A population census (also census , census or macro census ) is a legally mandated collection of statistical population data , whereby citizens are obliged to provide information using the conventional method of census by questionnaire. In the register census model, data in the population registers are used without asking citizens .
A population census (also census, census or macro census) is a legally mandated collection of statistical population data, whereby citizens are obliged to provide information using the conventional method of census by questionnaire . In the register census model, data in the population registers are used without asking citizens. The term census is misleading in that it counts more than the people, i.e. the number of inhabitants. Rather, censuses oblige people to provide a variety of personal data. In some countries, the conventional method of census has been replaced by the register census model , whereby the data in various population registers are used. No questionnaires are given to citizens.
Another census method is the rolling census method. A part of the population is regularly surveyed, the scope of the surveys mostly depending on the size of the community. There are mixed forms in which conventional population censuses (i.e. surveys) are combined with the evaluation of registers. Register-based counts are also possible, which are supplemented with surveys in random samples .
In modern industrialized countries , as a rule, censuses are carried out every ten years, usually at the beginning of a new decade, as was recommended at the Statistical Congress in Saint Petersburg in 1872.
A census is carried out in order to obtain as accurate information as possible about various statistical parameters that are to be used as a basis for political and administrative action. The planning of housing programs, measures to improve the public infrastructure , assessment bases for the financing of public budgets or tax estimates of the numbers can be carried out more precisely or in a more targeted manner through a census.
The International Statistical Congress in Saint Petersburg in 1872 recommended what personal data should be asked for in every census
- First and last names ,
- Gender ,
- Age ,
- Relationship with the head of the family or household,
- Marital status ,
- Occupation or employment,
- Creed ,
- language spoken in ordinary communication ,
- Knowledge of reading and writing ,
- Origin, place of birth and nationality ,
- Place of residence and type of stay on the day of enumeration (whether permanently or temporarily present or absent),
- Blindness , deaf dumbness , nonsense and cretinism , mental illness .
The aforementioned personal data generally form the basis of every survey even today.
A methodological peculiarity of classic population censuses is that, unlike normal empirical studies, not a representative sample , but all households and almost the entire population are surveyed directly ( total census , see also population ). The data obtained are updated through ongoing representative surveys (the microcensus ) that are carried out annually and extrapolated to the base data of the last census. The sample size for these surveys is usually set by law, around one percent of the population. Since the resulting error rate increases over the years, the census must be repeated at longer intervals and the basic data must be updated. In addition, larger interim counts are carried out over a larger, representative selection after about five years , limited to rapidly changing data sets.
The first censuses
Indirect determination of population figures for tax purposes can be seen as early as 2700 BC. In Egypt . Due to pottery shards from the city of Mari, it can also be said for the time around 1700 BC. A local census in Mesopotamia for military purposes. (Local) censuses are documented in Egypt around 1100 BC. From earlier epochs, censuses are also known in China (2 AD), in Persia and Greece , also in Egypt under Amasis (569 BC: decree on the recording of incomes). One often limited oneself to the registration of men capable of weapons.
In the Roman Empire there was since the 6th century BC Every five years censuses and surveys of the income of Roman citizens. The censor , an ancient Roman official, was responsible for the census and the tax estimates . It set the amount of tax each citizen had to pay and was responsible to the Senate. The censors were very influential and highly regarded.
The census in the Bible
In the Old Testament (Tanach)
In the book of Numbers ( 4th book of Moses ) of the Bible , two censuses carried out at God's instruction are described in detail. They took place shortly after the beginning and shortly before the end of the forty-year desert wandering of the Israelites under Moses. The Latin name of the book "Numbers" comes from these censuses. Only men capable of fighting were counted, with the exception of the Levites intended for priestly service , which also counts male children and old men. Both counts each give a total of over 600,000 men. In view of the lack of archaeological finds, historians consider it very unlikely that so many Israelites actually wandered through the Sinai Desert for such a long time.
The Old Testament of the Bible also mentions a census of the Israelites by King David ( Second Book of Samuel , First Book of the Chronicles ). This census was therefore not ordered by God and drew a divine punishment - an epidemic ravaged the country and claimed numerous victims. The temple in Jerusalem was built according to the report later in the place where David could turn away again by a guilt offering the wrath of God upon the non-arranged Census.
