Statistical Journal of the IAOS

For millennia the census has been an instrument of public administration; it was used in a wide variety of formats counting different populations in different scientific disciplines. Over time a standard format emerged based on the scientific method; the population and housing census was formalised in the second half of the 19th century. Its function was the enumeration of every individual in the population of a specific geographical area at specific time, through direct contact of the households to obtain information about each household member. Practically parallel another administrative instrument was developed: the register, an officially sanctioned list of objects or events. Throughout time these two administrative instruments with complementary functions have contributed to the management of society. Since the second half of the 20th century the sample survey became another source of statistical information. At the same time some countries started to redesign their statistical organisation favouring registers and replacing the census, with a system of combined registers and sample surveys. Proponents of this approach indicate that this procedure has no theoretical basis and that there are methodological challenges with its implementation. This paper will review these developments and make a call for a science-wide review.

This overview focuses on the implementation of the 2020 World Population and Housing Census Programme; the methodological framework and international recommendations for census-taking in the 2020 Round of Censuses (2015–2024); the state of the art of the world’s census-taking midway through the census decade; the preliminary results of the survey launched by the United Nations Statistics Division; and the trends in census-taking after the 2020 round.

The 2020 census round (2015-2024), is being implemented in a changing context, marked by widespread adoption of new technologies, increasing demands for disaggregated data for implementation of SDGs, capacity challenges in many national statistics offices, funding challenges, as well as conflict and unrest which are affecting census coverage in many countries. This article details the UNFPA strategy to ensure that everyone is counted, and accounted for, in the 2020 census round. The strategy responds to the changing contexts and builds on lessons learned in previous rounds. Within this framework, UNFPA provides technical and operational support (tools and capacity strengthening); promotes adoption of new and innovative technologies and methodologies; advocates for wider utilization of census data, including free and open access to data while ensuring confidentiality; leverages institutional partnerships at all levels for coordinated census support; strengthens resource mobilization, and advocates for the inclusion of questions on disability and migration to support efforts to leave no one behind.

For over 65 years, the U.S. Census Bureau’s International Programs (IP) has been providing capacity-strengthening-oriented technical assistance and training to National Statistical Offices around the world. This paper describes how IP has modernized and enhanced its public domain tools and technical assistance strategies to support low-and middle-income countries prepare for the global 2020 Round of Population and Housing Censuses.

Globally, with increase in need for evidence for relevant and effective responses in this century, holistic stocktaking and documentation of subjects has been heightened. In the same vein, Malawi has successfully conducted censuses from 1966 to 2018. Unlike other censuses, the latest round adopted use of Satellite imagery for mapping, Computer Assisted Personal Interviewing (CAPI) technology for data capturing and wireless and on-line data transmission to significantly reduce processing time and improve reliability and timeliness of census results.

However, in the face of CAPI and built-in data quality control features, data processing is irreplaceable and key since censuses are never perfect and errors do occur. Further, whilst incorporation of technology demonstrates benefits and merits, the approach comes with some logistical, financial and technical implications.

It is also evident that Knowledge sharing between National Statistical Offices (NSOs) is worthwhile, and incorporation of technological innovations has advantages irrespective of significant costs. Finally, being optimistic, positive minded and pro-active is key to implementation of a successful census.

As mobile and wireless technologies continually improve and become more affordable, reliable, powerful and user-friendly, Computer Assisted Personal Interviewing (CAPI) is expected to become the one of the effective approaches in field-based census data collection even in those countries where access to infrastructure for information and communication technologies is limited. African countries are at such a case. The introduction of electronic data capture into the business cycle of the census provides cost and time savings, and also allows users to take advantage of added features that can be programmed into mobile devices or linked to the data collection process. These features include, among others, integrated maps and Global Positioning System (GPS) and real time monitoring of fieldwork.

This paper shows how the Italian National Institute of Statistics (Istat) has supported the Ethiopia Central Statistics Agency (CSA) in designing and implementing such a monitoring system, which has been fully integrated with the census data collection process managed by CSPro, the public domain software package developed by the U.S. Census Bureau (USCB). The proposed architecture is generalised and provides a simple solution for monitoring electronic data collection operations, particularly in cases where the technical and financial resources to implement such a system from the ground up are lacking.

For the first time in Mexico, the 2020 Population and Housing Census data collection will be carried out through a CAPI (Computer-Assisted Personal Interviewing) scheme as the main enumeration method, but it will also include the CATI (Computer-Assisted Telephone Interviewing) and the CAWI (Computer-Assisted Web Interviewing) methods. These innovations, given the census structure size and the rapid changes in technologies, are a significant challenge for INEGI. Progress in census planning and field tests results will be presented, including the main challenges to be faced, the innovations considered for their implementation, as well as the successful experiences on the use of technologies for geo-referencing the information both in the data collection stage and for results dissemination.

