267. This chapter reviews the population bases adopted by countries in the 2020 census round, including in particular the definitions of “usually resident population” adopted in the countries and the practices regarding special population groups.

268. The definition of the population is not only central to the census operation, but it is at the very heart of the institutional and socio-economic setting of a country. Defining the rules according to which a person is part or not of the population of a country has indeed far reaching consequences, first-hand examples being the allocation of parliamentary seats or the attribution of funds depending on the population size. The definition of the population is therefore a fundamental component of the statistical information of a country ― if not its core ― and the population census is the primary framework in which such a choice is made.

269. This chapter recalls the main population issues dealt with in the CES Recommendations, defines the population concepts that countries followed in the 2020 round, and reports on the extent to which these complied with the CES Recommendations.

270. The definition of population is as complex as it is fundamental. While it may be safely assumed that there is a large awareness about the importance of the usual residence concept, it is also true that its implementation is not always straightforward and that full clarification was provided in the CES Recommendations.

271. The main definitions related to the population bases included in the CES Recommendations (paras. 388-393) are as follows:

(a) The population to be enumerated is the set of persons whom the country decides should be covered by the census, regardless of their subsequent exclusion from any specific population count, as defined below. The enumerated population base is composed of those persons who have actually been enumerated. This may or may not equate to the target population (the population to be enumerated), that is, the coverage of the census may represent either an under-count or over-count.

(b) The population base is the population used for the compilation of statistical aggregates in a particular tabulation. This may be a sub-set, or the whole, of the ”population to be enumerated”. A country may adopt more than one population base (for different statistical purposes), but one of these should always be the population base used for international comparisons purposes (more often the ”usually resident” population).

(c) A population count is the aggregate obtained by the simple addition of individual records from the enumerated population base. A population estimate is the aggregate obtained as outcome of a statistical method of estimation. Therefore, both the population count and the population estimate refer to a specific population base and are empirical measures.

(d) The place of usual residence is the geographic place where the enumerated person usually spends their daily rest, assessed over a defined period of time including the census reference time.

(e) The population base to be used for international comparisons purposes is the usually resident population. The usually resident population of a country is composed of those persons who have their place of usual residence in the country at the census reference time and have lived, or intend to live, there for a continuous period of time of at least 12 months. A “continuous period of time” means that absences (from the country of usual residence) whose durations are shorter than 12 months do not affect the country of usual residence. The same criteria apply for any relevant territorial division (being the place of usual residence) within the country.

272. Countries were asked in the survey to specify, for each of the population counts produced in the census (up to six counts), the label used (e.g. "usually resident population", "registered population") and give the corresponding definition.

273. Several countries provided label and definition for more than one population count used in their census, as follows:

(a) 8 countries reported two counts (Armenia, Belarus, Belgium, Czechia, Hungary, Poland, Russian Federation, United States);

(b) 4 countries reported three counts (France, Ireland, Slovakia, Tajikistan);

(c) 2 countries reported six counts, which was the maximum (Switzerland and United Kingdom).

274. The most common group of population counts is usually resident population or similar, with 27 countries mentioning it as the most important and by 4 more as an additional count. However, the respective underlying definitions diverge notably with some countries pointing to the 2020 CESR definition, others noting that they followed other (including EU regulated or national) definitions, or approximated the concept by the registered population.

275. The second most popular concept label adopted was registered population, used by eight countries as the most important count and by two countries as an additional count. The somewhat similar concept of legally resident population is used as most important count in two countries that use registers as part of a combined census: in Germany it is used according to the registration law and in Poland based on a national definition.

276. The permanent resident population is used as the most important count in 6 countries (Azerbaijan, Belarus, Kyrgyzstan, Russian Federation, Switzerland and Tajikistan). In Ireland the de facto is the most important count, but the usually resident count is also used. The temporarily present population was used as an additional count in Belarus, the Russian Federation and Tajikistan (where also the temporarily absent population was used).

277. Other population counts, deriving from national definitions or combinations of different criteria, were used by several countries. In France the municipal population² count is the most important, then the population counted apart (which includes certain people whose primary residence is in a different municipality, but who have maintained a residence in the municipality) and the total population resulting from the sum of the two counts.

