Chapter 13 Geography
13.1 Introduction 
783.
783. One of the distinguishing features of censuses of population and housing is the extent to which a comprehensive specification of geographic characteristics can be undertaken, as well as the cross-tabulation of other characteristics, such as demographic, economic or ethnocultural, by geography.
784.
784. As described in Chapter 3, one of the important features of a census is that, since it covers the complete universe of interest, characteristics of the population can be given at the smallest geographical levels, without being subject to the errors that can arise when using statistical estimates based on survey samples.
785.
785. Once the population base has been determined it is then possible to examine how this population is located geographically. As described in Chapter 12, the “location” of each individual is generally taken to mean where they have their place of usual residence, i.e. where they most often spend their nightly rest. Chapter 1Part AChapter 8Chapter 8 explains in detail how such location can be specified using geospatial information.
13.2 Location of place of usual residence (core topic)
786.
787.
787. The location of the place of residence should be coded to the smallest possible civil division and, to the greatest extent possible, geo-referenced to the precise geographical coordinates73 of the address point. In the absence of such coordinates, a precise and complete postal address should be recorded for geocoding purposes. The purpose is to enable tabulations and spatial aggregations to be referenced to any small geographic or administrative subdivisions and, if possible, to grid squares (see section 13.6). The link between the census information and the location of the place of residence should form a permanent and integrated part of the census information at the unit record level.
13.3 Locality (derived core topic)
788.
788. For census purposes, a locality is defined as a distinct population cluster or settlement; that is, the area defined by population whose places of usual residence are located in neighbouring or contiguous buildings.
789.
789. Such buildings referred to in this definition may either:
(a) form a continuous builtup area with a clearly recognizable street formation; or
(b) though not part of such a builtup area, comprise a group of buildings to which a locally recognized place name is uniquely attached; or
(c) though not coming within either of the above two requirements, constitute a group of buildings, none of which is separated from its nearest neighbour by more than 200 metres.
790.
790. In applying this definition certain landuse categories should not be regarded as breaking the continuity of a builtup area (and accordingly should not be counted in applying the 200metre criterion above). Such categories include: industrial and commercial buildings and facilities, public parks, playgrounds and gardens, football fields and other sports facilities, bridged rivers, railway lines, canals, parking lots and other transport infrastructure, churchyards and cemeteries.
791.
791. This definition is intended to provide general guidance to countries in identifying localities and in determining their boundaries, and it may need to be adapted in accordance with national circumstances and practices. The population not living in clusters as defined above may be described as living in scattered or isolated buildings. Whatever definition of a locality is adopted for the census, it should be given in detail in the relevant census report and/or metadata.
792.
792. Localities, as defined above, should not be confused with the smallest civil divisions of a country. In some cases, the two may coincide. In others, however, even the smallest civil division may contain two or more localities as defined. On the other hand, some large cities or towns may comprise two or more civil divisions, which should then be considered only as sub-divisions of a single locality rather than as separate localities.
793.
793. A large locality of a country (that is to say, a city or a town) is often part of an urban agglomeration, which comprises the city or town proper and also the suburban fringe or thickly settled territory lying outside, but adjacent to, its boundaries. The urban agglomeration is, therefore, not identical with the locality but is an additional geographical unit, which may include more than one locality. In some cases, a single large urban agglomeration may comprise several cities or towns and their suburban fringes. The components of such large agglomerations should be specified in the census results.
794.
794. Countries are recommended to develop their census statistics for localities in accordance with national needs and capabilities. In doing so, they should try to approach as closely as possible the concept of the population cluster as defined above. Countries which tabulate statistics only for civil divisions should, as a minimum, endeavour to compile data on the total population of each part of a civil division containing a population cluster, or part of a population cluster, of at least 2,000 inhabitants and so provide a basis for making a more clearcut distinction between urban and rural areas and their populations (see paragraphs 798-805 below).
795.
795. It is recommended that the population be classified by size of locality according to the following sizeclasses:
(1.0)1,000,000 or more inhabitants
(2.0) 500,000 999,999
(3.0) 200,000 499,999
(4.0) 100,000 199,999
(5.0) 50,000 99,999
(6.0) 20,000 49,999
(7.0) 10,000 - 19,999
(8.0) 5,000 9,999
(9.0) 2,000 4,999
(10.0) 1,000 1,999
(11.0) 500 - 999
(12.0) 200 499
(13.0) Population living in localities with less than 200 inhabitants or in scattered/isolated buildings
(13.1) Population living in localities with 50-199 inhabitants
(13.2) Population living in localities with less than 50 inhabitants or in scattered/isolated buildings
796.
