Methodological Challenges in Estimating the Size of the Jewish Population in Greater Philadelphia

• Activity Number: 475 - Recent Advances in Sample and Survey Design
• Type: Contributed
• Date/Time: Thursday, August 6, 2020 : 10:00 AM to 2:00 PM
• Sponsor: Survey Research Methods Section
• Author(s): David Marker* and Shelley Brock Roth and Darby Steiger and Jill DeMatteis
• Companies: Westat and Westat and Westat and Westat
• Keywords: Address-based sampling; Sponsorship; Coverage; Push-to-web

Abstract

The size of the Jewish community in Greater Philadelphia was last measured in 2009, by sampling membership lists and landline telephone numbers through random-digit dialing (RDD). To update the study in 2019 we replaced RDD with address-based sampling, greatly expanding the coverage. Rather than conducting a telephone survey, the 2019 survey was self-administered, using a push-to-web design followed by a paper questionnaire for non-respondents. Historical surveys had screened households by asking if anyone was Jewish; to get more unbiased estimates, we used more neutral language to screen sampled households. We also included a split-plot design that allowed us to determine the impact of sponsorship, using neutral language compared to identifying the survey sponsor. All of this interacted with the data collection environment of early 2019, including reactions to offering the survey in English and Russian. This paper will present each of the methodological advancements that were made, and the impact on response behavior, data quality, and population trends.