ANES: Data Products, Instrumentation, and Methodological Innovations

About This Grant

This project maintains the ANES gold-standard tradition of a scientifically valid probability-based public opinion survey while adding cutting-edge innovations that link AI methodology to advances in survey methodology. The project uses AI and automation processes to improve the sample validation processes and to improve accessibility and usability of ANES’ online resources. The ANES project continues a strong working relationship with the Comparative Study of Electoral Systems (CSES), the General Social Survey (GSS), and commercial companies. Furthermore, the project has developed a substantial user community, and more than 45,000 individuals have utilized ANES data since 2018. The ANES is a powerful educational tool, as 80 percent of the 45,000 individual users are undergraduate or graduate students at universities around the country. This current ANES project administers both pre- and post-election interviews, produces several new data products and methodological innovations, and includes the first-ever ten-year panel in the ANES time series. The project makes methodological advances in the use of AI in survey research by employing AI and automation processes to improve the matching ANES respondents’ files with commercial voter files and using AI to automate programs that evaluate the website and improve the accessibility and usability of ANES’ online resources. The 2026 study adds a 2026 wave to the ANES panel and delivers a Social Media Study that will be the longest and largest panel study of social media information linked to individual survey data in existence. The project continues ongoing collaborations with a domestic survey project (the GSS) and 66 international survey projects (the CSES) that yield new datasets of interest to scholars in sociology, economics, communications, comparative public policy, and international relations. Methodological innovations include a non-response follow-up study, innovations that improve sample representativeness, improvements in video interviewing, and the use of a mixed-mode design to yield further insights about survey mode effects. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.