796. This classification could also be applied to analyse other relevant population bases such as the economically active population, households, families and dwellings (though it may be necessary to adopt different scales).
13.4 Location of living quarters (core topic)
797.
797. As defined in section 17.2.2, living quarters are those types of housings that are the usual residences of one or more persons. It is therefore possible to classify living quarters to the same detailed geography as “location of place of residence” (paragraph 786 above), and the definitions and classifications set out in paragraphs 788-796 and 798-818 apply equally. The extent to which they are used will vary according to each country’s statistical needs for information on localities and smallest relevant civil division.
13.5 Urban and rural areas (derived core topic)
798.
798. For purposes of international comparisons, countries should define urban areas as localities with a population of 2,000 or more, and rural areas as localities with a population of fewer than 2,000 and sparsely populated areas. Some countries might also wish to consider defining urban areas in other ways, for example in terms of:
(a) administrative boundaries;
(b) built-up areas; or
(c) the area for which services such as shops, educational facilities, recreational facilities, employment, are provided.
799.
799. The approach taken should be described clearly in the relevant census report and/or metadata.
800.
800. For national purposes, as well as for international comparability, the most appropriate unit of classification for distinguishing urban and rural areas is the locality as defined in paragraph 788 above. The decision of whether to use the locality or the smallest civil division as the unit of classification depends on the needs and circumstances in each country.
801.
801. Countries that use the smallest civil division as the unit should endeavour to obtain results that correspond as closely as possible to those obtained by countries that use the locality as the unit.
802.
802. The approach to be adopted to achieve this aim will depend mainly on the nature of the smallest civil divisions in the countries concerned. In some countries the smallest civil divisions (and the average number of inhabitants) are relatively small and generally do not contain more than one population cluster (or part of a larger population cluster). If it is not feasible for some of these countries to use the locality as the unit, they should make use of the concept of the multicommunal agglomeration (that is, treat as single units groups of two or more contiguous minor civil divisions that form part of the same population cluster). It is also suggested that minor civil divisions at the periphery of such an agglomeration be included in the agglomeration if a major part of their resident populations live in areas belonging to the continuous builtup area of the agglomeration, and that minor civil divisions containing one or more isolated localities be classified according to the number of inhabitants of the largest population cluster within the unit.
803.
803. The situation is different in the case of countries in which the smallest civil divisions (and the average number of inhabitants) are relatively large and often contain two or more population clusters of varying sizes. If it is not feasible for some of these countries to use the locality as the unit, they should endeavour to use units smaller than minor civil divisions for this purpose; for example, parishes, enumeration districts and grid squares. They should endeavour to use these smaller units as building blocks and to aggregate them to correspond as closely as possible with the boundaries of localities in the same way as described above in the case of multicommunal agglomerations. If it is not feasible for some countries to adopt this approach, they should endeavour to develop new approaches to the classification of entire minor civil divisions in ways which will yield results that are as comparable as possible with those obtained by using the locality as the unit.
804.
804. It is recommended to group localities or similar units into the following categories:
(1.0)1,000,000 or more inhabitants
(2.0)250,000 to 999,999
(3.0)100,000 to 249,999
(4.0)50,000 to 99,999
(5.0)10,000 to 49,999
(6.0)2,000 to 9,999
(7.0)Fewer than 2,000 inhabitants.
805.
805. Countries should develop typologies of localities or similar areas based on additional criteria that could be used to distinguish different types of areas within particular categories of the recommended classification. For example, some countries may wish to subdivide category (1.0) (and in some cases category (2.0) as well) to distinguish agricultural localities from other types of small localities. Some countries may wish to subdivide one or more of the intermediate categories to distinguish market towns, industrial centres, service centres, etc. Some countries may wish to subdivide the large urban agglomerations included in categories (4.0) to (7.0) to distinguish various types of central and suburban areas. Extensions of the classification in these and other ways would enhance its analytical usefulness.
13.6 Grid square characteristics
13.7 Population grid (derived non-core topic)
806.
806. For census purposes, a population grid is a geo-referencing framework for population, with fixed and unambiguously defined locations of equal-area grid cells. The grid adopted should be fully specified with reference to a specific grid standard and projection.
807.
807. Each grid cell contains the aggregated number of persons for whom the location of place of usual residence (as described in paragraph 786 above) is geo-referenced to a point located within the grid cell. In countries in which such aggregation is not possible, the population grid could be disaggregated from the smallest civil division using estimation methods based on auxiliary information such as land use maps, built-up area detection, or cadastral information.
808.
