Conference Description

In the age of AI and rapidly expanding alternative data sources, household surveys remain indispensable for generating representative and comprehensive insights on jobs and living conditions in low- and middle-income countries (LMICs).  

Despite the value of household surveys for research and policymaking, there is room to improve their quality, timeliness and policy relevance. Developing better survey methods, including approaches leveraging the latest advances in AI, can address common criticisms of surveys and guide more effective policies and interventions to create more and better jobs and pave the path to prosperity. 

Against this background, the Living Standards Measurement Study (LSMS) – the World Bank’s flagship household survey program - in collaboration with the Global Poverty Research Lab (GPRL) at Northwestern University and the World Bank Data Academy, organized the conference Better Data for Better Jobs and Lives: Innovations in Survey Measurement in the Age of AI. It was held in person at the World Bank Headquarters in Washington, D.C. on December 8–9, 2025

This event brought together researchers from academia and international organizations to share and discuss recent advances in the design, implementation, and analysis of household surveys in LMICs. A special focus was on highlighting research for improved measurement of work and employment and applications of AI and machine learning for improving survey measurement. The goal was to strengthen the link between frontier research on survey methods and measurement and the needs of survey implementers, development practitioners, and broader data user community. 

The program showcased completed papers as well as early-stage methodological research, along with invited sessions on strategic priorities and transformative ideas in survey methods and measurement. As such, the conference fostered exchange across research applications at varying maturity levels, while maintaining a strong focus on methodological rigor and relevance for household survey practice. 

We included parallel sessions for full papers and shorter sessions for extended abstracts. The papers that were presented focused on innovations, challenges and empirical findings related to the design, implementation, or analysis of household surveys and measurement of development outcomes. This included, for example, work on questionnaire design and testing, survey operations, sampling and weighting, survey mode, interviewer behavior, data quality assessment, data harmonization, or meta-analysis.  

This meetup aimed to serve as a platform for sharing research that can inform the development of validated, practical, and policy-relevant survey methods.

Participants had the opportunity to engage with World Bank teams and external experts working at the frontier of survey methods and measurement. 

If you have any questions, please contact surveymethods@worldbank.org

KEYNOTE SPEAKER

 

• David Rothschild is an economist at Microsoft Research. He has a PhD in Applied Economics from the Wharton School of Business at the University of Pennsylvania.                                                                                                                                            

He has written extensively, in both academic and popular press. His work pushes the boundaries on varying data and methods: polling, prediction markets, social media, online data, and large behavioral and administrative data.  

 

Rothschild’s work focuses on solving practical and interesting questions including: interactions of humans and LLM-based tools, mapping and updating public opinion, the market for news, effect of advertising, finance, and an economist take on public policy. 

 

MODERATORS

 

• Haishan Fu is the World Bank Group’s Chief Statistician and Director of the Development Data Group. In these roles, she leads and coordinates the Bank’s development data agenda, including oversight of cutting-edge global data public goods, pioneering public-private partnerships, globally trusted technical advisory services and innovative financing solutions, including the World Bank-hosted Global Data Facility.

 

A lifelong advocate for unlocking the power of data to improve lives, Haishan has been at the forefront of the global data discourse as a prominent thought leader, expert advisor and academic and policy researcher for over three decades. 

 

Prior to joining the World Bank, she led the regional statistical development programs in Asia and the Pacific at UNESCAP and was the first Chief of Statistics of UNDP’s Human Development Report. 

Haishan holds a PhD in Demography from Princeton University and a Bachelor’s degree in Economics from Peking University.

 

• Kathleen Beegle is a Lead Economist in the Poverty, Inequality, and Human Development Team of the World Bank’s Development Research Group. Her research focuses on poverty, labor markets, economic shocks and the design and implementation of household survey data collection in developing countries. She also co-leads the new World Bank Center for Research on Women and Jobs.

In previous roles at the Bank, Kathleen served as Program Leader for Human Development in Ghana, Liberia and Sierra Leone, and as Lead Economist for the Gender Group. 

 

She has published extensively in peer-reviewed journals, co-led flagship World Bank reports, and co-authored chapters in several books. Kathleen is an IZA Research Fellow and a member of the IZA G²LM|LIC Advisory Board. She also taught at Georgetown University, USA. 

 

Kathleen holds a PhD in Economics from Michigan State University, USA, and completed a postdoctoral fellowship at the RAND Corporation. 

 

PANELISTS 

 

• Adeyemi Adeniran is the Chief Executive Officer of Nigeria’s National Bureau of Statistics (NBS) and serves as the Statistician General of the Federation.

