Trust Magazine

When 51 Percent Might Not Mean a 'Majority'

How to write precisely about survey data.

In this Issue:

  • Fall 2018
  • The Hollowing Out of Newsrooms
  • Actually, Millennials Are Planning for Retirement
  • Preparing for a Changing Future
  • The Big Picture
  • Noteworthy
  • When 51 Percent Might Not Mean a 'Majority'
  • Australia Commits to Expand Protections in the Outback
  • 'Like a Kid in a Candy Store': Remembering Gerry Lenfest
  • Close Encounters With Killer Whales Offer Clues to Southern Ocean Health
  • Mobile Food Banks Serve Isolated, Rural Poor
  • Juvenile Justice Reform Can Help Young People 'Turn Their Lives Around'
  • The Pew Research Center Remains Focused on the Facts
  • Return on Investment
  • Improving Public Policy
  • Informing the Public
  • Invigorating Civic Life
  • Fishing Subsidies Are Speeding the Decline of Ocean Health
  • End Note: How Americans Value Gender
  • When You Say You Believe In God, What Do You Mean?
  • View All Other Issues
When 51 Percent Might Not Mean a 'Majority'
Trust Magazine
Getty Images

Mark Twain once said that “the difference between the almost right word and the right word is really a large matter—’tis the difference between the lightning-bug and the lightning.” That advice is especially apt when writing about survey data.

For many people, “majority” is a word so common that they rarely have to think twice about what it means. But it’s a different matter for polling organizations like the Pew Research Center. By their nature, polls provide an estimate of what a large group of people say, since they’re based on a sample rather than the entire population. This basic reality can create challenges for writers who want to summarize poll findings in a precise way.

To better appreciate the problem, consider these three recent news headlines:

  • Majority of Canadians would boycott U.S. in response to trade war
  • A majority of travelers are picking destinations based on food
  • Majority of Americans approve of Trump’s handling of the economy for the first time

Each headline includes the word “majority.” But take a moment to consider the wide variation in the survey data that’s actually being cited in each of the stories above. In the first article, a “majority of Canadians” refers to 72 percent. In the second story, “a majority of travelers” refers to 64 percent. And in the third article, a “majority of Americans” refers to 51 percent. These differences are substantial.

In order to present survey findings in an accurate and impartial manner in its reports, the Pew Research Center has developed internal guidelines for using certain terms consistently. At the center, writers cannot label a survey finding a “majority” unless it meets specific criteria.

One of these criteria is the survey’s margin of error. Since surveys only question a sample of a larger population that is being studied—whether that population is a single city, an entire country or something else—the margin of error describes the estimated range within which we would expect the exact answer to fall. (The results we would have gotten if we had surveyed everyone in that larger population is the “true population value.”) For example, if a survey has a margin of error of plus or minus 3 percentage points at the 95 percent confidence level, that means we can expect the result to be within 3 percentage points of the true population value 95 out of 100 times.

Let’s say our hypothetical survey asks a yes or no question about whether the economy should be the top concern of the country’s political leaders. If 51 percent of all respondents say “yes,” we would estimate the true population value to fall between 48 percent and 54 percent. As such, 51 percent in this survey would not necessarily translate to a “majority” of Americans. But if 54 percent say “yes”—again keeping in mind the 3-point margin of error—then we would estimate the true population value to fall between 51 percent and 57 percent. Therefore, it would be a fair characterization by our standards to say that a response of 54 percent or higher is a “majority” share of the population. (The center’s writing guidelines note, however, that caution is always warranted when you’re close to the threshold.)

It is important to remember that there are several sources of uncertainty in survey estimates, some of which can’t be quantified. For instance, question wording can introduce error or bias into the findings of opinion polls. Another potential source of error is nonresponse bias—that is, when the people who respond to surveys differ in important ways from the people who don’t respond.

Although the center’s writers can’t quantify all of the uncertainties when summarizing survey data, they seek to acknowledge them. Or, put another way, precise writing requires an accurate description of imprecision. 

Mark Strauss is a writer and editor focusing on science and society at the Pew Research Center.

Australia Commits to Expand Protections in the Outback Noteworthy
America’s Overdose Crisis
America’s Overdose Crisis

America’s Overdose Crisis

Sign up for our five-email course explaining the overdose crisis in America, the state of treatment access, and ways to improve care

Sign up
Quick View

America’s Overdose Crisis

Sign up for our five-email course explaining the overdose crisis in America, the state of treatment access, and ways to improve care

Sign up