Positioning household surveys for the next decade
What is it about?
The paper identifies eight technical priority areas for innovations in household survey design, implementation, and analysis. They include: (1) enhancing the interoperability and integration of household surveys; (2) designing and implementing more inclusive, respondent-centric surveys; (3) improving sampling efficiency and coverage; (4) scaling up the use of objective measurement technologies, (5) building capacity for CAPI, phone, web, and mixed-mode surveys; (6) systematizing the collection, storage, and use of paradata and metadata; (7) incorporating machine learning and artificial intelligence for data quality control and analysis; and (8) improving data access, discoverability, and dissemination. With these in mind, the paper also presents a set of recommendations for fostering enabling environments at the national and international levels to support the production of more and higher-quality household survey data that are affordable and responsive to policy needs.
Why is it important?
The paper was written at a time when household survey programs around the world have been suffering due to reduced funding, increased concerns over their quality, and the disruption of traditional fieldwork operations by the pandemic. At the same time, these programs have been challenged by the rise of non traditional data sources, alongside data science skills such as machine learning that are relatively unfamiliar to national statistical offices. However, given the proven agility of NSOs during the COVID-19 pandemic, through experimenting with new mode of data collection and new data sources, we hope that this paper will help support further innovation in countries, turning the various crises faced by household survey programs into opportunities For example, decreased response rates can motivate the establishment of strong relationships with data users and survey respondents, reduced resources can provoke the use of innovative approaches to increase survey efficiency and the integration of survey data with other sources, and the new “competing” data sources and skills can drive NSOs towards building partnerships and taking a more active role as data stewards.