You’re not going to believe this poll.

Prepare to get meta: A new poll shows three-quarters of Americans, across all demographic subgroups, think public opinion polls are biased.

A poll. Of people. Telling us most people don’t believe polls.

Distrust is strongest for polls conducted by candidates, political parties and automated voice recording firms, but news media polls are not widely trusted, either, according to the survey of 1,011 Americans conducted July 24 through Aug. 4 on behalf of Kantar, the research and data management division of WPP, a British multinational advertising and public relations company.

The poll, called “The Path to Public Opinion,” found that although Americans believe polls are biased, they’re not certain who they favor: A very small percentage believes they are biased toward conservatives; a slightly larger percentage believes they are biased towards liberals; and a significant majority (68 percent) just think they are biased in some way.

Also, 67 percent of Americans claim to pay little to no attention to polls when considering for what or whom to vote. Yet 59 percent of Americans say they pay attention to consumer research when considering products or services to buy.

While poll participants are harder to find, Kantar’s research shows that the identity of a poll’s sponsor is a key determinant of people’s willingness to take part. Academics and foundations have the most positive impact on willingness (41 percent say they are more likely to participate) while social media sites get only 11 percent; news organizations, at 24 percent, run about even with political parties or candidates, at 23 percent.

Surprisingly, only 11 percent of Americans say they view social media as a viable source of information about public opinion on policy and politics, and 60 percent are less likely to take a poll conducted on a social media site. Only 6 percent say they use social media to communicate about issues and causes, while 61 percent say they use it only to communicate with friends and family.

Josh Richman

Josh Richman covers state and national politics for the Bay Area News Group. A New York City native, he earned a bachelor’s degree in journalism from the University of Missouri and reported for the Express-Times of Easton, Pa. for five years before coming to the Oakland Tribune and ANG Newspapers in 1997. He is a frequent guest on KQED Channel 9’s “This Week in Northern California;” a proud father; an Eagle Scout; a somewhat skilled player of low-stakes poker; a rather good cook; a firm believer in the use of semicolons; and an unabashed political junkie who will never, EVER seek elected office.

  • Joseph De Sante

    All poles are biased based on who puts them together and how they want the pole readers to react to the poled questions.

  • Alcoahead

    I don’t believe this article — so there. 😉

    @1 — It’s “polls”, not “poles” — Polish people worldwide are very upset and asked me to chime in on that statement you made wherein you called “all poles…biased”.

  • Bruce R. Peterson, Lafayette

    When a caller mentions a poll or survey, I always ask who their sponsor is. Usually the caller tells me their sponsor is secret. So I say, “If you won’t answer my question, I won’t answer yours”.
    It’s 100% certain, it’s a survey for a tax election, if they do name their sponsor. They want to know how much money fools will depart with, for a named cause.

  • M McL

    Ive been suspicious of Nielson ratings for TV for a long time. I don’t know anyone called by these people yet they state they know ratings.

  • jskdn

    There are many way to introduce bias into polling. I always want to see the methodology and the polling script in order to assess confidence in the results. It bothers me when news organization report on polls without providing links to that information, something any reporter would need to do their job adequately.

  • Rod Campbell

    Stay out of the Middle East fights……They don’t care about us and we cannot fix their problems…