278. Switzerland, in addition to the permanent resident population main count, five others were reported: non-permanent resident population; permanent foreign resident population; Swiss resident population; non-permanent foreign resident population; and population of Swiss citizens living abroad. Also the United Kingdom reported five counts in addition to the main usually resident count: all usual residents in households; all usual residents in communal establishments; all household reference persons; all households; and short-term residents.

Population count	Total (out of 45 countries)	Of which:
		as most important count	as additional count
Usually resident	31	27	4
Registered	10	8	2
Permanent resident	6	6	-
Legally resident	2	2	-
De facto / Present	1	1	-
Temporarily present	4	-	4
Temporarily absent	1	-	1
Other (national definition)	16	1	15

279. Given the importance of the concept of usual residence as recommended in the CES Recommendations, countries were asked whether they produced a population count based on the usually resident population and if the definition adopted complied with the recommended definition.

280. Almost three quarters of the responding countries used a definition of usual residence compliant with the CESR definition, 9 (20%) a definition that mostly complies with it, and 10 (23%) a definition based on legal or registered residence, which was compliant with the European Census Act (Table 49).

281. Of the 44 responding countries, 23 (52%) used a definition of usual residence that complies fully with the CESR definition, 9 (20%) a definition that mostly complies with it, and 8 (18%) a definition based on legal or registered residence, which was compliant with the European Census Act (Table 49). No countries used another non-compliant definition.

Question: “Did you produce a population count based on the usually resident population?”	Total (44 countries)		Type of census
			Field enumeration (15 countries)		Combined (15 countries)		Register-based (14 countries)
Yes, and the definition of usual residence complies fully with the CESR definition	23	52%	10	67%	8	53%	5	36%
Yes, and the definition of usual residence mostly complies with the CESR definition, as far as circumstances allow	9	20%	4	27%	3	20%	2	14%
Yes, and the definition of usual residence was based on legal or registered residence (compliant with European Census Act)	8	18%	1	7%	3	20%	4	29%
Yes, using another definition of usual residence that does NOT comply with the CESR	0	0%	0	0%	0	0%	0	0%
No	4	9%	0	0%	1	7%	3	21%

282. The countries that used a definition of the usually resident population that was entirely or partially different from that recommended by the CES for the total population census count were asked in the survey to indicate what were the main reasons for that. Multiple replies were possible.

The responses to this question, provided by 10 countries, are presented in Table 50: the reasons reported by most countries are the need to comply with a national definition (4 countries), to ensure better comparability with previous censuses (3 countries), or to use data from registers, whose definition was not fully compliant with the CES recommendations. Among these, Italy specified that the population count is register-based and the definition of the register is different from the CES definition, but the signs of life method is used to implement the definition, i.e. the population register data is corrected. Belgium noted that when register data is used it is not possible to measure the intention to stay in the country.

Reasons	Total (10 countries)	Type of census
		Field enumeration (3 countries)	Combined (4 countries)	Register-based (3 countries)
The definition used was imposed by the national definition	4	2	1	1
The definition used allowed better statistical comparability with previous censuses	3	2	1	0
The definition used was the only one available in the register used	3	0	1	2
The definition used was imposed by legislation	2	1	1	0
The definition used reflected requirements and/or needs of some of the main users	2	0	2	0
The definition used was more acceptable to respondents (for reasons of clarity, confidentiality or other)	1	1	0	0
The definition used was consistent with other statistical surveys	1	1	0	0

283. The CES Recommendations identified a number of population groups for which uncertainty may arise about their inclusion in the usually resident population of a country (para. 401), or about the definition of their place of usual residence within the country (paras. 407-408).

284. Countries were asked in the survey to indicate whether special criteria were adopted to define the place of usual residence of those population groups. When population counts were based on population registers, countries were asked to report criteria adopted by register-holding authorities. The results are presented in Table 51.

285. Special criteria for determining the place of usual residence were most frequently defined for the population groups “homeless or roofless persons, nomads” (22 countries), followed by “persons who have been an inmate of a hospital, hospice, nursing home, prison, etc.” (20), “persons who work away from their homes during the week and return at weekends” (16) and “students in tertiary education (ISCED levels 5-6) who are away from home while at college or university within the country” (16).

286. While almost all countries with a field enumeration or combined census defined special criteria for several of those population groups, only 4 of the 11 countries with a register-based census reported special criteria for some groups. This could be explained by the fact that some countries did not report criteria adopted by register-holding authorities. Norway reported that the National Population Registration manages a comprehensive body of rules, and therefore different kinds of "special criteria" abounds.