In the United States of America, the Census Bureau (USCB) is implementing four major areas of innovation in preparation for the 2020 Census. The first area focuses on the maintenance of our frame of addresses using satellite imagery. The USCB will compare old and new satellite images to identify geographic areas that are stable and will not require in-person field staff updating. The second uses the Internet, including Identified (ID) and Non-Identified (Non-ID) processing, and cloud services to make it easier for people to self-respond. This Non-ID methodology has significant impacts on the way USCB staff execute our advertising campaign and promises to increase our self-response rates. The third uses existing administrative data to reduce cost. USCB will use existing data in many components of the 2020 Census including identification of vacant, non-existent, and occupied living quarters for non-respondents. The last area of innovation leverages technology to make field staff more efficient. Much of the field tasks will be automated including recruiting people, hiring staff, training staff, payroll and expense reporting, case assignments, case management, route optimization, quality control, and communications. The 2020 Census promises to be innovative, and improve over past censuses that relied heavily on paper questionnaires.

The Office for National Statistics (ONS) is exploring models for the future of population statistics. The UK does not have a population register or a set of coherent identifiers across administrative datasets held by government. The current population statistics system is underpinned by the decennial Census, which is expensive and is arguably becoming increasingly unwieldy as a source of data in a rapidly evolving society and with ever increasing demands for more timely, relevant statistics. The system is also highly reliant on a port-based survey to measure migrant flows to and from UK, with the result that the intercensal population size estimates tend to have an increasing element of bias. The ONS is therefore researching how the population statistics system can be transformed within that context. This paper outlines the current plan, focusing on some of the methodological challenges underpinning the transformation.

Stats NZ is beginning to see the benefits of a programme of research into greater use of administrative data for the population census and the wider population statistics system. The research effort has been driven by a need to modernise and reduce costs of the current full-enumeration census model, and the opportunities afforded by a rich and growing set of integrated administrative data. While New Zealand does not have an administrative population register, or any mandatory requirement to register an address with authorities, we have developed a statistical admin resident population based on a ‘signs of life’ approach. This work has proved to be a significant advantage given an unanticipated level of non-response in the most recent full field enumeration census in 2018. The final 2018 Census dataset consists of 89 percent census responses and 11 percent admin enumerations. This paper focusses on the statistical methods used to determine high-quality administrative records to include in the final 2018 Census dataset.

Response rates for official statistical collections are falling globally, placing increased emphasis on methods for handling missing data. At the same time, linked administrative data provides new opportunities for National Statistical Organisations. In 2016, the Australian Census of Population and Housing addressed unit non-response through nearest neighbour donor imputation. This method used dwelling and location characteristics to inform donor selection for each non-responding unit. This paper proposes a variation to the 2016 Census imputation method that incorporates administrative data to improve the selection of donors. Our method, Administrative Data informed Donor Imputation (ADDI), strengthens support for the assumptions of donor imputation, while limiting exposure to risks inherent in direct use of administrative data. We apply the ADDI method to the 2016 Census non-responding population, and demonstrate how the accuracy of imputation can be improved to produce higher quality statistics.

The census method in France has changed since 2004 from a traditional decennial census to a rolling census. For 15 years, the rolling census has been providing, each year, results at municipal and infra-municipal levels with a non-response rate remaining very low (under 4% of the households). During these years, many changes occurred: questionnaires, introduction of an internet response mode, changes in the protocol.

It turned out that the existence of an annual survey of 9 million people is a chance to implement large tests in real context before making a decision. Thus, Insee tested during four years the internet response mode, which guarenteed the success of the project by optimizing the response time and the robustness of the internet questionnaire. Today, four years after the generalization, 60% of people answer by internet. Moreover, the existence of an annual census survey is a chance to assess a collection mode effect: Insee measured that the introduction of internet did not affect most of the variables (except maybe the measure of employment). However, the last change of questionnaire in 2018 did have an effect on the counting of the population, which could be estimated thanks to the annual census surveys.

Many countries use administrative registers as an alternative to a field census enumeration, in order to save time, effort and money. Using administrative registers enables periodically updates of the population count and characteristics, without the long collection and processing period with the completion of a field enumeration census. Timely periodically updates are particularly useful for countries that undergo rapid population changes, as is the case in the United Arab Emirates (UAE) in general, and the Emirate of Abu Dhabi in particular.