 

Prior to his appointment, he was the Director of Demographic and Household Statistics at NBS. He has also served as National Coordinator for large-scale national surveys, including the Multiple Indicator Cluster Survey, the National Household Survey on Corruption, and the National Nutrition and Health Survey.

 

Adeyemi is particularly interested in social statistics and has over 30 years of experience as a statistician in both national and international organizations.

 

He holds a Bachelor of Science degree in Demography and Social Statistics from the Obafemi Awolowo University, Ile-Ife, Nigeria. 

 

• Alaka Holla is a Senior Economist in the Human Development Chief Economist Office at the World Bank and the Program Manager of the Bank’s Strategic Impact Evaluation Fund (SIEF), a multi-donor trust fund that supports the use of randomized controlled trials and quasi-experimental methods to inform policies in education, health, water and sanitation, and early childhood development in low- and middle-income countries.

 

Alaka joined the World Bank as a Young Professional and has worked in both research and operational roles, across a wide range of countries and in multilateral initiatives focused on measuring child development and estimating program costs.

 

Her research examines the quality of health care, discrimination, early childhood development and education, and cost analysis. Her work has been published in leading peer-reviewed journals, including the American Economic Review and the Journal of Public Economics.

 

She holds a PhD in Economics from Brown University and completed postdoctoral research at Harvard University and Innovations for Poverty Action.

 

• Andrew Dillon is a Research Associate Professor of Development Economics and Director of the Research Methods Cluster at the Global Poverty Research Lab (GPRL), based at Northwestern University’s Kellogg School of Management, USA. 

 

His research explores how productivity improvements can enhance welfare in low- and middle-income countries (LMICs), as well as identifying the methods and measures that establish these causal relationships. 

 

Andrew’s current work examines market organization in developing countries and the productivity-enhancing investments that households can make in education, new agricultural technologies, health and nutrition. 

 

He is involved in ongoing projects implemented in partnership with governments, private sector firms and NGOs across Burkina Faso, Côte d'Ivoire, Ghana, Mali and Nigeria.    

 

• Charles Lau is the Deputy Global Research Director of Gallup, where he provides technical advice on study design, sampling, questionnaire design, training, data collection, data processing and analysis. He also drives methodological and technical innovations that enhance data quality, cost-effectiveness, and operational efficiency.

 

Before joining Gallup, Charles served as Chief Research Officer at GeoPoll and as Director of the Survey Research Program at RTI International. He has led end-to-end research projects in more than 50 countries.

 

Charles holds a PhD in Sociology and an MSc in Epidemiology from the University of California, Los Angeles (UCLA), as well as a Bachelor’s degree in Sociology from Brown University.

 

• Elisabeth Root is the Deputy Director for Gender, Vulnerability and Health Equity at the Institute for Disease modeling at the Gates Foundation. 

 

Before joining the foundation, she was a Professor of Geography and Epidemiology at The Ohio State University, USA, and served on the leadership team of the Institute for Population Research. 

 

Her academic research examined the long-term and intergenerational impacts of health and development programs on women and children in low- and middle-income countries (LMICs) as well as spatiotemporal patterns of mortality and disease. She also worked on modeling the contributions of social determinants to health and wellbeing. 

 

Elisabeth holds a Ph.D. in Health and Medical Geography from the University of North Carolina at Chapel Hill, USA. 

 

• Frauke Kreuter is Co-Director of the Social Data Science Center and a faculty member of the Joint Program in Survey Methodology at the University of Maryland, USA. She also serves as Professor of Statistics and Data Science at the Ludwig Maximilians University of Munich, Germany.

 

An elected Fellow of the American Statistical Association, Frauke was honored with the 2020 Warren Mitofsky Innovators Award by the American Association for Public Opinion Research.

 

Beyond her academic work, she founded the International Program for Survey and Data Science, designed in response to increasing demand from researchers and practitioners for the appropriate methods and right tools to face a changing data environment.

 

Frauke is also the co-founder of the German-language podcast Dig Deep and the Coleridge Initiative, which advances data-driven research and policy around human beings and their interactions for program management, policy development and scholarly purposes by enabling secure, efficient and effective access to sensitive data about society and the economy.

 

• Lisandro Martín is the World Bank Group Director of the Outcomes Department in the Senior Managing Director's Office.

