Comparability — Bangladesh [top]

In the 2011 sample, a person is classified as disabled if he/she responds as having any difficulty in seeing.

Comparability — Benin [top]

Both the 2002 and 2013 questionnaires contained separate questions on whether the person was visually impaired or totally blind. These variables are combined for the purpose of DISBLND.

Comparability — Botswana [top]

The 1991, 2001, and 2011 samples asked about a defect of seeing or blindness. A person that cannot count the fingers of a hand from a distance of 3 meters, even when wearing prescribed glasses, would be considered as having a seeing defect.

The unharmonized source variables in 1991 and 2001 allow identifying persons with these disabilities in only one or both eyes. DISBLND considers all of these cases as being blind or having limited vision.

Comparability — Brazil [top]

In 1991, only total permanent blindness qualifies as a "Yes" for DISBLND. In 1991, persons with more than one disability condition are coded "Unknown" for DISBLND, because the exact disabilities in those cases are not known. See DISABLED to identify these cases.

In 2000, any level of permanent vision impairment that is not correctable with glasses is coded "Yes". Note that this usage is different from the definition used for 2000 in DISABLED, which requires "significant" permanent difficulty seeing.

In 2010, the question asks for the self-reported level of permanent difficulty (the level is preserved in the unharmonized data) in seeing. A person is classified as disabled if he/she responds as having any level of permanent difficulty in seeing.

The 2000 and 2010 instructions for the question are the same, but there has been a significant increase in people reporting this type of disability.

Comparability — Burkina Faso [top]

A person is classified as disabled if he/she responds as being blind.

The 1985 sample only provides the type of disability for residents of the household. The 1996 and 2006 samples provide the type of disability for all persons. The samples do not distinguish between temporary and permanent disabilities.

Comparability — Cambodia [top]

The question in 2008 and 2013 defines a visual disability as the inability to see at all (blind), blurred vision even with glasses, or without proper vision in one eye; this variable is irrespective of whether the person was born with the disability or developed it since birth. In 2019, the data identifies persons with difficulty seeing, even if wearing glasses.

Comparability — Cameroon [top]

In the 2005 sample, a person is classified as disabled if he/she responds as being blind.

Comparability — Chile [top]

The 1992 question refers to total blindness; an affirmative response in 2002 also included having limited vision.

Comparability — Colombia [top]

The 1993 questionnaire simply said "blind", but the instructions suggest that a "limitation" of eyesight would qualify if it affected typical activities. The 2005 sample specified a limitation that could not be corrected with glasses.

Comparability — Costa Rica [top]

The 2000 census indicates "partial or total blindness" including any substantial deficiency that could not be corrected by lenses. In 2011 the affirmative response to the question indicates "difficulty with seeing".

Comparability — Dominican Republic [top]

The 1981 data indicate whether the respondent was blind. The 2002 sample has two source variables that distinguish in blindness in one eye or both eyes. Respondents who were blind in one or both eyes are included in the "Yes" category in DISBLND. The 2010 data indicate if the person had a permanent difficulty seeing.

Comparability — Ecuador [top]

In 2001, the affirmative response to the question indicates the permanent complete inability to see or the ability to see only shadows. Persons using lenses because of myopia or similar diseases are excluded from this category.

In 2010, persons reported having a permanent visual disability that has lasted for more than one year. If a person indicates that has to use glasses because he/she no longer sees well, this condition is not registered as a disability.

Comparability — Egypt [top]

Both censuses instructions included the loss of one or both eyes, with no mention of other sight limitations.

Comparability — El Salvador [top]

The universe changes from all persons in the 1992 sample to persons in private households in the 2007 sample.

The 1992 sample refers to persons with total blindness, while the 2007 sample identifies permanent limitations in seeing, even when wearing glasses.

Comparability — Ethiopia [top]

The 1984 data indicate whether the respondent was fully blind or blind in one eye. Similarly, the 1994 sample indicates whether the respondent was fully blind or partially blind, including blind in one eye. The 2007 data indicates if the person was blind or had difficulty seeing. Some blind persons in 1984-1994 are included in the "unknown" category because they had multiple disabilities that could not be disaggregated.

Comparability — France [top]

The "Yes" response for the French sample is the combination of three separate responses: "blind", "almost complete blindness", and "one-eye or unilateral blindness".

Comparability — Ghana [top]

The data indicate whether the respondent had a serious vision disability that limited his/her full participation in life activities such as mobility, work, social life, etc.

