RECAP: Cornell Collaborative Midterm Survey Hackathon and Data Launch
Kristen Conrad and Mickey Jackson Presented on the Data Collection Panel: Collaborating to Increase Survey Accuracy and Representation
- On January 20, 2023, Cornell Tech in NYC hosted experts from industry, academia, and media to offer their perspectives on the innovations, methods, and data from the 2022 Collaborative Midterm Survey, which had a combined sample size of nearly 20,000 adults living in the US, including oversamples in California, Florida, and Wisconsin.
- Key findings include that nearly half of white Americans recognize that the deck remains stacked against Black Americans. One out of four Americans think parents should decide whether their kids buckle up. And a majority of Americans believe in restoring voting rights to people previously convicted of a felony.
- The NSF-funded Collaborative Midterm Survey is a multimode survey (including probability and nonprobability samples) conducted by three data collection teams (SSRS, Gradient and Survey 160, University of Iowa) selected from an open call for data collection proposals.
- For its sample, SSRS used a combination of the probability-based SSRS Opinion Panel, a supplemental address-based sample, and multiple nonprobability panels. The SSRS sample was calibrated using our Encipher Hybrid methodology for blending probability and nonprobability data.
- The SSRS sample presented a highly accurate picture of the 2022 midterm electorate. The estimated House generic ballot from the pre-election SSRS sample was within 1 percentage point of the actual results.
- Mickey Jackson and Kristen Conrad of SSRS presented on the Data Collection Panel: Collaborating to increase survey accuracy and representation. Other panelists include Cory Manento (Gradient), Kevin Collins (Survey 160), and Julianna Pacheco (the University of Iowa).