Abstract

Households are the most basic units of coresidence among humans. They are fundamental institutions for social and economic reproduction across societies. Yet comparative data on how households and living arrangements are changing worldwide continues to be lacking today. The CORESIDENCE database aims to fulfill this data gap. It is based on three main repositories of global-scale individual microdata. First, we rely on 313 census samples from the International Integrated Public Use Microdata Series (IPUMS-I) in 94 countries worldwide. Second, we use 313 Demographic Health Surveys (DHS) at the household level for 75 countries. Third, we rely on 117 samples from the European Union Labor Force Survey (EU-LFS) for 24 countries. In addition, the analysis uses country-specific surveys and censuses not present in the previous repositories. These include the EU Statistics on Income and Living Conditions (EU-SILC) survey for the Nordic countries, the Household, Income, and Labour Dynamics in Australia (HILDA) survey and the Household Income and Expenditure Survey for South Korea. All these data sources allow the identification of family relationships among household members and provide information, among other variables, on age, sex, relationship to head, marital status, and educational attainment. At the individual level, the final dataset consists of over 148 million individual records for 142 unique countries, representing more than 98% world’s population spanning from the 1960s to the present.

This paper presents the extensive harmonization process of data sources to create and develop the CORESIDENCE database and introduces its main features and indicators. The CORESIDENCE database focuses exclusively on private households. Household indicators are organized at two levels: household level and individual level. From a household perspective, CORESIDENCE provides information on household size and composition. The composition includes data on age structure, relationship to head, and headship characteristics. At the individual level, CORESIDENCE adds information on living arrangements. It identifies who lives with whom to inform about the coresidential context of each individual within a household. Data on living arrangements are arranged by sex, age, and educational attainment. We provide national and subnational detail and contextual level data on basic economic and demographic conditions. The database allows for the examination of coherence across data sources. We compare results across various data sources on basic indicators and discuss the most significant inconsistencies.

The CORESIDENCE dataset offers numerous opportunities for empirical research on family change. By leveraging newly available large-scale international census and survey microdata, CORESIDENCE allows the development of comparative social theory by analyzing variation in family forms on several geographic, cultural, and developmental scales across social groups and over time spans from one to five decades. In so doing, researchers will be able to contribute to the theoretical and empirical understanding of the background factors of changes in family systems while highlighting their diversity.