Comparability — Guatemala [top]

The 1981 census instructions do not discuss degree of impediment. The 1994 census specifies that partial inability to see qualifies as disabled. 1981 has a specific source variable for blindness, whereas 1994 combines all disability responses into one variable, leaving no option for multiple disabilities.

Comparability — Guinea [top]

In 1996 and 2014 the data report complete blindness. In 1996, the variable reports responses of blind as well as blind-and-deaf, but other multiple-disability responses that might have included blindness are not identified.

Comparability — Haiti [top]

The 2003 sample, a person is classified as disabled if he/she responds as being blind.

Comparability — Honduras [top]

In the 2013 sample, a person is considered to have limited vision if they cannot see even though they wear glasses.

Comparability — Indonesia [top]

There are slight universe changes from 1980 to 2010.

The 1980 sample refers to persons that cannot see at all (total blindness), while the 2010 sample identifies total difficulties in seeing, even when wearing glasses. The 2005 sample only indicates the use of glasses but not whether the person has limitations in seeing even with them and thus it was not integrated for the DISBLND variable.

Comparability — Iran [top]

The variable indicates whether a person is blind, defined as having lost sight in both eyes or having only the ability to identify light while unable to recognize movement more than three feet away.

Comparability — Iraq [top]

The census instructions included the loss of one or both eyes, with no mention of other sight limitations.

Comparability — Ireland [top]

In 2011, the data indicate whether the respondent had long-lasting blindness or serious vision impairment.

Comparability — Israel [top]

In 2008, the census question asks for difficulty in seeing, rather than complete blindness; the response "does not have difficulty or has a mild difficulty" is coded as "no" in DISBLND.

Comparability — Jamaica [top]

In both samples, the variable indicates a disability in sight only, defined as a visual restriction or lack of visual ability to perform activities in the "manner considered normal for a human being". Those with multiple disabilities, due to the inability to determine the type of disability, are grouped in the unknown category. See DISABLED to identify these cases.

There are slight variations on the question universes but the samples are otherwise comparable.

Comparability — Kenya [top]

In 2009, a person is classified as disabled if he/she responds as having any difficulty in seeing. A person is considered to have an eyesight or vision disability if he/she does not have normal vision even when wearing eyeglasses or contact lenses. The respondents were to report up to three disabilities, listing most disabling one first.

In 2019, a person is classified as disabled if they have at least some difficulty seeing, even when wearing glasses. The response refers to vision with glasses that the respondent has and uses, not what their vision would be if provided glasses, or better glasses.

Comparability — Lesotho [top]

The census question referred to total or partial blindness.

Comparability — Liberia [top]

The "Yes" category includes persons who indicated that they had a sight difficulty or loss of sight. For more detail see the unharmonized source variable. Persons with multiple unspecified disabilities are coded as "Unknown".

Comparability — Malawi [top]

In 2008, the question asks if the respondent has any problems or difficulty in seeing which "inhibit the ability to work or participate in normal activities". In 2018, the question asks if the person has difficulty seeing. A second question, which is not used here, asks if the person has difficulty seeing even if wearing glasses.

Comparability — Malaysia [top]

The census instructions for both samples included limited sight as a disability. Persons could report up to three separate disabilities in 2000. They did not have to choose the most limiting condition.

Comparability — Mali [top]

The 1987 and 1998 source variables distinguish blindness resulting from onchocerciasis (river blindness) from other sources. The 1998 sample differentiates "blindness" from "difficulty seeing", a distinction not made in the 1987 sample. It is not clear if temporary or correctable decline in vision is interpreted as blindness or difficulty seeing. The source variable for each sample contains the full detail on nature of blindness.

The 2009 sample identifies persons with vision disabilities. The instructions do not provide further details. In 2009, persons with multiple unspecified disabilities are coded as "Unknown".

Comparability — Mauritius [top]

The 1990 and 2000 harmonized variables are constructed from the first, second, and third reported disabilities.

The 2011 sample reports different levels of difficulty in seeing, where persons with any level of difficulty seeing are classified as "yes". The difficulty levels are preserved in the corresponding unharmonized source variable.

Comparability — Mexico [top]

The 2000 and 2010 samples include persons with blindness affecting either or both eyes and those with visual difficulties, which are defined as the ability to distinguish only large shapes or shadows in the former sample and weak vision or advanced visual problems in the latter. Both the 2000 and 2010 samples refer to a disability that limits the person's ability to carry out daily life activities. The 2020 sample identifies people with difficulty seeing, even when wearing glasses.

