Comparability — Argentina [top]

The sample for 1991 asked only about the presence of a flush toilet. The other samples indicate all types of toilets (flush or non-flush).

Comparability — Armenia [top]

In 2001 and 2011 the census offered no response category to indicate complete lack of a toilet or access to one. Presumably, such cases are included with the "non-flush" category. In both years the category "other" was recoded as "no toilet".

Comparability — Austria [top]

All samples indicate only the presence or absence of a flush toilet within the dwelling.

Comparability — Bangladesh [top]

The 1991 and 2001 samples do not distinguish flush and non-flush toilets: the indicate "sanitary" or not, which appears to include latrines. See the unharmonized source variables for details. The 2011 sample distinguishes toilets with and without a water seal, which is interpreted here to indicate a flush toilet.

Comparability — Belarus [top]

The question for Belarus indicates presence of a flush toilet in the household.

Comparability — Bolivia [top]

Only the 1992 sample distinguishes between flush and non-flush toilets. The 1976, 2001, and 2012 samples only indicate the presence of any kind of toilet facilities, and do not specify the type.

Comparability — Botswana [top]

In 1991, 2001, and 2011, the census question distinguished between private, communal, and neighbor facilities. These distinctions are not included in TOILET but they are available in the unharmonized source variables for each census sample. In 2011, dry compost (enviro-loo) toilets were categorized with "Non-flush, other and unspecified".

Comparability — Burkina Faso [top]

The samples are comparable.

Comparability — Cambodia [top]

The 1998 sample indicates the presence or absence of any toilet facilities; the 2004, 2008, and 2013 samples specify toilet type. In all years, the question asks about toilet facilities "within the premises," which may explain the relatively low proportion of households with a toilet.

The 2019 sample distinguishes whether the toilet is shared with other households in the unharmonized source variable. The ''latrine overhanging field or water (drop in the field, pond, lake, river, sea)'' category is classified as ''no toilet.''

Comparability — Cameroon [top]

The "other" category in 1976 and 1987 is interpreted as having no toilet. The source variables for 1987 and 2005 differentiate by exclusive use (private and shared) and location (indoors and outdoors).

Comparability — Canada [top]

Presence of a flush toilet includes shared use with another household in the dwelling.

Comparability — Chile [top]

The variable is comparable across samples. The 1960 sample is a sample of individuals, not households. Only heads of households have most household-level information.

Comparability — Colombia [top]

The Colombian censuses are comparable with each other. The question refers to the dwelling, and the toilet may have been used by one or more households.

Comparability — Costa Rica [top]

The variable is consistent across samples. The 1960 sample is a sample of individuals, not households. Only heads of households have most household-level information.

Comparability — Cuba [top]

The census does not indicate the type of toilet.

Comparability — Côte d'Ivoire [top]

In 1988, the non-flushing toilet is considered as "non-flush, latrine".

Comparability — Dominican Republic [top]

Question universes vary across years, but the response categories are comparable across samples.

Comparability — Ecuador [top]

The variable is consistent across samples. The 1962 sample is a sample of individuals, not households. Only heads of households have most household-level information.

Comparability — Egypt [top]

The data are comparable across samples and indicate only whether the household had a toilet of some kind, including shared use with another household.

Comparability — El Salvador [top]

The samples are mostly comparable and classify the type of toilet used regardless of ownership (private or shared). Information on exclusive use of facilities is retained in the unharmonized source variables.

Comparability — Ethiopia [top]

There are significant universe differences across samples: in 1984 the question was asked only of urban households; in 2007 only long form recipients were asked the toilet question. In 1994, separate forms were used for urban and rural areas. The rural form groups pit latrines and flush toilets, while the urban form distinguishes between the two.

Comparability — Fiji [top]

Water sealed toilets are included with flush toilets. They are separately identified in the unharmonized source variables.

Comparability — Germany [top]

The 1971 and 1981 East Germany samples are comparable, whereas the 1987 West Germany sample only indicates presence of a flush toilet. For the 1971 and 1981 samples, the no toilet category consists of households that use a toilet outside of the building, regardless of type. Those that use a toilet outside the dwelling yet still inside the building are regarded as having a toilet.

Comparability — Ghana [top]

This variable is comparable across samples. Use of a public toilet or toilet in another house is coded as no toilet; bucket/pan is coded as "other non-flush."

