Disability Statistics – Estimates Database Methods

Appendix 1: Disability Breakdown/Disaggregation and Groups

Appendix 1 describes the methods used in the Disability Statistics – Estimates (DS-E) database to disaggregate estimates based the disability status of adults ages 15 and older (DDI 2024). It includes and expands the description of methods in Carpenter et al (2024).  DS-E makes it possible to compare indicators across groups by disability status within and across countries. Disaggregating an indicator (e.g. ever attended school) by disability status aims to establish the size of the gap that may be associated with disability, i.e. the disability gap, or inequalities associated with disability.

Disability is measured through questions on functional difficulties in household surveys and population and housing censuses. All the datasets have questions on functional difficulties that

1. meet the United Nations (2017) Principles and Recommendations for Population and Housing Censuses with questions in at least four core domains (seeing, hearing, walking, cognition) and
2. have graded answer scales that capture the severity of functional disability.

Some datasets have the internationally tested Washington Group Short Set (WG-SS) of disability questions covering six domains (seeing, hearing, walking, cognition, self-care, and communication) (Altman 2016). The WG-SS is included in Appendix 1 Table 1 and more information is available at . Other datasets have other functional difficulty questions. These are similar to the WG-SS but with differences in the wording of the answer scale or the questions, and some do not have questions for the self-care domain and/or the communication domain.

Disability is diverse, both in terms of the type and severity of disability. Severity requires the establishment of a threshold using the answer scale of functional difficulties. Disability classification is challenging (Fujiura and Rutkowski-Kmitta 2001) and variation in the threshold of functional difficulty under consideration may lead to varying results regarding group sizes and inequalities (Hanass-Hancock et al 2023). Therefore, the DS-E Database uses four ways to disaggregate adults based on answers to functional difficulty questions:

• Disability versus No disability;
• Disability by type;
• Severe versus Moderate versus No Disability;
• Severe versus Moderate or No disability.

These four methods are described in Appendix 1 Table 2. The first breakdown ((i) Disability versus No disability) compares persons with any level of difficulty to persons with no difficulty in all domains.  It is the most inclusive in that persons with any level of difficulty are considered disabled.

The second breakdown ((ii) Disability by type) compares persons with any level of difficulty in each functional domain (seeing, hearing, walking, cognition, self-care, and communication) to persons with no difficulty in all domains.

The third breakdown ((iii) Severe versus Moderate versus No Disability) is a three-way disaggregation comparing persons with a lot of difficulty or unable to do, persons with some difficulty, and persons with no difficulty. This method makes the comparison of prevalence and disaggregated indicators possible for two levels of severity. In particular, it has the advantage of highlighting the situation of persons with moderate disability, who have been found in recent studies to be worse off than persons with no difficulty for various indicators (Hanass-Hancock et al 2023).

The fourth breakdown ((iv) Severe versus Moderate or No disability) follows the initial recommendation of the Washington Group (Washington Group on Disability Statistics 2020). However, by grouping persons with moderate disability with persons with no disability, persons with moderate disability are not included as part of the disabled population. It mixes the experience of people with moderate and no disabilities and captures only the inequalities that persons with severe disability face. Additionally, as noted above, when people with moderate disabilities are worse off than persons with no difficulty, combining them in this way artificially reduces the size of the disability gap for persons with a lot of difficulty compared with persons with no or some difficulty.

The DS-E database includes results using all four disaggregation methods. The three-way disaggregation via degree ((iii) Severe versus Moderate versus No Disability) provides a very informative approach to identifying potential disadvantages that may vary with the degree of functional difficulty. When sample sizes of persons with moderate disability and severe disability taken separately are too small (e.g., under 50 persons) for further disaggregation or analysis by gender, ethnicity, geography and other circumstances, disaggregating based on the first method (i) Disability versus No disability) makes further disaggregation and intersectional analyses possible.

Besides, for each dataset, we set 50 observations as the minimum required to produce estimates for subgroups following common practice. Due to this constraint, for a given data set, disaggregation may be possible for some indicators but not others, especially when some indicators are constructed for particular subgroups such as youth or workers.

