Opinion polls deserve much of the derision we pile upon them.

But I’m not sure we should blame the pollsters so much as the rest of us: the people who consume the polling news, and the politicians, media and special interests who pay to conduct the polls.

Polling is a fascinating industry, dually dependent on social sciences and math. To be credible, opinion polls must use a random sample that will produce a true reflection of the group being surveyed, and the questions must be posed in an unbiased manner.

Often, polls cited by politicians fall short of those marks. They cite polls that were designed to elicit certain responses. Mailers you receive from politicians often include these kinds of questionnaires or polls. They claim they want your input, but the questions are worded to get you to agree with their agenda.

And some polls conducted during political campaigns aren’t polls at all. Known as push polls, their purpose is to smear an opponent. These have grown more common in recent years, with “pollsters” asking potential voters such questions as: Would you be less likely to support (opponent’s name) if you heard he had been arrested for beating his wife?

Note, the pollster didn’t claim the assault or arrest actually happened, but merely posed the question to create suspicion.

Voters who value character and integrity ought to reject candidates who resort to such tricks. But in this era of lax campaign rules, it’s often not even possible to identify the responsible parties. So we shrug, rationalize that “they all do it,” and let ourselves be influenced by dishonest insinuation.

Media outlets also have been known to cite polls that aren’t scientific. These include many newspapers’ online polls, radio show call-in polls, Twitter polls and so on. If the poll relies on respondents to take the initiative to participate, it’s highly improbable that the sample accurately reflects the population that it’s supposed to.

Some of these polls are harmless, asking such things as your favorite book or movie. They should be viewed more as conversation starters than as accurate measures of public opinion.

Even among scientifically valid polls, there can seem to be big differences that undermine the public’s trust – not only of the results but of the motives of the pollsters.

One example are polls that measure the approval ratings of President Donald Trump.

A Washington Post-ABC News poll recently found that among all Americans, Trump’s disapproval rating had reached 60 percent.

However, Rasmussen Reports, which tracks the president’s approval rating daily, found during the same time frame that 50 percent of likely voters disapproved of Trump.

Both Rasmussen and Washington Post/ABC are considered reputable polling outfits.

The difference is explained by who was surveyed. The Post/ABC poll reflected the population of all adult Americans. The Rasmussen poll is a reflection of likely voters – a subset of our population that would account for fewer than half of all adult Americans.

As voters and consumers of news, we need to discern and appreciate such differences.

In an era when truth so often loses out to propaganda, we need to appreciate the significant difference between legitimate measures of public opinion and dishonest polls that aim to further a cause or candidate.

As campaigns work to persuade us to vote a particular way, all of us will be subjected to more polls, more quasi-polls and political shenanigans disguised as polls.

We should be ready to do some homework about what politicians and media say the polls are, what they measure and what they mean.

Online, RealClear Politics, a conservative news aggregation site, does a good job of compiling poll data from credible sources. Another option is Fivethirtyeight, which leans just a little to the left.

Together the two show how political bias doesn’t automatically translate into dishonest news and information.

Scientific polls can provide solid information about what the public is thinking. The difficulty is in distinguishing between credible polls and the propaganda meant to manipulate opinion rather than measure it.


A native of Garden City, Julie Doll is a former journalist who has worked at newspapers in California, Indiana and New York, as well as across Kansas.