Uruguay's 2023 combined census: Integrating administrative registers and traditional enumeration

Abstract

Uruguay's 2023 Population Census implemented a combined methodology that integrated traditional field enumeration with administrative registers to achieve near-complete population coverage in the absence of a central population register. This paper describes the design, implementation, and validation of this approach. Approximately 10% of the usual resident population was added through administrative enumeration to compensate for non-response and omissions in the field census. All individuals included correspond to real persons identified in administrative records; no synthetic persons, i.e., model-based population imputations, were included in the final database. A statistical population register constructed from multiple administrative sources, combined with a “Signs of Life” model to verify current residency, was used to identify eligible individuals and assign them to households or small geographic areas where possible. Data quality was assessed through multiple independent validation exercises, including a post-enumeration survey, analysis of operational outcomes from the web-based census, and long-term demographic reconstruction. These evaluations show that administrative enumeration effectively corrected the field undercount without introducing meaningful demographic or geographic bias. The Uruguayan experience demonstrates that combined censuses can substantially improve coverage and data quality in countries without central population registers, offering a replicable pathway toward census modernization based on integrated administrative data.