The Statistics Centre of Abu Dhabi (the Centre) has thus decided to launch a project to develop a register-based census for the census 2020. This requires building a database, and maintaining its sustainability, and populating it with data on a continuous basis.

For this purpose, coordination has been established with the entities owning the administrative registers, with the support of senior management in those data providing agencies. These efforts have resulted in the signing of service level agreements, and data have begun to flow to the Centre by virtue of these agreements. The Centre has established specialised registers, such as population register, employees register, unemployed register, education register, and people with disabilities register.

The paper presents the different steps of preparations for the 2020 register-based census of Abu Dhabi, in general, and the process to conduct an administrative records based pilot census in particular.

The National Institute of Statistics (Istat) is moving from a decennial census to a permanent census, which will produce census data every year. The georeferencing of statistical information will allow Istat, through the integrated system of registers, to release statistical and geographical information with a strong territorial detail and will allow users to know more on the structure and composition of the territory of Italy. The new Statistical Base Register of territorial and geographical entities of Istat (RSBL) is a multidimensional register integrating several components: addresses, regular grids and micro zones, buildings and housing units, administrative zones, statistical and functional zones. In the panorama of statistical registers, it represents an innovative way of considering in an integrated way all the most relevant components of territorial data. In this paper, we present some of the main features of the register.

This paper aims to demonstrate the programme of censuses conducted in Kingdom of Saudi Arabia, which includes the Population and Housing Census and Establishment Census.

In order to improve and further develop efficiency and prepare the Kingdom for an efficient move to an administrative based approach in the 2030 Vision, the General Authority for Statistics will use a Combined Census approach between traditional and register-based census. This Combined approach will use administrative data supported by fieldwork. The new approach has required prior testing of administrative data such as the Saudi National Postal Address and data held by the National Information Centre. The paper will further discuss the progress in the preparation and implementation of this new approach. This will include a thorough review of how the National Address system is being used as the framework for identifying buildings and households, as well as reviewing the results of the pilot studies initiated. The paper will also discuss the implications for the fieldwork presented, which include Computer Assisted Interviewing and the publicity and awareness censuses campaigns. Finally, it will also discuss preparations for the release and publication of statistics in the 2020 round.

The next census in Estonia is planned to be register-based – all the necessary information would be derived from 25 registers. During the preparation period, the algorithms for creating census variables from register variables were prepared and tested and criteria for quality checks were developed. The first step for the state was to create a legal basis for using administrative data, with the aim to produce statistics. It was also necessary to link all administrative registers with each other by identifiers (ID codes of persons, enterprises and addresses) and form a uniform system. The existence of a rapid and secure data exchange and transportation environment was also required. Big efforts were made to achieve good mutual understanding and collaboration between Statistics Estonia and register holders.

This paper discusses some pre-conditions for the transition from traditional census to combined census and to register-based census. It also sheds light on the practical use of certain methodological experiences and quality aspects in the 2020 census round. The paper provides an overview of potential use of administrative registers in population and housing censuses and formulates basic recommendations for their use. It focuses on the case study of Poland, which carried out a population census using combined methods of linking administrative registers with ad hoc surveys.

With the goal of producing timelier, more useful, sustained and less costly annual population statistics, the Turkish Statistical Institute (TurkStat) plans to change its census methodology from combined to register-based. It also plans to produce census data on 1 km 2 grids, in 2021. Turkstat’s shift to a register-based census is also driven by Eurostat’s goal of producing census variables annually in the post-2021 period, unfavorable features of previous census methods (traditional or combined), and significant improvements in national administrative registers. TurkStat plans to establish an integrated system that includes related registers for Turkey’s 2021 Population and Housing Census (PHC). TurkStat’s Address Based Population Registration System is sufficient to meet almost all the country’s population statistical needs, except for labor force statistics and building and dwelling characteristics. Thus, TurkStat must carry out further research to set up a qualified database. With respect to the United Nations Economic Commission for Europe’s recommendations for the 2020 PHCs, as well as European Union (EU) regulations, TurkStat is likely to produce approximately 70% of the core census topics using the currently available registers. With the 2021 PHC approaching, TurkStat’s aim is to increase the quality and variety of register-based statistics.

Spain plans to carry out its first register-based census in the next 2021 round, becoming one of the biggest countries in the world using this approach.

Since 1996, when the Population Register was created, census methodologies have evolved enormously and Spain has vastly increased the use of administrative registers in official statistics. In the last 2011 census, it was too soon to completely rely on administrative data, so a combined census was performed.

