Workshops

Full count census data represent a rich class of source material for social scientists. By linking individuals and families across censuses, the IPUMS Multigenerational Longitudinal Panel (MLP) allows analysts to trace the characteristics of individuals over their lives and follow families across multiple generations. These data can be leveraged for innovative analyses of spatial change, support consistent analyses of the impact of neighborhood context on individual outcomes, and permit studies of the smallest subpopulations. However, linking and analyzing data at this scale is different than working with sample data. This workshop, hosted by the IPUMS Big Microdata Network, will include presentations designed to lower barriers to utilizing linked, full count census data for demographic and health research, with a specific emphasis on data availability and access, linking methods, and analytical considerations when working with linked data. 

The National Health and Aging Trends Study (NHATS) is an NIA-funded national panel study that conducts annual interviews with a representative sample of Medicare beneficiaries ages 65 and older. Designed as a platform for scientific study of late life disability trends and trajectories, NHATS fosters research to guide efforts to reduce disability, maximize health and independent functioning, and enhance quality of life at older ages. Caregivers to NHATS participants are interviewed in the supplemental National Study of Caregiving (NSOC) to provide the perspective of family and friends who help older adults experiencing limitations in daily life. This workshop aims to introduce researchers to the NHATS and NSOC study designs, data file structure and access, and enclaves for linking restricted files (geographic, genetic) and Medicare claims data. By the end of the session, users will 1) gain knowledge about the survey design and content of NHATS and NSOC; 2) be familiar with how to access and use NHATS and NSOC data; and 3) understand other resources (e.g., geographic-based data) and learning opportunities available. More information about the study is available at nhats.org.

In an era of rapid data collection and evolving research technologies, designing surveys and selecting the right data collection approach are critical to producing valid, reliable, and actionable insights. This interactive workshop brings together three essential topics—best practices in survey design, the strategic use of probability-based online panels, and practical applications for fit-for-purpose research.

This four-hour workshop introduces the theory and practice of demographic projections in an accessible and practical way. Participants will learn how to implement population projections with fixed or time-varying schedules of fertility, mortality, and migration, grounded in the probabilistic median scenario from the UN World Population Prospects (WPP).

The workshop covers both standard age-structured models (Leslie matrix) and multistate projection models, a method often perceived as technically demanding but made simple and fully operational in this workshop. Participants will gain hands-on experience in simulating population dynamics not only by age and sex but also across additional dimensions such as region, health status, marital status, labor force participation, and race/ethnicity.

A key feature of the workshop is the ability to generate projection scenarios and counterfactuals in seconds. Participants will be able to modify demographic rates—such as fertility decline, mortality improvement, or transition probabilities between social states—and immediately see how these assumptions shape future population outcomes. This makes the workshop particularly useful for researchers interested in exploring alternative scenarios, policy impacts, or demographic what-ifs.

The workshop will cover core functionalities of leslie, a Stata-based program specially designed for demographic projections.

This workshop is designed to teach basic to intermediate tools to implement panel data analysis using Stata. The participants will learn how to formulate their regression models, estimate the parameters, and interpret the results. We will discuss theoretical aspects of panel data modeling, and we will work with real data examples to illustrate how to use panel data estimators to fit our model. We will cover linear panel data models, dynamic panel-data estimators that explicitly model the dynamic structure of panel data, and panel data models for binary outcomes.

The Future of Families and Child Wellbeing Study (FFCWS) is the longest running contemporary birth cohort study in the US based on a national probability sample. The FFCWS provides free access for researchers to its public use data, which covers a wide range of topics, including education, employment, housing, relationships, parenting, health, income, and more for a sample of ~5000 families including a large number of Black, Hispanic, and low-income families. It also contains restricted use contract data on residential contexts, school contexts, and biological and health data. We welcome those who are interested in learning about using the FFCWS data for the first time, and current data users who would like a deeper dive into navigating our data, as well as those who are just curious to learn more about this study. This workshop will introduce researchers to 1) a brief overview of the history and data collection of the FFCWS including our ongoing field work at Year 27 ; 2) an overview of our latest contextual data files;  3) a guide to the file structure, contents, and data conventions used in FFCWS including tips and advice from experts in FFCWS data support; and 4) a tutorial on the contents and use of our metadata web interface. Our goals for this workshop are to provide a solid background for working with FFCWS data and to support participants as they begin variable exploration and selection relevant to their own research questions.

This workshop introduces three pioneering city-based longitudinal datasets from China: the Hong Kong Panel Study of Social Dynamics (HKPSSD), the Shanghai Urban Neighborhood Survey (SUNS), and the Guangzhou Metropolitan Panel Survey (GPS). Together, they open new frontiers for studying social and spatial inequality, family life, health and well-being, as well as neighborhood change in rapidly transforming urban contexts. Participants will gain practical insights into accessing, merging, and linking these datasets with external sources such as census, administrative, and spatial data. The session will also highlight opportunities for comparative and cross-city analyses, showcasing how these resources can support innovative research on urban transformation in China. Ideal for researchers and students eager to explore cutting-edge data and collaborative opportunities in population and urban studies.

