Beyond the data gap: Unlocking insights to identify and analyse hard-to-survey rural and vulnerable populations
Hybrid Event, 06/03/2025
Date: Thursday, 6th March 2025,
Time: 15h-18h (Rome-CET), 9h-12h (USA-ET), 6h-9h (USA-PT)
Location (hybrid): FAO Headquarters, room D587 (Coffee & nibbles provided)
This seminar explores the concept of "hard-to-survey" (HTS) populations, particularly in agricultural and rural contexts, the methods for identifying such groups and their relevance to policymakers and practitioners. The seminar provides an overview of key concepts and their implications, beginning with a definition of these groups. Tourangeau (2014), categorizes these groups into four types: 1) Populations that are difficult to sample or identify, 2) Populations that are difficult to locate or contact, 3) Groups with members who are reluctant to participate, and 4) Groups that are willing to participate but difficult to interview.
By examining these distinctions, participants will gain insights into the challenges and characteristics of these groups, particularly in rural and vulnerable settings. The challenges of identifying HTS populations stem from the limitations of traditional data collection methods, such as censuses and sample surveys, which often lack detail and are conducted infrequently. Challenges arise from factors such as seasonal labor patterns, reliance on informal or illegal employment, restrictive social norms, geographic isolation, limited resources, and legal obstacles to data collection.
Participants will learn about innovative methods for identifying and studying these populations. The discussion will explore quantitative and mixed-method approaches, illustrated by real-world examples from intermediate actors in agri-food value chains in Uganda and Bangladesh, marginalized and vulnerable groups such as Roma communities in Italy, and migrant agricultural workers in California.
Traditional methodologies for identifying these populations often rely on social networks, where an initial sample of individuals—known as "seeds"—helps recruit others with the target characteristics. However, advances in geospatial and satellite data, along with improved access to government registries, provide new opportunities to enhance these traditional approaches.
For instance, Lee and Beatty (2024) used remote sensing land-use data and mobile phone services to connect day laborers with agricultural fields. Similarly, Betty, (2024) leveraged the IPUMS Geomarker tool to integrate survey or census data with AI-generated datasets and government registries, allowing for the geolocation of households participating in the Supplemental Nutrition Assistance Program (SNAP).
Finally, the session will critically analyze the limitations and trade-offs of these methods, considering factors such as cost, representativeness, and accessibility. As part of the THINK-PA Technical Network, this discussion introduces key theories, frameworks, and methods to enhance understanding of identifying and studying hard-to-reach populations in diverse contexts.
SPEAKERS:
- Brad Edwards: Vice President and Lead Scientifical/Methodology Advisor, at Westat
- Valeria De Martino and Nadia Nur: Senior Researchers, Italian National Institute of Statistics – ISTAT)
- Jeffrey Bloem: Research fellow - Markets, Trade, and Institutions Unit – International Food Policy Research Institute (IFPRI)
- Prof. Tim Beatty: Chair of the Department of Agricultural and Resource Economics, University of California, Davis