Xi Song is an associate professor of sociology and demography at the University of Pennsylvania. Her research interests include social stratification and mobility, poverty, inequality, population studies, and quantitative methodology. Her current research examines social, economic, and demographic processes that govern the persistence of inequality across life stages and generations in the United States and China from the eighteenth century to the present.

As a demographer, Song uses mathematical, statistical, and computational methods to study the rise and fall of families in human populations across time and place. Her past research has demonstrated the values of genealogical microdata for studying long-term family and population changes. These data sources include historical data compiled from family pedigrees, population registers, administrative certificates, church records, and surname data; and modern longitudinal and cross-sectional data that follow a sample of respondents, their offspring, and descendants prospectively or ask respondents to report information about their family members and relatives retrospectively.

As a quantitative methodologist, Song has developed Markov chain demography models for genealogical processes, population estimation methods for overlapping lifespan between generations, multivariate mixed-effects location-scale models for inter- and multigenerational data, and weighting methods for reconciling prospective and retrospective mobility estimates.