Now, in 2018, 22 years after the creation of this register, INE faces this challenge with confidence. The availability of a variety of administrative registers and the fact that INE has access guaranteed by law to them, endorses this step forward.

There are still some difficulties to overcome and a great effort is being made today, collecting any suitable administrative source, processing and integrating data, developing new IT tools, and of course, evaluating the quality of the whole statistical product.

A complete population census test referenced to 2016 proved very satisfactory which enables for Spain planning to conduct a register-based census in 2021.

Small Island Developing States (SIDS) like those of the Caribbean Community are subject to a number of challenges that include susceptibility to external economic and financial shocks, vulnerability to natural disasters and to the effects of Climate Change that when combined with domestic policy choices lead to reduced funding for Development. This key constraint, affects the availability of resources for statistical development and specifically for the conduct of the decennial Population and Housing Census. In addition to the specific challenges that affect the entire census operations such as-political factors, accessibility to the households that are be enumerated and Crime, the efficiency of the census operation and the quality and timeliness of the results require intensive attention to enable a successful outcome.

US census data on Indigenous American Indians have progressed from no data at all to very poor and inconsistent data to smaller (but still significant) undercounts on Indigenous lands. The first US census, conducted in 1790, explicitly excluded American Indians from being counted, in accordance with the US Constitution. Much has changed. As the U.S. gears up for the 24 th decennial census in 2020, the very first Americans to be counted reside in the remote Indigenous village of Tooksook Bay, Alaska. This paper describes how the Indigenous American Indians and Alaska Natives came, first to be counted and secondly, to be counted accurately in national censuses. Joint efforts by American Indian Tribes, the National Congress of American Indians and the U.S. Census Bureau have evolved to obtain accurate and complete counts. The results of these efforts may be useful in other countries to improve participation in censuses of Indigenous people in other countries.

Population and housing censuses are large operations mobilizing the whole country. The National Statistical Office, usually in charge of the preparation and implementation of the census, does not have the permanent resources to conduct it on its own and needs support from the national government as well as from local authorities. There are three areas where the government plays a role in a census. It provides the legal framework to conduct the census; funding for the census; and logistical support. Local authorities provide also logistical support and, in some countries, a ‘democratic control’ of the fairness of the enumeration operations. However, the involvement of national and local authorities in the census operations is not without pitfalls. The independence of the census may be jeopardized, and the census perceived by the population as an administrative operation rather than a statistical one. This may entail lack of trust from the population on the use of the data and the confidentiality guaranteed to the information provided. Based on examples in the Western Balkans, Asia and Africa, the paper discusses the optimal balance between governance support and influence avoided in the population and housing census operations.

The timeliness dimension of data quality has not received sufficient scholarly attention even though the publication of official statistics of various countries are often delayed. This paper examines the growing delays in the release of census data in India amidst the technocratisation of policy-making, public professions of faith in evidence based policy-making and growing fascination with big data. We show that the growing delay in the release of census data of India is a fact, rather than a mere allegation, and contextualise the problem by comparing seven decennial censuses conducted between 1951 and 2011. We suggest that delays can be measured vis-à-vis usual, desirable, declared and feasible schedules of publication. Further, delays can be understood from the perspectives of the unwillingness of the government statistical system to face public scrutiny, decline in the quality of public goods, political interference and, in case of data on identity, communalisation of politics.

In many developing countries, Population and Housing Censuses (PHC) and Censuses of Agriculture (CAs) remain two key sources for basic and structural data on the characteristics of the population and the agriculture sector. However, these two operations are highly expensive and many countries find it difficult to conduct them in the internationally recommended time frame of ten years. This is particularly true for the CA, which is conducted with very irregular frequency (or not at all) in many developing countries.

In addition to funding issues, constraints include inappropriate institutional set-up often characterized by a lack of coordination and cooperation between the institutions concerned (particularly National Statistical Offices and Ministries of Agriculture), the non-integration or linkage of CAs with agricultural surveys, administrative sources and other major censuses and surveys in the national statistical system, particularly the PHC.

The adoption of the Sustainable Development Goals (SDGs) which have created increased data demands and emerging technological tools provide new opportunities that can help in better linking the CA with the PHC and a more integrated statistical system. This can result in a wider database for more in-depth analysis of interactions between agriculture and the socio-economic environment and contribute to monitoring the SDGs.

This paper will provide an overview of some of the advantages and issues of linking the two censuses in light of the new international guidelines on the PHC 2020 and the World Programme for Census of Agriculture (WCA) 2020, and present some country examples.