Population groups	Total (out of 39 countries)	Type of census
		Field enumeration (14 countries)	Combined (14 countries)	Register-based (11 countries)
Homeless or roofless persons, nomads	22	12	7	3
Persons who have been an inmate of a hospital or hospices, older persons in nursing homes, prisoners, juvenile detention centres, etc.	20	11	6	3
Persons who work away from their homes during the week and return at weekends	16	11	5	0
Students in tertiary education (ISCED levels 5-6) who are away from home while at college or university within the country	16	10	4	2
Persons in military service	15	10	3	2
Refugees and asylum seekers	15	9	2	4
Children who alternate between two households	14	10	3	1
Students in primary and secondary education (ISCED levels 1-4) who are away from home during the school term	13	8	4	1
Students in tertiary education (ISCED levels 5-6) studying abroad but returning home at weekends	13	9	3	1
Persons who regularly live in more than one residence during the year	13	9	4	0
Persons working for international organizations (not including diplomats or military forces)	11	8	3	0
Other	9	2	3	4

287. As mentioned above, the CES recommended whether those population groups were to be included in the usually resident population of the country, or excluded. In the survey, countries were asked if they included – or attempted to include, even partially – in the population those groups of person, or if they were intentionally not included.

288. The results are presented in Table 52, where the percentage of the countries that included the population groups in the usually resident population in the 2010 census round is also presented, allowing a comparison with the 2020 round.

289. The population groups “holders of temporary residence permits (and their families) staying in the country for more than 12 months” and “persons living in remote areas” were included in the usually resident population in all countries where these groups were defined in the census. Other groups were included in almost all countries: “homeless or roofless persons” and “civilian residents who cross a frontier daily to work in another country, but who return home every day or at weekends” (both groups 97%), “merchant seamen and fishermen resident in the country but at sea at the time of the census”, “tertiary students who study in another country, but who return home every day or at weekends”, and “persons who have applied for or been granted refugee status or similar types of international protection” (all these groups 94%).

Population group	Countries including the group	Percentage of countries*	Percentage of countries in the 2010 round
Holders of temporary residence permits (and their families) staying in the country for more than 12 months	36	100%	80%
Persons living in remote areas	32	100%	68%
Homeless or roofless persons	36	97%	94%
Civilian residents who cross a frontier daily to work in another country, but who return home every day or at weekends	34	97%	86%
Merchant seamen and fishermen resident in the country but at sea at the time of the census (including those who have no place of residence other than their quarters aboard ship)	29	97%	66%
Tertiary students who study in another country, but who return home every day or at weekends	32	94%	80%
Persons who have applied for or been granted refugee status or similar types of international protection	30	94%	64%
Foreign persons working for international organizations (not including diplomats or military forces) located in the country	32	91%	60%
National military, naval and diplomatic personnel and their families, located outside the country	29	81%	74%
Nomads or other travelling people	20	80%	48%
Asylum seekers	25	74%	58%
Illegal, irregular or undocumented migrants	10	33%	38%
Short-term international migrants (staying in the country less than 12 months)	10	30%	16%
Foreign military, naval and diplomatic personnel and their families temporarily located in the country	9	26%	14%

* Calculated over the number of countries that responded having included or excluded the population group (not considering countries where the group was not defined in the census).

290. The comparison with the 2010 round shows that for most of the population groups the percentage of countries including them in the count has increased significantly in the 2020 round. Particularly significant is the increased inclusion of “persons who have applied for or been granted refugee status or similar types of international protection” (from 64% in the 2010 round to 94% in the 2020 round), “nomads or other travelling people” (from 48% to 80%) and “asylum seekers” (from 58% to 74%).

291. On the contrary, the group “illegal, irregular or undocumented migrants” was included only in one third of the countries, compared to 38% in the 2010 round. This is explained by the increasing number of countries using data from registers that normally do not include this population group. Spain is the only country with a register-based census where this group was included, since all migrants – regardless of their legal status – can be registered in the municipal registers, that are also used for census purposes.³

292. The group “short-term international migrants (staying in the country less than 12 months)” was included by 30% of the countries (increased from 16% in the 2010 round), although the CES recommended not to include this group in the usually resident population (paras. 402 and 403).