808. In addition to assigning usual resident individuals to a grid, the same grid can be used to assign people to their place of work or location of school, college or university. The same grid could also be used for assigning households, families and dwellings.
809.
809. The size of each grid cell should be 1km². Countries may also wish to create other grid cell sizes to meet national requirements.
810.
810. To permit cross-national interoperability, it is recommended that countries make use of a shared, internationally agreed grid (such as INSPIRE74 for Europe).
811.
811. Although the topic is included in these Recommendations as a non-core topic, countries are strongly urged to use grid square data. For those in continental Europe that do so, the grid system should be defined in line with the INSPIRE legal framework.75 In regions outside of continental Europe, countries may define their own grid based on a geodetic coordinate reference system compliant with the International Territorial Reference System (ITRS) and a Lambert Azimuthal Equal Area projection, following the same principles as laid down for the INSPIRE grid. In this case, an identifier for the coordinate reference system should be created and included in the metadata of the population grid.
812.
812. Census statistics have, in the past, usually been produced in relation to administrative areas or specific census output areas. Population grids are very useful complementary output systems offering several advantages. Grid cells are all of the same size, making them perfect for area- or distance-based comparisons (e.g. for comparing accessibility to services), and for the construction of customizable, interactive mapping tools for dissemination (see section 7.2). Grids are stable over time and hence independent of changes in civil divisions or borders. They permit the easy integration of data from other sources and on other topics (such as climate data). Grid cells are flexible as they can be assembled to form areas reflecting a specific purpose or study area. Grid systems can be constructed hierarchically in terms of cell size, thus matching study areas from the local to the global level.
813.
813. Grids are very powerful for spatial analysis in an international and cross-border context as they are not affected by variation in municipality size between countries. However, grid statistics may give rise to concerns over confidentiality or significance of the data in thinly populated areas, or when data come from a sample source. Therefore the topics for which grid statistics are to be produced should be selected carefully, and the choice should always represent a compromise between the demand for detailed and flexible statistics and data protection concerns (for more details on this, see section 7.1).
13.8 Degree of urbanization (derived non-core topic)
814.
814. Degree of urbanization76 classifies the areas in which persons have their place of usual residence as being densely populated areas, intermediate density areas, or thinly populated areas. The classification is based on a combination of geographical contiguity and minimum population thresholds applied to the 1 km² population grid cells (defined in paragraph 806 above). These grid cells all have the same shape and size, which avoids distortions caused by units varying in size.
815.
815. The degree of urbanization creates a classification of local administrative units/municipalities as follows:
(a)Densely populated areas have more than 50 per cent of their population living in high-density clusters77 (urban centres).
(b)Intermediate density areas have more than 50 per cent of their population living in urban clusters, but are not densely populated areas.
(c)Thinly populated area has more than 50 per cent of their population living in rural grid cells.
816.
816. Degree of urbanization may also be used to create an alternative classification of areas as being either “urban” or “rural”, with intermediate and densely populated areas being classified as “urban” while thinly populated areas are classified as “rural”.
817.
817. In the above classification of local administrative units/municipalities, the following definitions, population and density78 thresholds should be used:
(a)High-density clusters (urban centres) are clusters of contiguous grid cells of 1 km² with a density of at least 1,500 inhabitants per km² and a minimum population of 50,000. Contiguity for high-density clusters does not include the “diagonal” (that is, cells with only the corners touching), and gaps in the cluster are filled (that is, those cells surrounded by high-density cells).
(b)Urban clusters are clusters of contiguous grid cells of 1 km² with a density of at least 300 inhabitants per km² and a minimum population of 5,000. The same “contiguity” rule for high-density clusters, described at (a), also applies, but gaps in the cluster are not filled (that is, cells surrounded by urban cells).
(c)Rural grid cells are grid cells which do not belong to an urban cluster.
818.
818. As local administrative units/municipalities vary considerably in area, this methodology will lead to a closer match between high-density clusters and densely populated local administrative units/municipalities in countries with small local administrative units/municipalities than in those with large local administrative units/municipalities. To take this difference into account, the classification can be adjusted as following:
(a) A densely populated local administrative unit/municipality can be classified intermediate as long as 75 per cent of its high-density cluster population remains in densely populated local administrative units/municipalities.
(b)A thinly populated or intermediate density local administrative unit/municipality can be classified as densely populated if it belongs to a group of local administrative units/municipalities with a political function and if the majority of population of this group of local administrative units/municipalities lives in a high-density cluster.
13.9 Commuting characteristics
819.