The US Panel Study of Income Dynamics (PSID) is the world’s longest-running longitudinal household panel study. It is used in the fields of demography, economics, sociology, public health, and public policy to investigate individual and family socioeconomic status, health, and well-being in longitudinal and intergenerational contexts. This workshop familiarizes new and prospective users with the design, content, and applications of the main PSID interview and two youth-centered supplements: The Child Development Supplement (CDS) and Transition into Adulthood Supplement (TAS). One segment of the workshop will introduce the available genomic data collected from CDS children age 0-17 years and their primary caregivers, as well as additional biomarker collections. The workshop also provides a hands-on introduction to data access, key data files, and user education resources.

This workshop introduces the theory and practice of demographic projections in an accessible and practical way. Participants will learn how to implement population projections with fixed or time-varying schedules of fertility, mortality, and migration, grounded in the probabilistic median scenario from the UN World Population Prospects (WPP).

This workshop introduces the fundamentals of digital trace data for migration studies. Topics include data collection and analysis, as well as the main advantages and challenges of using Facebook, Instagram, Twitter/X, Bluesky, LinkedIn, Google Trends, Wikipedia, and bibliometric data. The workshop includes a brief section on Google Trends and Wikipedia as case studies to demonstrate how to gain insights into migration patterns through migrants’ information-seeking behavior. The workshop then discusses the main sources of bias and potential ethical issues, and concludes with a general discussion on digital trace data in migration studies.

Spatial demographic research is a booming sector of population science that can, to beginning and veteran demographers alike, appear complicated and daunting. This workshop is designed to orient new potential users to the exciting capabilities and applications of spatial demographic research—and the range of new empirical questions that it invites. We will pair this enthusiasm with a practical introduction to finding data, using simple software tools (Stata, R, QGIS, Geoda), and making appropriate choices about describing and analyzing spatial demographic phenomena. All users will walk away with an understanding of the menu of options in spatial demographic analysis, a compass for how to make research decisions and find more information, and exposure to tools that enable them to get started immediately.

Artificial intelligence can accelerate your research workflow and help solve complex analytical challenges. This workshop covers practical AI applications for researchers. You will learn approaches for common research challenges where AI can provide solutions. We begin with an introduction to how AI works, and what it can – and cannot – help with. Our primary focus is on real world use cases and lessons learned from them. For example, we will cover ways AI can help with manuscript preparation as well as a range of data analysis challenges. We close with small group sessions where participants explore AI opportunities within their specific research contexts.

There is now a widely held consensus that more transparent research provides better outcomes for individual researchers, for scientific communities, and for society at large. Most efforts towards transparent research to date have focused on quantitative research and the data that underpin it. The demand for scientific transparency, however, is equally relevant to qualitative research and data, even though sharing qualitative data poses unique challenges, both logistically and ethically. Qualitative researchers, therefore, may want to share their data and make their work more transparent, but lack the tools and resources to do so successfully.

The last decade has seen an exponential rise in the interest in sequence analysis (SA) in the social sciences, especially in population studies. With this half-day workshop, we aim to introduce SA to novices as well as current users and to highlight common mistakes in various methodological decisions, such as choosing distance measures and conducting sequence cluster analysis. The topics include the SA research process, using R for SA, sequence visualization, choosing an appropriate distance measure beyond the often-misused OM, conducting sequence cluster analysis with the knowledge from stability assessment and comparisons to the null model, and strategies of multidomain/multichannel SA beyond a reliance on OM-based multichannel distance computation. The goal is to help interested PAA 2026 attendees gain a better understanding of this popular yet less understood demographic methodology.

The U.S. Census Bureau strives to make each census better than the last, adapting and enhancing our methods, tools and processes to improve on the prior decade's experience. This workshop will share updates and solicit stakeholder feedback on several ongoing research projects that will inform the selection, design, implementation, and evaluation of the Disclosure Avoidance System that will be used to protect the confidentiality of individuals' and households' responses to the 2030 Census. This important research includes the development of new algorithmic approaches to improve the accuracy of high value statistics while also ensuring effective protections; novel approaches to statistical post-processing that can improve accuracy and produce measures of uncertainty for Census estimates; research to bridge differences in reporting categories for race and ethnicity and their implications for confidentiality protection; and innovative approaches to generating synthetic Census microdata files that can be used to test and evaluate Disclosure Avoidance System performance and tuning, as well as to publish demonstration statistical products for external stakeholders to evaluate and provide iterative feedback on our ongoing research and testing. These research projects represent key components of the Census Bureau's broader 2030 Census Disclosure Avoidance Research Program. Data users, researchers, civil society groups, and other Census stakeholders are encouraged to participate in this workshop to provide their important feedback and reactions to this ongoing research and to inform future decision-making on the selection, design, and implementation of the 2030 Census Disclosure Avoidance System.

Bayesian modeling approaches are widely used for producing estimates and forecasts of demographic and global health indicators, and for assessing population differentials. This workshop provides a conceptual introduction to such models, complemented with hands-on exercises and analyses in R and Stan.