Comparability — Mongolia [top]

In 2010, visual impaired refers to vision lost or impaired by "visual organ injuries, defects, and diseases". The response of "other disabilities" is classified as unknown, as that may include persons with multiple disabilities. In 2020, it refers to persons who have difficulty seeing, even when wearing glasses. Persons with multiple, unspecified disabilities are considered unknown.

Comparability — Morocco [top]

The 2004 sample identifies persons unable to practice daily activities in a normal way because of a total or partial visual disability (blindness).

The 2014 sample indicates whether the person has any difficulty in seeing, where persons with any level of difficulty seeing are classified as "yes". The difficulty levels are preserved in the corresponding unharmonized source variable.

Comparability — Mozambique [top]

The 2007 questionnaires only indicate "blind" without any further description. The 2017 sample identifies persons who responded yes to "Blind" or "Difficulty seeing, even when wearing glasses".

Comparability — Myanmar [top]

The 2014 sample indicates whether the person has any difficulty in seeing, where persons with any level of difficulty seeing are classified as "yes". The difficulty levels are preserved in the corresponding unharmonized source variable.

Comparability — Nepal [top]

In the 2001 sample, blindness includes inability to clearly see surroundings, blurred vision, and total loss of vision. The 2011 instructions are more detailed but are similar in defining the disability to include blindness and "low vision."

Persons with more than one disability condition are coded "Unknown" for DISBLND, because the exact disabilities in those cases are not known. See DISABLED to identify these cases.

Comparability — Pakistan [top]

In the 1973 sample, blindness is defined as the person's inability to "count the fingers of the enumerator before him/her or without spectacles at the distance of one foot". Respondents were not asked if their inability to see objects up close was a permanent condition.

DISBLND is not available in the 1981 and 1998 samples.

Comparability — Panama [top]

The census instructions for all years include uncorrectable "visual weakness" as a disability. The universe for the 1980 sample is limited to persons age 40 or younger.

In 2010, the affirmative response to the question indicates the complete inability to see or any visual impairment that cannot be normalized using lenses, treatments and other optical aids. This includes people who are blind from birth, or who suffer from diseases such as glaucoma, toxoplasmosis, and infections caused by various accidents (hit in the eye, various injuries, among others). In previous samples, the question only considered blindness or very limited vision as a physical impairment.

Comparability — Paraguay [top]

The 1962-1992 censuses simply report whether the person was "blind." The 2007 data indicates if the person was blind, including blind in one eye, or had difficulty seeing even with glasses.

Comparability — Philippines [top]

For the 1990, 1995, and 2000 samples, the question was directed to non-visitors while the in the 2010 sample, the question was directed to persons 5 or more years old.

There is are major comparability issues between the samples. An affirmative response in 1990 indicates complete blindness in one or both eyes. In the 1995 and 2000 samples any level of limitation (total, partial or low vision) is considered a disability. The 2010 sample broadly indicates functional difficulty in seeing, even if wearing eyeglasses.

Comparability — Portugal [top]

Visual disability was defined as complete or significant sight loss not correctable with glasses.

There was a single question on type of disability. If a person had more than one disability, the one causing the greatest degree of incapacity was chosen.

Comparability — Puerto Rico [top]

In 2015 and 2020, the variable indicates whether the person has any difficulty in seeing.

Comparability — Rwanda [top]

There is a minor universe difference between samples.

For the 2002 census, there was a single question on type of disability. If a person had more than one disability, the most severe one was to be reported. There were no specific instructions regarding the category "blind". This is not the case for the 2012 census, when respondents could list up to six types of disabilities.

Comparability — Saint Lucia [top]

The 1991 sample asked persons to report permanent disability in sight. The criteria used to interpret partial loss of sight as disability are not clear.

The question was not asked in the 1980 census.

Comparability — Senegal [top]

The 1988 sample identifies blind people as "visually impaired". The 2002 sample defines blindness as a disability that limits daily activities; in this sample, the loss of one eye is not interpreted as a disability.

It is not clear if visual impairment that is not limiting daily functioning or loss of one eye is interpreted as blindness in the 1988 sample. Permanent total loss of vision is implied in both the 1988 and 2002 samples.

The 2013 census question reports different levels of difficulty in seeing. Persons with any level of difficulty are classified as "yes" in 2013. These details are preserved in the unharmonized source variable.

Comparability — Sierra Leone [top]

In the 2004 sample, the variable indicates if the person's most serious disability is either blindness or sight difficulty. Therefore, a person may go unrecorded yet still have sight difficulty because it is not his/her most "serious" disability (subjective).

