every responsible pollster has already taken note of this trend and adjusted accordingly. So, this move is no surprise…

As more Americans have made a mobile phone their main phone, polls that contact only landline phones look increasingly behind the times. HuffPost Pollster has decided we need to evolve.

By the end of last year, 48 percent of American adults didn’t have a landline phone. Another 16 percent had a landline but relied mostly on their mobile phone to make and receive calls.

That’s 64 percent of Americans who are only or primarily reachable via mobile phone. The numbers go even higher for those under age 45 and for minorities. Sixty percent of Hispanic adults don’t have a landline at all, and more than two-thirds of Americans between ages 25 and 34 rely solely on a mobile phone.

HuffPost Pollster has always sought to base our charts on as many sound polls as possible. As long as a pollster makes a good-faith professional effort to obtain a representative sample of the population, we include the survey’s results. In our opinion, polls that call only landlines with no attempt to sample the other half of the population no longer fit that description. So from now on, we will no longer include the results from landline-only telephone polls in our charts.

* So, consumer beware. If you see a published poll and it doesn’t include the percentage of mobile phone respondents, you can safely assume that there are no mobile phone respondents, particularly if it’s a robopoll. And media outlets that publish telephone poll results from surveys which don’t include mobile phone users are either clueless or living in the past.

Demand better.

*** UPDATE *** And right on cue…

A recent poll commissioned by The Illinois Observer shows Governor Bruce Rauner losing in a hypothetical re-election matchup by nine points.

Illinois Working Together Campaign Director Jake Lewis released the following statement regarding Governor Bruce Rauner’s sagging poll numbers:

“It’s no surprise that the vast majority of Illinoisans would not re-elect Governor Rauner. After all, it was Bruce Rauner’s reckless pursuit of his radical, anti-worker agenda that created a crippling and unnecessary budget impasse, devastating the state’s social safety net. Governor Rauner’s term in office has been a total failure since Day One, and the harm that he has inflicted on the people of Illinois may take years to heal. Illinoisans are sick of the governor’s divisive brand of politics that has achieved nothing except causing widespread damage to Illinois’ seniors, families, and economy.

“Though the 2018 election is still some time away, it is clear that Illinoisans have already rejected the extreme Rauner agenda. It is now up to the governor to listen to the people and drop his demands once and for all.”

The “poll” is problematic for numerous reasons. First and foremost, the pollster’s identity isn’t disclosed, which is highly unusual. Second, there is no disclosure of the type of polling that was done, live calls or robocalls. And third, no stats were disclosed on mobile phone users, if any.

posted by Rich Miller
Tuesday, Aug 2, 16 @ 10:22 am

Comments

This is a very important aspect of public opinion research and Rich is dead on. Any poll you see, is reported on or commissioned by you or your organization should be making a methodologically reliable and sincere effort to secure a representative sample of interviews conducted via cell phone. The percentage and number of cell phone interviews will vary based on where the interviews are conducted and other factors, but not disclosing that cell interviews were conducted or vaguely saying “cell phone interviews were included” without a definitive percentage and number of interviews is a red flag. Its also important to note that IVR (Robo) polls cannot legally call cell phones (significant FCC fines). They must have prior permission, or supplement using interviewers or other techniques to interview cell phone only (CPO) homes. We recognized this trend back in 2002 when our firm first began actively conducting interviews via cell phone (one of the first firms in the country to do so).

Dave Fako

Comment by Dave Fako Tuesday, Aug 2, 16 @ 10:35 am
Great information!

Comment by Precinct Captain Tuesday, Aug 2, 16 @ 11:11 am
“media outlets that publish telephone poll results from surveys which don’t include mobile phone users are either clueless or living in the past.”

Or they know that the accuracy of a poll has no bearing on its value as click-bait.

– MrJM

Comment by @MisterJayEm Tuesday, Aug 2, 16 @ 11:16 am
So it’s good news for Durbin that a biased poll, taken before Rauners resources have dug into him, while Rauner is in the middle of a heated standoff, only shows him at 43%?

Comment by Anonymous Tuesday, Aug 2, 16 @ 11:36 am
===a biased poll===

It’s not a biased poll as far as I know. It is, however, highly suspect.

Comment by Rich Miller Tuesday, Aug 2, 16 @ 11:57 am
Anon @ 11:36

Not sure if it’s a biased poll because, as Rich points out, we know nothing about it. But, be warned, land line-only polls (if this is one) usually lean Republican.

Comment by Teddy K Tuesday, Aug 2, 16 @ 11:57 am
We can demand better, but given that you need bodies to run cell phone polls and their increasing costs because they have to manually dial the numbers it drives up the cost into prohibitive ranges for many organizations.
The Pollsters podcast has been discussing a lot of these issues and the growing reliability of online polls.

Comment by Anonish Tuesday, Aug 2, 16 @ 12:35 pm
Rather than the constant demand polling how about media covering news?

Comment by Annonin' Tuesday, Aug 2, 16 @ 1:55 pm
Not to lead to another tangent but the podcast, The Axe Files with David Axelrod has had a few guests over the last few months dealing with this issue. Reporters and the Press no longer have people in the field talking to people and rely on the laziness of polling.

Comment by Come on man! Tuesday, Aug 2, 16 @ 1:59 pm
**drives up the cost into prohibitive ranges for many organizations.**

Don’t disagree. But polls that don’t include cell phones are pretty worthless at this point.

Comment by JoeMaddon Tuesday, Aug 2, 16 @ 2:27 pm
===It’s not a biased poll as far as I know. It is, however, highly suspect.===

I think a lot of folks misunderstand the concept of bias vs. validity. One can construct a completely honest and straight forward poll that lacks validity with no attempt for bias.

What Rich is saying is if your poll is only calling land lines the poll results lack validity.

Comment by Anon Tuesday, Aug 2, 16 @ 3:13 pm