Courtney Kennedy is director of survey research at Pew Research Center. Her team is responsible for the design of the Center’s U.S. surveys and maintenance of the American Trends Panel. Kennedy conducts experimental research to improve the accuracy of public opinion polls. Her research focuses on nonresponse, weighting, modes of administration and sampling frames. Her work has been published in Public Opinion Quarterly, the Journal of Survey Statistics and Methodology and the Journal of Official Statistics. She has served as a co-author on five American Association for Public Opinion Research (AAPOR) task force reports, including chairing the committee that evaluated polling in the 2016 presidential election. Prior to joining Pew Research Center, Kennedy served as vice president of the advanced methods group at Abt SRBI, where she was responsible for designing complex surveys and assessing data quality. She has served as a statistical consultant for the U.S. Census Bureau’s decennial census and panels convened by the National Academies of Science, Engineering, and Medicine. Kennedy has a doctorate from the University of Michigan and a master’s degree from the University of Maryland, both in survey methodology. She received bachelor’s degrees from the University of Michigan in statistics and political science. Kennedy has served as AAPOR standards chair and conference chair.

A typical public opinion survey cannot provide reliable insights into the attitudes and experiences of relatively small and diverse religious groups, such as adults identifying as Jewish or Muslim. Not only are the sample sizes too small, but adults who speak languages such as Russian, Arabic, or Farsi (and not English) are excluded from interviewing. This presentation discusses how Pew Research Center has sought to address this research gap by fielding large, multilingual probability-based surveys of special populations. Examples include the Center’s 2017 Survey of Muslim Americans and the 2020 Survey of Jewish Americans. These studies present numerous challenges in sampling, recruitment, crafting appropriate questions, and weighting. The presentation will also discuss the Center’s methods for studying racial and ethnic populations with the goal of reporting on diversity within these populations, as opposed to treated them as monolithic groups.