Comparability — Greece [top]

All samples indicate access to a flush or non-flush toilet, either located inside or outside the dwelling.

Comparability — Guatemala [top]

In all years, washable toilet (excusado lavable) is categorized as "flush toilet." The 1981, 1994 and 2002 source variables provide additional detail on connection to sewer or septic tank and exclusivity of use.

Comparability — Haiti [top]

The universe of households asked the question is restricted in 1982 to those in the communes of Port-au-Prince, Pétion-Ville and Croix des Bouquets.

Comparability — Honduras [top]

The 1961 census is a sample of individuals, not households. Information is only available for households consisting of a head.

The latrine connected to a drain is classified as "non-flush, latrine" in 1961, 1988, and 2001.

Comparability — Hungary [top]

Prior to 2001, the question for Hungary indicates the presence of a flush toilet within the building. It could be shared with other households. In 2011, the census question asks for the number of flush toilets.

Comparability — Indonesia [top]

The 1971 and 2010 data are derived from separate source variables for access and type of facilities. For 1990-2010, toilets "with septic tank" are interpreted as flush toilets, and those "without septic tanks" are classified as "non-flush."

The 1980 and 1985 samples have variables on toilet facilities, but they could not be integrated into this variable, because they do not distinguish between households with shared toilet access and those without any toilet facilities.

See the unharmonized source variables for each sample for category detail that was lost during harmonization.

Comparability — Iran [top]

The data indicate the presence of a water closet within the housing unit.

Comparability — Iraq [top]

The data indicate the presence of a water closet in the dwelling unit with no indication as to type or whether access was exclusive.

Comparability — Ireland [top]

The variable is consistent across samples. The data indicate the presence of the type of toilet regardless of whether it is inside the building and regardless of shared use.

Comparability — Israel [top]

The sample indicates the presence or absence of any toilet facilities, not the specific kind of toilet.

Comparability — Italy [top]

The Italian census indicates presence or absence of a flush toilet -- one "connected to a plumbing and a drainage system."

Comparability — Jamaica [top]

The variable is consistent across samples, except for in the 1991 and 2001 samples when the question was asked only of private households, and in the 2001 sample when there is a lack of an "other" category.

Comparability — Kenya [top]

The Kenyan censuses do not mention of whether the toilet is physically in the household. The residual non-flush toilet facilities include buck latrines and cesspools.

Comparability — Laos [top]

In 1995, the ''other'' category is coded as "non-flush, other and unspecified."

In 2015, ''hanging toilet or hanging latrine'' which is built on the sea, a pond, river, and or surface water where feces falls directly into the water is classified as ''no toilet'', and ''composting toilet'' and ''bucket'' categories are coded as "non-flush, other and unspecified."

Comparability — Lesotho [top]

In 1996, use of a public toilet is coded as no toilet.

Comparability — Malawi [top]

The 1987 and 1998 samples indicate the type of toilet only for the first dwelling of the household. The 1987 sample includes shared use. The 1998 and 2008 samples do not offer instructions on classification of shared use, but ask for which type of toilet the household uses. The data for 1998 and 2008 include a category for "VIP latrine," which is distinguished from a traditional pit in the unharmonized data.

Comparability — Malaysia [top]

A "pour flush" toilet is considered a flush toilet in both samples.

Comparability — Mali [top]

In all samples the "other" category is interpreted as having no toilet, which seems implied by the instructions.

Comparability — Mexico [top]

The 1960 and 1970 samples are comparable and indicate if the toilet has a sewer connection (flush). The 1960 sample is a sample of individuals, not households. Only heads of households have most household-level information.

The 1990, 2000 and 2010 samples indicate the presence of a toilet, and if so, whether it has a direct discharging (flush) mechanism; the 1990 sample only counts toilets that are for exclusive use whereas the 2000 and 2010 censuses count private and shared use toilets. The 2015 and 2020 samples distinguishes flush toilets, non-flush toilets, and latrines.

The 1995 sample differs in that it specifies the type of toilet (latrine, black hole, etc.) and the 2005 sample classifies toilets connected to the sewer system or septic tanks (flush), those without water discharge (non-flush), and those without toilets.

Comparability — Mongolia [top]

Use of a toilet outside the dwelling is coded as no toilet.

Comparability — Morocco [top]

The samples indicate whether the household uses a toilet or water closet regardless of ownership (private or shared). The 2004 sample offers more detail on the type of toilet facility.