In the New Testament
The New Testament of the Bible mentions a census ordered by the Roman emperor Augustus . According to the Gospel of Luke , the emperor had ordered that everyone should register in the tax lists in their place of origin. For this reason, Mary and Joseph traveled to Bethlehem , where Jesus Christ was born:
“But it happened in those days that Emperor Augustus issued the order to enter the whole world in tax lists. This record was the first; at that time Quirinius was governor of Syria. Everyone went to their city to be registered. So Joseph also went up from the city of Nazareth in Galilee to Judea, to the city of David, which is called Bethlehem; for he was of the house and tribe of David. He wanted to be registered with Maria, his fiancée, who was expecting a child. "
Here Luke mentions two clearly differentiated processes in the Roman administration in two immediately following sentences, namely
- Imperial census ( Lustrum ), the estimate of Roman citizens in the entire Imperium Romanum
- Provincial census, the estimate for the inhabitants of a province who did not have Roman citizenship ( Civitas Romana ).
Both the Luke and Matthew Gospels (2.1–19 EU ) report that the birth of Jesus took place when Herod the Great was still alive. But since this 4 v. Chr. Died, the mentioned count should have taken place before that (cf. Lk 2.1 EU with 1.5 EU ). Accordingly, the census mentioned by Luke cannot be the provincial census - also mentioned by Flavius Josephus - which was carried out in 6/7 AD under Publius Sulpicius Quirinius in the province of Syria . Shortly before, Herod Archelaus , a son of Herod the Great, had been deposed and banished, and the previous ethnarchy of Judea had been incorporated into the province of Syria.
The tax returns of the residents of the provinces were probably made according to a uniform tax form , which was the same in all imperial provinces. A form from AD 127 was found in a cave west of the Dead Sea .
The census was rarely included as a theme in Christian art and in nativity scenes. A well-known example is the painting “ The Census at Bethlehem ” (1566) by Pieter Bruegel the Elder, who relocated the scene to the Flanders of his time.
Counts from the Middle Ages to the beginning of the modern era
In the 7th century, the first census was carried out in the Arab city of Medina under the Prophet Mohammed (623). In particular, the aim was to define the groups listed in the Constitution of Medina and to clearly record the tax revenues of the new state under the Prophet.
Another 7th century census took place in China. The survival of small farmers and thus social peace ensured tax laws (619) and the associated agricultural ordinances (624). In it, the farmers were given evenly distributed plots for life. The award was based on exact population censuses taking age into account and a cadastral system for land evaluation / distribution. The system was undermined by manipulated figures and the expansion of private and ecclesiastical property as early as the late 7th century. It was also too susceptible to outdated numbers, migration and changes in cultivation.
After the Norman conquest of England in 1066, King William the Conqueror initiated a census. The dates of the census carried out in 1086 were published in the Domesday Book (Anglo-Saxon: domesdaeg = "Judgment Day").
In the Middle Ages there were few censuses in Europe; Most of the fireplaces were registered, but the data collected was often imprecise, so that information on the population is usually only extrapolations (e.g. one fireplace = approx. 10 residents) and not traditional figures. In some parts of Europe, such as France , Spain and Italy , the sources go back to the 14th and 15th centuries. B. a census in the county of Foix is known.
Reliable data are mostly the land records of the monasteries and monasteries, some of which date to the High Middle Ages, but often only allow indirect conclusions about the population. Church records of the parishes were important in recording the population. So the pastors had to keep books about the “souls” (Latin liber status animarum ). In some countries the interest in these records was particularly great. Sweden has excellent parish records for the 17th century. There are said to have been censuses in Venice since the 13th century. The oldest surviving document of such a system dates back to 1509 and population surveys have been carried out regularly in Venice since the 16th century, the documents of which have largely been preserved in the archives.
The nobility always resisted listing their serfs, as a purely private matter, especially as long as they were exempt from taxation. As late as 1753, the British Parliament rejected a census because it "would expose England's enemies' weaknesses". A member of parliament stressed that he was strange "that there are human beings who are so cheeky and shameless" to propose such a thing.
In Central Europe , a few cities, for example Nuremberg in 1449 , first attempted censuses. In the later nationwide censuses, the number of fireplaces was determined, whereby the population could be determined using an estimated average "number of people per fire". In Zurich , Antistes Breitinger carried out a census for the first time in 1634, which would then be repeated every six years.
The first country population census in the modern sense took place in Lithuania around 1528 , initially as a pure recording of the rural population or the able-bodied men, then as a tax cadastre. In French Louisiana and Canada (Nouvelle-France, Acadie) a census was carried out from 1665.
