Determining which candidate is "ahead" is the most visible and potentially dangerous element to report in any election polls. Public opinion can change quickly and dramatically.

Here are some areas where journalists should focus their attention when reporting on polls that compare voting intention:

  • Margin of sampling error: You hear or read statements such as, “Candidate A leads Candidate B by four percentage points, 50 percent to 46 percent, with a sampling error of plus or minus 3 percentage points.” The term conveys that the pollster has taken a small sample and is using that to represent the underlying population of all voters with it. Election polls are an estimate of candidate preference. Noting that different samples from the population would produce slightly different estimates as a result of chance variation is what is referred to as sampling error and is expressed as plus or minus a certain number of percentage points.
  • Undecideds: In addition to reporting the level of support for each candidate, what's the percentage of undecided voters? And note that, as the election gets closer, the number of undecided voters generally decreases.
  • Question Wording: Be careful when reporting on electability (“Of all of the candidates running for president, which one do you think is most likely to win the election?”) versus voter preference (“In the upcoming election, which candidate will you vote for?”). These questions do not measure the same thing and you will see a range of results. The difference between whom a voter supports and whom the voter thinks will win can be significant.

Taken from Understanding and Interpreting Polls, a self-directed course at Poynter NewsU, developed in partnership with the American Association for Public Opinion Research (AAPOR).

Take the full course

Have you missed a Coffee Break Course? Here's our complete lineup. Or follow along at #coffeebreakcourse.