Comparability — Mozambique [top]

The source variable for 2007 has additional information about the type of latrine (traditional, improved, non-improved).

Comparability — Myanmar [top]

In the 2014 sample, buckets/surface latrines are included in the "non-flush, other and unspecified" category.

Comparability — Nepal [top]

The samples are comparable. Ordinary toilet is classified as "Non-flush, other and unspecified."

Comparability — Nicaragua [top]

The samples are generally comparable. Any responses that indicate a system that discharges water are classified as flush toilets, regardless of where the waste goes.

Comparability — Palestine [top]

The variable is consistent across samples except for slight differences in the question universes.

Comparability — Panama [top]

All samples indicate the presence of toilet facilities in the dwelling, whether they were shared with other households or exclusive.

Comparability — Paraguay [top]

The variable is comparable across samples.

Comparability — Peru [top]

In both samples, responses that indicate the toilet was connected to a public system or (in 2007) a septic tank are interpreted as having a flush toilet.

Comparability — Philippines [top]

The data are fully comparable across samples. Flush toilets include all "water-sealed" devices.

Comparability — Poland [top]

The 1978, 1988, and 2002 samples ask about flush toilets in the dwelling. The 1978 sample includes both connections to public or local systems. In 1988, flush toilets include also those households which have access within the building.

Comparability — Portugal [top]

The variable is consistent across samples.

Comparability — Puerto Rico [top]

The question is nearly comparable across samples. In 1970, 1980, and 1990, respondents reported whether they had a flush toilet or privy inside the house. In 2010, information about whether or not the house had a flush toilet was recorded.

Comparability — Romania [top]

The samples indicate the presence or absence of a flush toilet.

Comparability — Rwanda [top]

The samples are comparable. The use of a neighbor's facilities in 1991 is classified as no toilet.

Comparability — Senegal [top]

In the 1988-2013 samples, responses that indicate the toilet was connected to a public system or a septic tank are interpreted as having a flush toilet.

Comparability — Sierra Leone [top]

The 2015 sample distinguishes between private and communal facilities. These distinctions are not maintained in TOILET, but they are available in the unharmonized source variable.

Comparability — South Africa [top]

Chemical toilets are included with flush toilets in the 1996 sample, but are in the residual non-flush category in the 2001, 2007, 2011, and 2016 samples. In all the South Africa samples, bucket latrines are included in the "other non-flush" category. There are some universe differences between the samples.

Comparability — South Sudan [top]

The 2008 sample identifies the type of toilet used in the household.

Comparability — Spain [top]

The samples of 1991 and 2011 indicate the presence or absence of toilet facilities, not whether there was a flush toilet. In 1991 the question was asked to all dwellings except lodgings, while in 2011 the universe was composed by all households.

Comparability — Suriname [top]

In 2012, an open pit or hole in the ground is included in the "non-flush, other and unspecified" category.

Comparability — Tanzania [top]

The 2012 sample includes an option for "composting or ecoson latrine" which is categorized as "non-flush latrine".

Comparability — Thailand [top]

The data are consistent across all samples, though there are underlying differences in category availability in the source variables. For the 1970 and 1980 samples, the "other or none" category is coded as having no toilet.

Comparability — Trinidad and Tobago [top]

The 1980, 1990, 2000, and 2011 samples identify different types of toilet facilities.

Comparability — United Kingdom [top]

The data indicate only the presence of a flush toilet, whether shared or used exclusively by the household.

Comparability — United States [top]

The 2010 sample identifies whether there is a flush toilet in the household either for exclusive or shared use.

Comparability — Uruguay [top]

Other than slight variations in questions universes, the samples are comparable and explicitly indicate whether the household uses a toilet with or without a flushing system.

Comparability — Venezuela [top]

The variable is consistent across samples.

Comparability — Vietnam [top]

Many households in the Vietnam 1989 sample lack dwelling information, including all households in some provinces. The 1999 and 2009 samples differ slightly in the wording of responses and question universes but are otherwise comparable. The 2009 and 2019 samples are comparable.

Comparability — Zambia [top]

In 1990, the "other" category is coded as having no toilet. The later samples had a separate response category for "no toilet," and "other" is coded as "non-flush, other and unspecified."

Comparability — Zimbabwe [top]

Communal toilet is classified as "non-flush, unspecified".