In Sweden - Finland the longest closed time series in the world began in 1686 when the parish registers were compiled. This is where the earliest establishment of a specially competent authority can be found ( Tabellverket as a full census from 1749, Table Commissions 1756). Comprehensive censuses are also known from Friesland 1689, Iceland 1703, Overijssel 1748, Holland 1747.
The first census in Austria was carried out after the administrative reforms under Maria Theresa in 1754, the first in the Hungarian part of the monarchy 1767–1775. In Spain (Castile) the first censuses beyond land use took place in 1768–1769, and in 1584 in Liechtenstein .
After 1800 more or less regular censuses took place in almost all European countries. In 1801 Great Britain , France (first local censuses by Vauban from 1676) or Portugal started their first complete data collection, Prussia in 1816, Württemberg in 1818 (largely all-German: Zollverein from 1834). In the later 19th century the first censuses took place in India and some South American countries, in most of the third world countries the first censuses were mostly not held until the 20th century (for example the first time in Afghanistan in 1979, Ethiopia in 1984). Due to various domestic, political turmoil, there were also long interruptions in the census in some countries; in Argentina there was no census between 1914 and 1947, and in Uruguay not even from 1908 to 1963.
In the United States , censuses are required every ten years by the constitution, which came into effect in 1789. The first census took place in 1790, and every decade since then.
In Switzerland , censuses have been carried out from 1850 (since the state was founded), usually up to every ten years. In Austria, too, the (today's) Statistics Austria has carried out a census every ten years since 1869.
The Statistical Congress in Saint Petersburg in 1872 recommended the ten-year cycle that is still in use today, with years ending with zero being counted. This recommendation was adopted by the League of Nations and later by the United Nations . However, counts can also be made in the previous or following year. Since the majority of countries in the world adhere to this recommendation, official population figures are available worldwide. This material forms, for example, the basis for population projections for these countries. In special cases, however, there are reasons to mistrust the population figures.
The coverage of the world population by censuses was around 17 percent in the middle of the 19th century and reached a first high of 78 percent in the middle of the 20th century. Since no census was carried out between 1954 and 1982 in China , the most populous country in the world, the degree of coverage fell again. At the end of the 20th century, a population census in the modern sense was held in almost all countries in the world, the large differences in accuracy and large temporal fluctuations (survey times and / or intervals) in many countries make it difficult to compare the surveys directly.
The knowledge based on international experience and the technical development led for the first time in 1890 in the censuses in Austria and the USA to the use of punch card technology according to Herman Hollerith in data processing with tabulating machines .
Censuses in the present
At the present time, the use of information technology for the storage, compression and evaluation of data is standard, although in most countries the written questionnaire is still used for direct collection. Citizens were able to fill out the census form via the Internet - for the first time in Europe - at the census in Switzerland in 2000. Each household received a user name and a password with the questionnaire, with which the participants could dial into the Internet at a certain address. So far, the Internet has been used for censuses in Singapore and the USA - but only for random samples and as a test.
In countries with a long tradition of register-based statistical surveys, such as Denmark, Finland, Norway and Sweden, the traditional method of census was replaced by the model of register census . Denmark was the first country in the world to switch to the new method in 1981. In the northern European countries, local population registers were set up as early as after the Second World War . These are already networked nationwide and equipped with personal identifiers. There are also central resident registers. For the 2011 census , the government in Germany also switched to the register-based model, while Austria implemented a full register census without any questioning. Switzerland continues to conduct an annual structural survey with at least 200,000 people. Countries without a population register will not be able to do without the traditional method of counting via questionnaires in the future either.
Population recording problems
For reasons of state , parts of the population are intentionally not fully recorded or are increased in number by the population statistics. In colonies, for example, the colonial powers were interested in keeping the indigenous population as small as possible. Ethnic rivalries make it seem right for some groups to downsize other groups and to enlarge one's own group. Such examples are significant when parliamentary seats are determined according to the size of the population or the proportion of civil servants.
Furthermore, the thought of higher development aid from outside can lead to an increase in the population. If more people live in an area, then it is more likely to be eligible for funding, or the respective gross domestic product is distributed among more people, which leads to a reduction in the statistical per capita income and thus makes economic aid appear more urgent. However, this assumption corresponds to an antiquated term of "development aid" of the 1960s, especially since today the World Bank or the European Development Fund strictly control the statistics carried out with their funds.
