NORC at the University of Chicago conducted the study in collaboration with the American Press Institute and funding and partnership from Democracy Fund. The survey was co-designed and written by API and NORC with support and engagement throughout the process by Tom Glaisyer and Jessica Mahone of Democracy Fund. Data were collected using AmeriSpeak Omnibus®, a monthly multi-client survey using NORC’s probability-based panel designed to be representative of the U.S. household population.
During the initial recruitment phase of the panel, randomly selected U.S. households were sampled with a known, non-zero probability of selection from the NORC National Sample Frame and then contacted by U.S. mail, email, telephone, and field interviewers (face-to-face). The panel provides sample coverage of approximately 97 percent of the U.S. household population. Those excluded from the sample include people with P.O. Box only addresses, some addresses not listed in the USPS Delivery Sequence File, and some newly constructed dwellings.
Interviews for this survey were conducted between December 13 and 16, 2018, with adults age 18 and over representing the 50 states and the District of Columbia. Panel members were randomly drawn from AmeriSpeak, and 1,067 completed the survey—992 via the web and 75 via telephone. Interviews were conducted in both English and Spanish, depending on respondent preference. The final stage completion rate is 20.1 percent, the weighted household panel response rate is 34.2 percent, and the weighted household panel retention rate is 85.1 percent, for a cumulative response rate of 5.8 percent. The overall margin of sampling error is +/- 4.1 percentage points at the 95 percent confidence level, including the design effect. The margin of sampling error may be higher for subgroups.
Once the sample has been selected and fielded, and all the study data have been collected and made final, a poststratification process is used to adjust for any survey nonresponse as well as any non-coverage or under- and oversampling resulting from the study-specific sample design. Poststratification variables included age, gender, Census division, race/ethnicity, and education. Weighting variables were obtained from the 2018 Current Population Survey. The weighted data reflect the U.S. population of adults age 18 and over.
All differences reported between subgroups of the U.S. population are at the 95 percent level of statistical significance, meaning that there is only a 5 percent (or lower) probability that the observed differences could be attributed to chance variation in sampling.
From the American Press Institute
- Tom Rosenstiel
- Kevin Loker
- Jeff Sonderman
From NORC at the University of Chicago
- David Sterrett
- William Bonnell
- Jennifer Benz
- Trevor Tompson
From Democracy Fund
- Jessica Mahone (now Duke University)
For printing and offline viewing, a PDF version of the topline survey results are available for download.