The 2015 sample records if the person's main or second type of disability is either blindness or partially sighted.

Comparability — South Africa [top]

The 1996-2007 samples asked roughly the same question, seeking to identify a "serious" sight disability that could not be corrected. The 2011 and 2016 questionnaires asked about degree of difficulty seeing, with DISBLND coding as a disability any amount of difficulty. The level of difficulty reported is preserved in the corresponding unharmonized source variable. The frequencies of blind or vision-impaired are very different across years.

The 1996-2007 samples have information for persons of all ages, while in 2011-2016 only for persons age 5+. The 2011 sample asked this question of all persons, but the census documentation recommended including only persons age 5+ in the analysis of disability. Thus, DISBLND has responses restricted to persons age 5+ in the variable universe. The original responses for persons under this age are available in the corresponding unharmonized source variable.

Comparability — South Sudan [top]

The variable indicates whether the person has any difficulty in seeing. Blindness and difficulty seeing are separable using the unharmonized source variables.

Comparability — Sudan [top]

The variable indicates whether the person has any difficulty in seeing. Blindness and difficulty seeing are separable using the unharmonized source variables.

Comparability — Suriname [top]

The 2012 sample reports different levels of difficulty in seeing, where persons with any level of difficulty seeing are classified as "yes". The difficulty levels are preserved in the corresponding unharmonized source variable.

Comparability — Tanzania [top]

The 2002 sample identifies blindness within a question about whether the person has any disability. The blindness recorded is severe enough to constitute a disability, but it is otherwise unclear whether permanent or total loss of vision is required.

Persons with more than one disability condition are coded "Unknown" for DISBLND, because the exact disabilities in those cases are not known. See DISABLED to identify these cases.

The 2012 census question asked about any difficulty seeing, even if wearing glasses. Persons with any level of difficulty were considered to have a vision disability.

DISBLND is not available in the 1988 sample.

Comparability — Thailand [top]

Blindness is defined as complete loss of sight. It is not clear if the data include people with multiple disabilities in DISBLND.

Comparability — Togo [top]

Blindness is not defined. The data are derived from separate variables recording up to two disabilities.

Comparability — Trinidad and Tobago [top]

The 2000 and 2011 questionnaires only indicate whether the person had visual impairment even with glasses.

Comparability — Turkey [top]

In the 1985 sample, persons with blindness in either or both eyes are coded as "Yes". In the 2000 sample, visually impaired refers to persons who cannot see at all or have limited sight in either one or both eyes. Persons with more than one disability in the 2000 sample, due to the inability to determine the type of disability, are grouped in the unknown category. Both samples refer to a disability that limits daily life activities or work.

Comparability — Uganda [top]

The 2002 sample stipulated that a disability was a condition expected to last 6 months or more. The threshold for disability in both samples is the ability to perform normal activities. The 2014 sample asks for difficulty in seeing even with the use of glasses. It reports different levels of difficulty, where persons with any level of difficulty seeing are classified as "yes"; these details are preserved in the unharmonized source variable.

Comparability — United States [top]

In 2010, 2015, and 2020, the variable indicates whether the person has any difficulty in seeing.

Comparability — Uruguay [top]

The 2006 sample identifies persons with permanent sight limitations on either or both eyes, even when wearing glasses, which affect the ability to perform daily life activities. Data are only available for persons interviewed in the third trimester of the household survey.

The 2011 data indicate if the person had any permanent difficulty seeing. The level of difficulty is preserved in the source variable unharmonized data.

Comparability — Venezuela [top]

The question for both samples asked if the person was totally blind.

Comparability — Vietnam [top]

The question asks for the self-reported level of difficulty (the level is preserved in the unharmonized data) in seeing, even with glasses. A person is classified as disabled if he/she responds as having any level of difficulty in seeing.

Comparability — Zambia [top]

The 1990 data indicate if the respondent was totally blind and does not include those who have lost sight in only one eye. In the 2000 sample there are two separate source variables on vision disabilities. One records if the person had complete vision loss and one records if the person had partial vision loss. Respondents who indicated complete or partial vision loss are included in the "Yes" category in DISBLND. See the unharmonized source variables for more detail. Persons who were blind or partially sighted are included in the "Yes" category in the 2010 sample. Persons with multiple unspecified disabilities are coded as "Unknown".

Comparability — Zimbabwe [top]

The 2012 questionnaire contained separate questions on whether the person had difficulty seeing or was totally blind. These variables were combined in the harmonized version.