 

In this role, he oversees the setting of outcome targets, building of internal and external capacity on results and approaches to measuring outcomes across critical cross-cutting themes, such as climate, gender, jobs and private sector capital mobilization. Lisandro also engages with multilateral development banks and key partners to align common metrics and approaches.

 

He has over two decades of experience in the World Bank Group Bank and served in several leadership positions at the International Fund for Agricultural Development (IFAD), contributing to the organization’s impact and results. Earlier in his career he worked at the African Development Bank on operations and impact measurement.

 

Lisandro holds a MSc degree in Public Affairs from Princeton School of Public and International Affairs, where he was a Fulbright scholar. 

 

• Talip Kilic is the Acting Manager and Senior Program Manager of the World Bank Development Data Group's Survey Unit, which houses the International Comparison Program (ICP) – an initiative that collects comparative price data and GDP expenditures to produce purchasing power parities - the Living Standards Measurement Study (LSMS) - the World Bank’s flagship household survey program - and Survey Solutions – an open-source software for design, implementation and management of large-scale surveys and censuses.

 

As a manager, Talip oversees the ICP’s worldwide data collection and processing activities, the LSMS portfolio of face-to-face and phone surveys, research on survey methods and training on survey design, as well as Survey Solutions’ software development, maintenance and training activities. 

 

As a researcher, he focuses on poverty, agriculture and labor in low- and middle-income countries, as well as survey methodology and data integration to improve the quality of household and farm surveys. Talip was also a core team member of the World Development Report 2021: Data for Better Lives. 

 

He holds a PhD in Economics from American University in Washington, D.C., and a Bachelor of Arts degree in Economics and International Relations from Knox College in Galesburg, Illinois, USA. 

Download the agenda here.

Day 1 – December 8, 2025

 

9.00-10.30 Parallel Session 1 

Session 1A: Improving Agricultural Data through Technological and Methodological Innovation 

Chair: Gero Carletto, Senior Manager, World Bank

 

Presentations: 

 Weather or Not: Leveraging in Situ Sensors to Improve Measurement of Weather, Climate Variability, and Their Links to Socioeconomic Outcomes in Low-Income Countries, Anna Josephson, Associate Professor, University of ArizonaDiscussant: Sydney Gourlay 

 

Presentation

 

 Scanning for Soil Health: Assessing the Accuracy and Scalability of Innovations for In-Situ Soil Measurement, Sydney Gourlay, World BankDiscussant: James Stevenson

 

Presentation

 

Agricultural labor, technological change and structural transformation: a measurement agenda, James Stevenson, SPIA-CGIARDiscussant: Jeffrey Dickinson

 

Presentation

 

Surveying Land with Satellite Imagery: A Pie in the Sky? Jeffrey Dickinson, Dollar Scholars FoundationDiscussant: Anna Josephson

 

Presentation

 

Recording of Session 1A

Session 1B: Jobs I – Data Innovations in Employment and Time-Use Surveys   

Chair: Amparo Palacios Lopez, Program Manager, World Bank

 

Presentations: 

 Things take the time they take: Lessons on measurement error from high-frequency time use surveys, Tammy McGavock, Grinnell CollegeDiscussant: Farnaz Safari

 

Presentation

 

Bias from Aggregation? Rethinking Occupational Transitions & Diversification with High-Frequency Data, Farnaz Safari, Cornell UniversityDiscussant: Tyler Rossow 

 

Presentation

 

 Transforming Zambia’s Labour Force Survey Using Artificial Intelligence, Tyler Rossow, IMFDiscussant: Tammy McGavock

 

Presentation

 

Recording of Session 1B

 

11.00-12.30 Opening Panel: Priorities in Survey Methods and Measurement: Data Gaps, Methodological Challenges, and Opportunities 

Moderator: Kathleen Beegle, Lead Economist, World Bank

 

Panelists:

Andrew Dillon, Research Associate Professor of Development Economics, Director of Research Methods Cluster in the Global Poverty Research Lab (GPRL)

Alaka Holla, Lead Economist, Office of the Chief Economist for People & Office of the Chief Statistician, World Bank  

Talip Kilic, Acting Manager, Senior Program Manager, Survey Unit, World Bank & Co-Chair for United Nations Intersecretariat Working Group on Household Surveys (ISWGHS)

Elisabeth Root, Gates Foundation Deputy Director Infectious Diseases

 

Recording of the Opening Panel

 

1.30-3.00 Parallel Session 2

Session 2A: Jobs II – Firms, productivity, and welfare

Chair: Andrew Dillon, Research Associate Professor of Development Economics and Director of Research Methods Cluster in the Global Poverty Research Lab (GPRL), 

Northwestern University

 

Presentations:

 When Labor Counts: Measurement Timing, Agricultural Productivity, and Gender Gaps in Nigeria, Paula Gonzalez, Northwestern UniversityDiscussant: Yannick Markhof

 

Presentation

 

 Lies in disguise: Respondent behavior and survey measurement uncertainty, Yannick Markhof, ETH ZurichDiscussant: Akuffo Amankwah

 

Presentation

 

 Using Household Surveys and Specialized Enterprise Surveys to Measure Informal Enterprises: Methodological Evidence from Ghana, Akuffo Amankwah, World BanK. 

