Surveying likely voters not as easy as it used to be

Here’s something you are probably not worried about: the plunging response rate for public opinion surveys.

Yes, worry about global warming, racial tensions, public education, homelessness, the Chinese economy and the University of Hawaii football team — but seriously, do we also have to stew about whether the public refuses to answer pollsters?

Maybe no, but because this is a column about politics, maybe yes.

Looming in the rear view mirror and closing at a fast rate is the 2016 election. Coming along with it are scads of political public opinion polls. This newspaper will take them; so will other media outlets and most of the major politicians and political entities such as unions and business group.

For the day they were taken, most professionally prepared public surveys are likely to draw an accurate picture, although that doesn’t mean they can predict the future, says Jim Dannemiller, president of Honolulu-based SMS Research.

“When a poll comes out saying that 41 percent of the vote belongs to a certain candidate, or something, I could say it is probably true right now, although it may not be true tomorrow,” says Dannemiller.

Polling as a science is doing fine, but polling has new challenges. The caution is that folks today are not as willing to be questioned.

“Regarding the response rate, we have never really stopped going downhill,” reports Dannemiller.

In an interview, Dannemiller explains that at the beginning of a political season, voters are more willing to answer questions because the political campaign is fresh and new — but as the season drags on, voters “just don’t want to be bothered.”

Becki Ward, president of Ward Research, adds that the increased use of cellphones means many polls are designed to take up to 70 percent of the sample from cellphone users without landlines.

She notes that people with landlines are usually older, so getting the right balance of respondents is tricky.

“This is all new territory and there are no fixed rules for designing the news polls,” Ward says.

For voters, however, the real concern is: Can you believe the polls, and what are the telling signs of a good poll and a misleading one?

Dannemiller advises that voters should watch out for polls that don’t tell you how they were taken, including the questions, the exact results and who paid for it.

Obviously, a poll conducted by a candidate’s supporters is not worth much consideration.

“Campaign workers can’t hide their enthusiasm for the candidate, when they ask poll questions,” says Dannemiller.

Other pollsters also note that not just candidates are playing in the polling business. Issue campaigns are also sending out press releases announcing that a candidate favorable to a cause is ahead.

The carefully placed distortions mean that voters have to be more questioning and less accepting of poll numbers meaning that someone is really ahead in a race.

So as the “I’m winning” proclamations start to flow and politicians hail their own predictions of future success, it is wise for voters to consider the source.

Richard Borreca writes on politics on Sundays, Tuesdays and Fridays. Reach him at rborreca@staradvertiser.com.