Population censuses

Population censuses were among the first type of data used to study migration and often directly ask respondents about their current residence, their residence at a previous point in time, for example at birth or at a specified time interval prior to enumeration or about their length of stay in their current residence (Kirchberger, 2021). Censuses produce statistics on migrant stocks, socio-economic characteristics, migrant flows (limited) and some emigration figures. The UN Statistics Division recommends including three core variables in population censuses to identify international migrants. They are: (1) country of birth, (2) country of citizenship and (3) year or period of arrival in the country for foreign-born persons (IOM, 2020a).

The UNDESA Statistics Division’s 2022 Handbook focuses in particular on the use of population and housing censuses as a source of information for international migration statistics. As discussed in the Handbook, the population and housing census enables migrant stock and migrant flow to be estimated in an internationally comparable manner through the inclusion in the census questionnaire of a few key questions, for example, on country of birth, country of citizenship, country of residence one or five prior to enumeration and year of arrival in the country. Integrating “the year or period of arrival in the country” for foreign-born persons can be used to determine whether the person is an international migrant, i.e., residing in the country for at least 12 months or most of the last 12 months, and to distinguish between migrants who recently arrived and those who arrived a while ago, thereby permitting an analysis of migrant integration. The question might also be directed to persons who have lived abroad regardless of their country of birth, if returning citizens are also of policy interest (Global Migration Group, 2017).

Figure 1: Availability of census data on migration

Figure 2: Examples of questions on migration from United States Census Bureau American Community Survey

Population censuses are a major source of data on migrant stock but can also be used to (roughly) estimate international migration flow for specific periods ending with the time of the census. Data from the current and previous census can be used to estimate the net migration flow for the period between the two censuses (Global Migration Group, 2017).

What are the strengths of population censuses?

The most distinct strength of the population census is its universal coverage, as it includes all residents and therefore also international migrants, or foreigners who have established residence. Another strength of the census is its ability to collect a substantial amount of information on each individual, which accounts for its potential to delineate international migrants based on certain demographic and socioeconomic characteristics (IOM, 2020a). For example, censuses allow for a comparison of access to housing, employment and social services between migrants and non-migrants (Global Migration Group, 2017) The overwhelming majority of countries in the world conduct a population and housing census at least once in 10 years, and the censuses have been the most commonly available source of migration statistics. This means that there is greater uniformity across countries in the type of data on international migration produced by censuses compared with any other data-collection system (UNDESA, 2022).

What are the limitations of population censuses?

There are, however, several limitations. First, since censuses are generally carried out only once every decade (or once every five years in a few countries), they cannot capture many of the current trends in international migration in a timely fashion. Second, there is limited space for including include direct, probing questions related to the causes and consequences of migration, which would need to be detailed in nature. Third, some countries employ random sampling strategies in their population census and the spatial distribution of migrants throughout the population might not be random. While censuses seek to cover the whole population, people are inevitably missed, and migrants are especially susceptible to not being enumerated. Finally, many international migration-related questions require responses regarding events that occurred within a specified period in the distant past, which increases the potential for recall bias (UNDESA, 2022).

Case study: Technical cooperation between IOM Brazil and The Brazilian Institute of Geography and Statistics (IBGE) on the 2022 Demographic Census (Migration Network Hub, 2023)

For the 2022 Census, IOM provided support in:

-Translation of the questionnaire and information materials into Spanish, English, French, Mandarin and Arabic;

-Mobilization and dissemination of the census among the migrant population, through the promotion of information sessions and dissemination of information materials;

-Accompanying surveyors in conversations with community leaders and in the administration of questionnaires in areas where IOM operates;

-Training institute census takers on migration, indigenous migrant populations, and spontaneous settlements;

-Disseminating monthly reports containing georeferenced statistics and information on Venezuelan refugees and migrant populations outside of shelters by using IOM’s DTM methodology;

-Participating in discussions related to the analysis of census results, reviewing publications related to migration topics, and constructing questionnaires for the institute's intra-censal sample surveys (i.e., sample surveys on diverse topics, such as economy, health, environment, among others)

Chapter overview