 

Presentation

 

Recording of Session 2A

Session 2B: Methodological Challenges and Advances in Poverty Measurement

Chair: Dean Jolliffe, Research Manager, World Bank

 

Presentations:  

The Missing Poor, Ana Radu, Princeton UniversityDiscussant: Robert Andrew Marty

 

Presentation

 

Global poverty estimation using private and public sector big data sources, Robert Andrew Marty, World BankDiscussant: Daniel Mahler

 

Presentation

 

The Impact of Questionnaire Design on Global Poverty, Daniel Mahler, World BankDiscussant: Amparo Palacios-Lopez

 

Presentation

 

Intrahousehold allocation of public and private goods: Evidence from Bangladesh, Amparo Palacios Lopez, World BankDiscussant: Ana Radu

 

Presentation

 

Recording of Session 2B

 

3.30-5.00 Parallel Session 3

Session 3A: Jobs III – Informality, Job Quality, and the Future of Labor Statistics

Chair: Louise Fox, Nonresident Senior Fellow, Brookings Institution

 

Presentations: 

 Job Quality in times of AI: Measurement Innovations and Data Challenges for the Future of Work, Kirsten Sehnbruch, International Inequalities Institute, LSEDiscussant: Zander Prinsloo

 

Presentation

 

Estimating employment transitions while accounting for classification error: evidence from South Africa, Zander Prinsloo, World BankDiscussant: Gisella Kagy

 

Presentation

 

Beyond informality: The importance of residential-based businesses, Gisella Kagy, University of Wisconsin-Madison. Discussant: Parth Chawla

 

Presentation

 

Supply Chain Networks, Risk Sharing and Resilience to Shocks Among Small Firms in Tanzania, Parth Chawla, University of California, Davis. Discussant: Kirsten Sehnbruch

 

Presentation

 

Recording of Session 3A

Session 3B: Advances in Human Capital, Health, and Nutrition Measurement

Chair: Wendy Cunningham, Lead Economist, World Bank

 

Presentations:

 The Impact of Respondent Selection on Dietary Diversity and Quality Measurement in Surveys: Findings from Ethiopia and Nigeria, Alemayehu Ambel, World Bank. Discussant: Smita Das

 

Presentation

 

 Socioemotional Skills in Sub-Saharan Africa: Validating and Comparing Behavioral and Self-Reported Measures, Smita Das, IPA. Discussant: Ivette Contreras

 

Presentation 

 

 Do men really have greater socio-emotional skills than women? Evidence from Tanzanian youth, Clara Delavallade, World Bank.Discussant: Alemayehu Ambel

 

Presentation

 

 Uncovering Hidden Care Network: Evidence from a Childcare Survey Experiment in El Salvador, Ivette Contreras, World BankDiscussant: Clara Delavallade

 

Presentation

 

Recording of Session 3B

 

Day 2 – December 9, 2025

 

9.00-10.30 Parallel Session 4

Session 4A: AI for Text Analysis and Data Classification 

Chair: Philip Wollburg, Senior Economist, World Bank

 

Presentations:  

 Understanding the Microstructure of Labor Demand: Innovations in Measuring Firms’ Hiring Beliefs and Decision Processes in Ethiopia, Tsegay Tekleselassie, Wellesley CollegeDiscussant: Peter Brueckmann

 

Presentation

 

 AI-Powered Multi-Level Classification Tool for Survey Data: Improving Open-Ended Response Coding, Peter Brueckmann, RowsquaredDiscussant: Roy Burstein

 

Presentation

 

Large language models for analyzing open text in global health surveys: why children are not accessing vaccine services in the Democratic Republic of the Congo, Roy Burstein, Gates FoundationDiscussant: Aletheia Donald

 

Presentation

 

  Whose Job Is it? Implicit Gender Bias Toward Occupations in India and Uganda, Aletheia Donald, World BankDiscussant: Tsegay Tekleselassie

