Shining Light in Dark Places: Increasing Pollster Disclosure

Two weeks ago, I posted a rather glib post about fake public polls in Ukraine, arguing that from a strategic point of view or as evidence of stolen elections, they're meaningless. However, in the long run, the spate of fake pre-election polls make it that much harder to get information on what Ukrainians really think about issues facing the country. That's a problem.

Our friends at Pollster.com with the support of the the American Association of Public Opinion Research, has started the Disclosure Project, designed to give readers, journalists and opinion leaders more information about pre-election public polls in the U.S. For any poll that runs on the site, pollsters are asked to provide the following information:

  • The questions or procedures used to select or define likely voters or likely caucus goers.
  • What share of the voting-age population do "likely voters" represent?
  • The results to demographic questions and key attributes measures among the likely primary voter samples. In other words, what is the composition of each primary voter sample (or subgroup) in terms of gender, age, race, etc.?
  • What was the sample frame (random digit dial, registered voter list, listed telephone directory, etc)? Did the sample frame include or exclude cell phones?
  • What was the mode of interview (telephone using live interviewers, telephone using an automated, interactive voice response [IVR] methodology, in-person, Internet, mail-in)?
  • What was the verbatim text of the trial heat vote question or questions?

These are the types of questions that analysts everywhere should be asking every time they run an article about a pre-election poll. Many of these questions are already enshrined in AAPOR's Code of Ethics, but even the U.S., media outlets too frequently neglect to ask them.

There's no reason why journalists, opinion leaders and even readers in Ukraine or Poland or Turkey or anywhere else where polls are being used to shape public opinion can't ask these questions and insist that the information be disclosed. In the U.S. the sponsor of a poll is typically revealed so readers can make up their own mind about any biases that influence the result. In Ukraine, that's basic information that is too often left out.

While I don't believe disclosure is a cure-all when it comes to campaign finance, media or public polling, it's almost always the best place to start. And in the Former Soviet Union, where everything is opaque and independent sources of information difficult to find, it's the only place to start. More disclosure isn't going to stop a newspaper from running polls that advances the agenda of its owner or a big advertiser or a political party, but it will at least give readers more information that will help them make up their own minds.