For political reasons, people are being outsourced or incorporated into some regions. At the time of apartheid, the Republic of South Africa incorporated tens of thousands of black people into the so-called homelands in order to increase the white population in the home country South Africa.
Wars and natural disasters
In some countries there was and is still no population recording or population extrapolation. The last census in Angola took place in 1970. The civil war between 1975 and 2002 made a census of the country's population impossible (see History of Angola ). In Afghanistan , the first and last census took place in 1979. Earlier and later data on the population are always based on estimates. The civil war in Afghanistan, which escalated in 1979, has since prevented further censuses as it culminated in a series of interrelated armed conflicts shaped by the Soviet invasion and US military intervention with its allies.
Wars prevent censuses and continuous population surveys. In Cambodia the died during the dictatorship Khmer Rouge from about eight million people, three million in 1974, according to various sources to 1979 well. However, until the 1998 census there was no reliable information about the country's population. During the 1994 genocide in Rwanda , around 10 to 25 percent of the eight million inhabitants in the 1991 census died, a quarter fled to neighboring countries and a further quarter of the population within the country.
Due to natural disasters (drought, floods, storms, earthquakes, volcanic eruptions), censuses have not been possible for years in the affected regions of Africa , Asia and Latin America . Refugee flows in and out of these areas make it difficult to record the population.
Insufficient statistical infrastructure
Groups in underdeveloped areas sometimes deliberately avoid being recorded because they assume - rightly or wrongly - that a census merely serves to collect taxes or to call up military service. Other reasons are rather unconscious: due to a lack of communication, an unfavorable geographical location. There are only estimates of the number of Brazilian jungle inhabitants over 50,000 to 150,000 people, which cannot be recorded by censuses . In fact, the partial non-recording ensures the existence of some indigenous peoples such as the Tagaeri and Taromenani in Ecuador that are threatened by the destruction of the rainforest .
Other problems are that there is no exact and statistically usable knowledge about certain information about the person, for example if the exact age information about the person is not available or it remains unclear whether a person should be recorded in certain regions, such as during stays abroad.
There are still problems with the population forecast in developing countries . If one only considers the births and deaths, the under-reporting of deaths is pronounced. In Africa and Latin America in particular, it has been found that numerous deaths in children and newborns among certain population groups have not been recorded. For these reasons, the difference between births and deaths is actually probably smaller, and with it of course the population growth, so that the annual growth rates of some countries are a little too high. These problems can be avoided by using censuses as it helps identify the unregistered deaths.
Even in countries with excellent population statistics (as in most highly developed industrialized countries) there are possibilities of uncertainty in the population survey. However, these are generally of a different order of magnitude than those in developing countries. In countries with many years of statistical experience, the data on the population censuses are considered exact, whereby as a rule not only a census takes place, but also data sets from other areas of social life are recorded and evaluated. The accusation of attempting to network data in order to create a transparent person is therefore objectively justified. This increases the tendency to refuse to output data or, out of fear of persecution, deliberately incorrect data is supplied where possible.
In the USA there is still no reporting obligation comparable to the reporting law in Germany , which results in a constantly estimated error in the recording of five to six million people. Due to the mandatory social security number on the one hand and the registration of voters before elections, there are indicators for determining a certain statistical population of the population, but no reliable data. Despite the above-mentioned circumstances, US sociology, political science and economics come to globally recognized results without a census taking place there as a total.
Development in individual countries
- Statistical service with a list that can be sorted chronologically
For the individual development in the German-speaking countries see:
- Census in Germany and list of censuses in Germany
- Working group for population registers in Northern Germany
- Census in Austria and list of censuses in Austria
- Census in Switzerland
- For the first EU-wide uniform census, see 2011 census .
- There are further country-specific articles on the censuses in the category: census
- Spyros Missiakoulis: Cecrops, King of Athens: the First (?) Recorded Population Census in History . In: International Statistical Review . tape 78 , no. 3 , December 28, 2010, p. 413-418 , doi : 10.1111 / j.1751-5823.2010.00124.x .
- Mario Martini : The 2011 census as a problem of intermunicipal equal treatment. Duncker & Humblot, Berlin 2011, ISBN 978-3-428-13590-5 .
- Frank Unruh: "That all the world would be valued." - Census in the Roman Empire . Theiss-Verlag, Stuttgart 2001, ISBN 3-8062-1639-8 .