819. The following topics examine issues associated with commuting from home to workplace, school, college or university. Accurate commuter flows are important for a whole raft of reasons including transport planning, housing development and economic development.
13.10 Location of place of work (core topic)
820.
820. The location of place of work is the precise location in which a currently employed person performs his/her job. The location should preferably be coded to the precise address and/or geographic coordinates79 or, if this is not possible, to the smallest possible civil division.80
821.
821. The main reason why place of work information is collected is to link it with place of usual residence in order to shed further light on commuter flows in addition to that provided by mode of transport to work, distance travelled and time taken. The precise address of place of work should be collected and the data coded to the smallest possible civil division in order to establish accurate commuter flows from the place of usual residence to the place of work. Information on persons who do not have a fixed place of work but who report to a fixed address at the beginning of their work period (for example bus drivers, airline crew, operators of street market stalls that are not removed at the end of the workday) should refer to that address. This group may also include individuals who travel to work, on a regular basis, across the border to a neighbouring country. However, it may not be possible to allocate the place of work of some persons, such as sailors, fishers and offshore workers. Such persons should be coded as having “no fixed place of work” (see classification in paragraph 823 below).
822.
822. In order to monitor commuter flows, a classification comparing location of place of work and location of place of residence is recommended. Commuting should be measured between place of origin and place of destination of the daily journey. It should be noted, however, that for some persons in employment, the place of origin may not necessarily be their place of usual residence if they have, of example, a separate week-day address from which they usually travel to work.
823.
823. The following classification is recommended:
(1.0)Same minor civil division as place of residence
(2.0)Other minor civil division within same major civil division as place of residence
(3.0)Other major civil division
(4.0)Abroad
(5.0)Other place of work
(5.1)Offshore installation
(5.2)No fixed place of work
824.
824. Minor civil division refers to the lowest level of geography for which commuter flows should be measured. Depending on national circumstances and user requirements this could be small areas (such as wards or communes) or areas at a higher level (such as municipalities or districts). Major civil division refers to a higher level of geography such as municipalities (if the minor division is small areas) or counties/departments (if the minor division is municipalities or districts).
825.
825. An alternative and more detailed two-digit classification may be considered by some countries that wish, for example, to identify, separately, persons working at home or longer-distance commuters:
(1.0)Same minor civil division as place of residence
(1.1) Working at home
(1.2)Working elsewhere within the same minor civil division
(2.0)Neighbouring minor civil division
(3.0)Other minor civil division
(3.1)In same major civil division
(3.2)In some other major civil division
(4.0)Abroad
(5.0)Other place of work
(5.1)Offshore installation
(5.2)No fixed place of work.
13.11 Location of school, college or university (non-core topic)
826.
826. By including this topic in their census, countries can extend the scope of their data on commuting patterns to cover pupils and students in addition to the coverage of the employed provided by place of work. In order to maintain comparability with the place of work topic, the location of school, college or university should be coded to the same level of geography.
13.12 Mode of transport to work (or to place of education) (non-core topic)
827.
827. Mode of transport to work relates to the journey made on most days. For people making several journeys or using more than one mode of transport, the mode of transport used for the greatest distance should be indicated. People not travelling to work should be classified as having no journey.
828.
828. The following classification is recommended81:
(1.0) Rail
(1.1) National/international rail network
(1.2) Metro/underground
(1.3) Tram/light railway
(2.0) Bus, minibus or coach
(3.0) Car or van
(3.1) Driver
(3.2) Passenger
(4.0) Other
(4.1) Motorcycle
(4.2) Pedal cycle
(4.3) Walk
(4.4) Boat or ferry
(4.5) Other
(5.0) No journey made
(5.1)Working at home
(5.2)Other reason
829.
829. Some countries may wish to amend or augment this classification to allow identification of such modes of transport such as e-bikes or e-scooters which are increasing rapidly in popularity, or consider further subdividing category (3.1) to distinguish person who drive alone from those who drive with other passengers, though this would require collecting additional information in the census.
830.
830. In order to provide more complete information on commuting patterns, some countries may wish to extend the topic of mode of transport to include daily journeys made by children and students to their respective place of education. As for the mode of travel to work, the mode of transport used for the greatest distance in the journey should be indicated. The classification set out in paragraph 828 should be adopted.
13.13 Distance travelled to work (or to place of education) and time taken (non-core topic)
831.
831. Countries may wish to collect information on the distance travelled to work (or place of education) on a daily basis and the time taken with a view to monitoring the extent to which persons are living at greater distances from their place or work (or education), and the impact which traffic congestion has on the travel time taken. When collecting this information, consideration should be given to the address from which the journey commenced.