"...we have designed the poll to meet four aims: to speak to current events, to inform applied policymaking, to support research, and to provide training opportunities for students."
Susan Moffitt, director of the Taubman Center
July 9, 2018
A new national poll conducted by the Taubman Center for American Politics and Policy at Brown University’s Watson Institute for International and Public Affairs is part of a revamped approach to polling that aligns with the Center’s core research and programming themes: the cost of living, the value of democracy, and the price of security.
Susan Moffitt, director of the Taubman Center, explains the new partnership with YouGov and how the survey will advance discussions of public opinion and academic research.
How does the new polling strategy relate to the Taubman Center's mission?
The Taubman Center’s research and programming focus on three core themes: the cost of living, the value of democracy, and the price of security. We have designed the Taubman Poll to examine various aspects of these core themes.
In this context, we have designed the poll to meet four aims: to speak to current events, to inform applied policymaking, to support research, and to provide training opportunities for students.
By fielding our surveys using the nationally regarded YouGov platform, we are able to use a wide range of formats to ask our questions. This includes asking questions connected to particular images and conducting survey experiments, in addition to asking standard survey questions. The high-quality YouGov sample is matched with voter registration records, which allows us to assess the opinions of likely voters. It also offers the potential for longitudinal analysis (asking the same people questions), which also offers us more opportunities to assess changes over time.
YouGov is well-known for its partnership with The Economist, and it is widely used in academic research. YouGov (Sample I) was recently rated and ranked by Pew as “the best performing sample” among “commercially available online nonprobability samples.”
What are the topics for the spring poll and how were Brown faculty involved in choosing them?
Each of our polls will routinely ask a standard battery of questions aligned with Taubman Center themes. Our spring poll includes:
-- Political questions (i.e. feelings toward the current administration; likelihood of voting for Democrats/Republicans in upcoming elections)
-- Economic questions (i.e. feelings about personal and national economic wellbeing)
-- Democracy questions (i.e. feelings about government and representation in the political system)
-- Security questions (i.e. feelings about safety and security in different contexts)
In addition to the standard battery of questions, we offer Taubman affiliates and their students opportunities to pose questions aligned with research and applied public policy.
How will the data be released?
We will release the data on the Taubman website in two ways. For one, we will offer an easy to use webtool that will allow Taubman website users to quickly create visual representations of the data. Users can use these for a range of purposes, including as images to be embedded in media articles. For the second, we will provide the “raw” data from the standard battery of questions to allow website users to conduct their own analyses.
Can you explain the methodology of an online poll?
The Taubman Center poll results were based on an online panel of 1000 respondents that nonpartisan polling organization YouGov selected using a technique called matched random sampling.
YouGov starts by drawing a random sample from the target population—U.S. adults—using the U.S. Census’s 2016 American Community Survey (ACS) one-year sample. Next, YouGov uses an algorithm to select respondents from its online panel of 1.8 million U.S. residents who are a close match to the individuals sampled from the ACS on gender, age, race and education. The final results are then weighted by respondents’ 2016 Presidential vote choice, and a four-way stratification of gender, age race and education.
This combination of matched sampling and post stratification yields a survey of U.S. results whose representation is comparable to and often better than other common survey techniques, such as telephone surveys via random digit dialing.