Reproducibility as part of code quality control
In this post, we discuss reproducibility as a part of Pew Research Center’s code review process.
A behind-the-scenes blog about research methods at Pew Research Center.
For our latest findings, visit pewresearch.org.
In this post, we discuss reproducibility as a part of Pew Research Center’s code review process.
After venturing into the world of computational social science in 2015, the Center needed to develop new tools and workflows.
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.
To search or browse all of Pew Research Center findings and data by topic, visit pewresearch.org
This is a blog about research methods and behind-the-scenes technical matters at Pew Research Center. To get our latest findings, visit pewresearch.org.
Copyright 2024 Pew Research Center