How keyword oversampling can help with text analysis
Keyword oversampling can be a powerful way to analyze uncommon subsets of text data.
Numbers, Facts and Trends Shaping Your World
Read our research on:
Decoded
All
Decoded
Publications
-
-
How many Taiwanese live in the U.S.? It’s not an easy question to answer
Surveys can produce widely different estimates depending on how people are asked about their backgrounds.
-
Using cognitive interviewing to design survey questions about democracy
Asking follow-up questions can help make sure that poll respondents are interpreting questions as intended.
-
Developing survey questions on sensitive topics in India
Asking balanced questions required investing considerable time and effort into developing and testing the questionnaire.
-
‘What’s the difference between income and wealth?’ and other common questions about economic concepts
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.
-
Q&A: How Pew Research Center surveyed nearly 30,000 people in India
To represent the views of Indians from a wide range of backgrounds, we fielded our largest-ever single-country survey outside the U.S.
-
Measuring caste in India
Despite the caste system’s significance in India, there is no consensus on what proportion of Indians belong to each caste category.
-
Estimating U.S. newspaper circulation is a challenge — especially for 2020
Using data from outside organizations has some advantages, but also poses challenges, especially when it comes to media industry data.
-
Improving the way we categorize family income
We’ve made adjustments to account for differences in purchasing power by survey respondents’ household size and geography.
-
Examining how survey mode affects Americans’ views of international affairs
Responses to survey questions asked online can differ from those asked on the phone.