 

Presentation

 

Recording of Session 4A

Session 4B: Methodological Evidence from Mixed-Mode and Hybrid Data Collection

Chair: Charles Lau, Deputy Global Research Director, Gallup

 

Presentations:  

  The effect of survey mode on data quality: Experimental evidence from Nigeria, Yannick Markhof, ETH ZurichDiscussant: Steven Glazerman

 

Presentation

 

  Survey Mode Effects on Financial and Livelihood Outcomes: Evidence from a Modes Experiment in Uganda, Steve Glazerman, IPADiscussant: Ismael Yacoubou Djima

 

Presentation

 

 Using Survey-to-survey Imputation for More Frequent and Lower-cost Poverty Estimates: Evidence from Malawi, Ismael Yacoubou Djima, World BankDiscussant: Juan Baron

 

Presentation

 

 Collecting Functional Disability Data in Children via Phone Surveys: Overcoming Sensitive Data Challenges using Randomized Experiments, Juan Baron, World BankDiscussant: Yannick Markhof

 

Presentation

 

Recording of Session 4B

 

11.00-12.30 Keynote - Surveys and AI: State of the evidence, opportunities and pitfalls, David M. Rothschild, Microsoft Research

 

Artificial intelligence is reshaping how surveys are designed, implemented, and analyzed, offering powerful new tools as well as real risks for data quality, reliability, and equity. In this keynote, David M. Rothschild explores what AI can and cannot do across the survey lifecycle, with a focus on the realities of producing reliable development data in low-capacity settings.

 

The session will highlight emerging opportunities, key pitfalls, and what these shifts mean for the future role of household surveys and official statistics in an AI-driven information and evidence ecosystem.

 

The keynote will combine a focused lecture with a fireside-style conversation and conclude with an open Q&A, creating space for both conceptual grounding and interactive discussion

 

Presentation

 

Recording of Keynote 

 

1.30-3.00 Parallel Session 5

Session 5A: AI Applications in Survey Data Collection

 

Presentations:  

 Harnessing AI to Map Complexity in Sub-Saharan Africa: An AI-Powered Conversational Agent for Measuring Multi-Activity Agricultural Livelihoods, Amine Amar, Al Akhawayn UniversityDiscussant: Clemence Pougue

 

Presentation

 

Leveraging AI for Screening Mental Health Assistance Needs, Clemence Pougue Biyong and Quy Toan Do, World BankDiscussant: Ceren Koca

 

Presentation

 

AI-Powered Voice Agents for Behavioral Data Collection: Comparing Quality and Efficiency with Human Interviewers in Kenya, Ceren Koca, Human TruthsDiscussant: Steven Glazeman

 

Presentation

 

Automating Survey Quality Control Using AI: Applications to Employment and Livelihood Measurement. Noureini Sayouti, IPA. Discussant: Amine Amar

 

Presentation

 

Recording of Session 5A

 

Session 5B: Machine Learning and Data Integration 

Chair: Kathleen Beegle, Lead Economist, World Bank

 

Presentations:  

 CLARE: A Causal Machine Learning Approach to Resilience Estimation, Pierluigi Montalbano, Sapienza University. Discussant: Alemayehu Ambel

 

Presentation

 

Predicting Drinking Water Quality from Household Survey Data using Machine Learning: Case Study from Tanzania, Alemayehu Ambel, World Bank. Discussant: Losira Nasirumbi Sanya

 

Presentation

 

Measuring Women’s Agency with Unsupervised Machine Learning Approaches, Losira Nasirumbi Sanya, Makerere University. Discussant: David Newhouse

 

Presentation

 

Improving Estimates of Human Capital Outcomes in Developing Countries by Integrating Survey and Geospatial Data, David Newhouse, World BankDiscussant: Pierluigi Montalbano

 

Presentation

 

Recording of Session 5B

 

3.30-5.00 Closing Panel: Repositioning Household Surveys in the Age of AI 

Moderator: Haishan Fu, World Bank Group Chief Statistician and Director of Development Data Group

 

Panelists: 

Yemi Adeniran, Nigeria NBS Statistician General

Frauke Kreuter, Chair of Statistics and Data Science in Social Sciences and the Humanities (SODA) & Deputy Head of Department, LMU Munich & University of Maryland

Charles Lau, Deputy Global Research Director, Gallup

Camilo Mondragon-Velez, Manager of the Outcomes Measurement Unit, Department for Outcomes, World Bank Group

 

Recording of the Closing Panel