Understanding error bars in charts
Error bars illustrate the margin of error for a survey estimate by showing how precise that estimate is. Here are some answers to common questions that might help you better understand charts with error bars.
A behind-the-scenes blog about research methods at Pew Research Center.
For our latest findings, visit pewresearch.org.
Error bars illustrate the margin of error for a survey estimate by showing how precise that estimate is. Here are some answers to common questions that might help you better understand charts with error bars.
The typology study examines U.S. politics through the prism of people’s values and attitudes, not just their party labels.
Showing margins of errors in graphics can help prevent readers from thinking that survey estimates are more precise than they really are.
Nonresponse rates for open-ended survey questions on Pew Research Center’s American Trends Panel can range from 3% to just over 50%.
Surveys that ask about voting can be made more accurate by validating respondents’ self-reported turnout with official voting records.
The final post in our series examines how topic models can and can’t help when classifying large amounts of text.
Keyword oversampling can be a powerful way to analyze uncommon subsets of text data.
Surveys can produce widely different estimates depending on how people are asked about their backgrounds.
Asking follow-up questions can help make sure that poll respondents are interpreting questions as intended.
Asking balanced questions required investing considerable time and effort into developing and testing the questionnaire.
Economic concepts aren’t always as understandable to the rest of us as they are to economists, and the jargon can be difficult to parse.
To search or browse all of Pew Research Center findings and data by topic, visit pewresearch.org