logo
Please note!
This fact brief was originally published as an experiment to test the concepts behind fact briefs.
Readers should be aware that while there is still a lot of useful information in fact briefs like this one, not all of them reflect Gigafact's current methods and standards for fact briefs. If you come across any that you feel are out of date, don't hesitate to contact us at support@gigafact.org.
Brief Publisher Logo
FACT BRIEF

Do most election polls have a higher margin of error than commonly reported?

By Bakar Wilson
Copy link
email
Copy featured image
email
YES

A 2016 study from the American Statistical Association examining 4,221 late-campaign polls found "the historical margin of error is plus or minus 6 to 7 percentage points" rather than the typically reported plus or minus 3 points that is usually cited as sampling error.

This higher margin of "total survey error" is affected by coverage error, i.e., when not all of a target population has an equal chance of being sampled. For example, phone surveys exclude people without phones. A Pew Research analysis found "politically and civically engaged individuals" were overrepresented in online surveys. Most of the online surveys Pew studied also had a disproportionate number of childless, low-income adults.

The higher margin of error also allows for nonresponse error—which can arise, for example, when supporters of a trailing candidate become less likely to respond to surveys.

This fact brief is responsive to conversations such as this one.
Sources
David M. Rothschild: Disentangling bias and variance in election polls (as published in Journal of the American Statistical Association)
New York Times: When you hear the margin of error is plus or minus 3 percent, think 7 instead
Pew Research Center: Evaluating online non-probability surveys
Pew Research Center: Key things to know about election polling in the US
ABOUT THE CONTRIBUTOR
Gigafact
Between 2020 and 2022, under close editorial supervision, Gigafact contracted a group of freelance writers and editors to test the concepts for fact briefs and provide inputs to our software development process. We call this effort Gigafact Foundry. Over the course of these two years, Gigafact Foundry writers published over 1500 fact briefs in response to claims they found online. Their important work forms the basis of Gigafact formats and editorial guidelines, and is available to the public on Gigafact.org. Readers should be aware that while there is still a lot of relevant information to be found, not all fact briefs produced by Gigafact Foundry reflect Gigafact's current methods and standards for fact briefs. If you come across any that you feel are out of date and need to be looked at with fresh eyes, don't hesitate to contact us at support@gigafact.org.
FACT BRIEF BY
Brief Publisher Logo
Latest fact briefs
facebook
twitter
email
email
About fact briefs published on Gigafact
Fact briefs are short, credibly sourced summaries that offer “yes/no” answers in response to claims found online. They rely on publicly available, often primary source data and documents. Fact briefs are created by contributors to Gigafact —a nonprofit project looking to expand participation in fact-checking and protect the democratic process. See all fact briefs here.