- Rainer Wehrhahn, Verena Sandner Le Gall: Population geography. Wissenschaftliche Buchgesellschaft, Darmstadt 2011, ISBN 978-3-534-15628-3 , pp. 17-22.
- Worldwide data from censuses of State (English)
- Wilhelm Kubitschek : Census . In: Paulys Realencyclopadie der classischen Antiquity Science (RE). Volume III, 2, Stuttgart 1899, Sp. 1914-1924.
- Census dates for all countries. United Nations Statistics Division, accessed February 19, 2013 .
- Statistical offices of the federal and state governments: Statistical monthly magazine Baden-Württemberg 6/2006: Population censuses abroad
- Ian Shaw: The Oxford History of Ancient Egypt . Oxford University Press, 2004, ISBN 978-0-19-280458-7 , pp. 4-5 .
- Daniel E. Fleming: Democracy's ancient ancestors: Mari and early collective governance . Cambridge University Press, January 26, 2004.
- Jaroslav Černý: Consanguineous Marriages in Pharaonic Egypt . In: Journal of Egyptian Archeology . tape 40 , December 1954, p. 28-29 .
- John D. Durand: The Population Statistics of China, AD 2-1953 . In: Population Studies . tape 13 , no. 3 , March 1960, p. 209-256 , JSTOR : 2172247 .
- Isaiah McBurney: The Student's Handbook Of Ancient History: From The Earliest Records To The Fall Of The Western Empire . Richard Griffin and Company, 1856, ISBN 1-104-40124-X , pp. 34 (Reprint by Kessinger Publishing).
- Frank Tenney: Roman Census Statistics from 508 to 225 BC In: The American Journal of Philology . tape 51 , no. 4 , 1930, p. 313-324 .
- Walter Scheidel : Rome and China: comparative perspectives on ancient world empires . Oxford University Press, 2009, pp. 28 .
- This form was found in 1961; see Frank Unruh: "That all the world would be valued." - Census in the Roman Empire . Theis-Verlag, Stuttgart 2001, ISBN 3-8062-1639-8 .
- Population census / census (AT) . In: Bibelwissenschaft.de
- Data transmitted in historiographies and other historical documents are not necessarily reliable. Earlier “fiscal surveys counted fuoche (hearth or fireplace) and left out tax-exempt households, such as those of the clergy. Surveys for the bread allocation counted 'mouths' and ignored infants not yet weaned ... Church surveys counted 'souls' ”(Peter Burke: Städtische Kultur in Italien between High Renaissance and Baroque . Frankfurt 1996, p. 43).
- See section "II. Population" in Landwehr pp. 193–325.
- Zemininku , 1528 (peasant census), mentioned in the sources, first surviving census 1861; The main Historical Moments of Lithuanian Statistics ( Memento from April 19, 2012 in the Internet Archive ) , stat.gov.lt “About Us” History
- Statistics of Canada (Ed.): Censuses of Canada, 1665 to 1871. Volume IV (online article Introduction to Censuses of Canada, 1665 to 1871 , statcan.gc.ca); first full census 1841.
- Statistics Sweden's History ( Memento of September 18, 2012 in the Internet Archive ) , scb.se ›About us› Main Activity ›Statistics Sweden's History
- History of statistics in Finland ( Memento of November 15, 2012 in the Internet Archive ) , stat.fi
- All of the Netherlands 1795/96; → nl: Volkstelling #Nederland
- Manntal , cf. en: Icelandic census of 1703
- → Census in Austria
- closed land registers from the Middle Ages, Castile 1768–1769; Census Floridablanca 1785-1787; in Denmark - Norway 1769 (first authority 1797), 1797–1819 Dansk-Norsk Tabel-Kontor , 1833–1848 Tabelkommission ; according to Danmarks Statistics historie , dst.dk → Om Danmarks Statistics
- Liechtenstein Statistics Office , www.as.llv.li; the Principality did not institutionalize official statistics until 1950
- The History of the Office for National Statistics ( Memento of August 17, 2012 in the Internet Archive ) , ons.gov.uk → About ONS
- → census in Germany ; List of censuses in Germany
- Census Office publications with the results of the 1890 census ( Memento of April 18, 2012 in the Internet Archive )
- Statistical Monthly Bulletin Baden-Württemberg: Population Censuses Abroad , 6/2006.
- Structure. Federal Statistical Office , accessed on June 13, 2019 .
- Indigenous Peoples Today . OroVerde, Tropical Forest Foundation