References

Altman, B. M. (Ed.). International measurement of disability: Purpose, method and application, the work of the Washington group. Social indicators research series 61. Switzerland: Springer 2016. Accessed on 24th July 2024 at:

Carpenter, B., Kamalakannan, S., Patchaiappan, K., Theiss, K., Yap, J., Hanass-Hancock, J., Murthy, GVS, Pinilla-Roncancio, M., Rivas Velarde, M., Teodoro, D., and Mitra, S.. The Disability Statistics – Estimates Database (DS-E Database). An innovative database of internationally comparable statistics on disability inequalities. International Journal of Population Data Science. 2024.

DDI (2024). Disability Statistics – Estimates Database (DS-E Database). Disability Data Initiative collective. Fordham University: New York, USA. 2024.

Fujiura, G.T. and V. Rutkowski-Kmitta. Counting Disability. In Albrecht, G.L. and K.D. Seelman and M. Bury, Handbook of Disability Studies, Thousand Oaks, CA: Sage. 2001; pp.69-96. Accessed on 24 July 2024 at:

Hanass-Hancock, J., Kamalakannan, S., Murthy, G.V.S., Palmer, M.,  Pinilla-Roncancio, M., Rivas Velarde, M., Tetali, S., Mitra, S.  What cut-off(s) to use with the Washington Group short set of questions?, Disability and Health Journal 2023; Volume 16, Issue 4, 101499.

Washington Group on Disability Statistics. An introduction to the Washington Group on Disability Statistics question set. 2020. Retrieved from Washington:

Go to top

Appendix 2: Population Groups

There may be patterns of intersectional disadvantage that affect subgroups of persons with disabilities and their households, such as women or rural residents.

For each dataset under consideration, DS-E includes disaggregated estimates at the individual level based on disability as well as sex (female, male), rural/urban residence or age group (15 to 29, 30 to 44, 45 to 64, 65 and older).

At the household level, DS-E has disaggregated estimates based on disability as well as rural/urban residence.

Go to top

Appendix 3: Geography

Estimates are representative and available at the national level and at the first subnational level as per ISO (International Organisation for Standardisation) 3166-2 for subnational divisions. One exception is the Mongolia 2020 Population and Housing Census dataset, which is representative at the national level and not at the subnational level.

For some datasets, results are available at an alternative subnational level that is different from ISO 3166-2 and is relevant to the country’s policy or historical context. This is the case for Guatemala Census 2018, Haiti DHS 2016-2017, Maldives DHS 2009, Mauritania DHS 2019-2020, Mexico Census 2020, Nigeria WHICH DATASET?, Pakistan DHS 2017-2018, Palestine HIES 2009, Uganda WHICH DATASET?, Vietnam WHICH DATASET?.

For some population and housing censuses, results are also representative and available at a second subnational level below the ISO 3166-2 level (Cambodia 2019, Guatemala 2018, Ghana 2021, Kenya 2019, Kiribati 2015 and 2020, Mauritius 2011, Mexico 2020, Morocco 2014, Myanmar 2014, Philippines 2020, Senegal 2013, Tanzania 2012, South-Africa 2011, Uganda 2014, Uruguay 2011, Vanuatu 2009, Vietnam 2009 and 2019).

Go to top

Appendix 4: Indicators

The Disability Statistics – Estimates (DS-E) database covers indicators that are defined below. Each indicator has a short name that is noted below as it is used on the interactive platform.

Proportion with disabilities (Prevalence)

Adults with Disabilities

“Adults with disabilities”, “Individual prevalence” for short, is the proportion of adults ages 15 and older with disabilities. The proportion is available for adults with any disability (disability), by disability type, and by degree (moderate and severe disability). More details on the disability groups are in Appendix 1.

Households with Disabilities

“Households with disabilities”, “Household prevalence” for short, is the proportion of households who have at least one adult ages 15 and older with disability. The proportion is available for households with adults with any disability (disability), and by degree (moderate and severe disability). More details on the disability groups are in Appendix 1.

Education

Adults Who Have Ever Attended School

“Adults who have ever attended school”, “Ever attended school” for short, is the proportion of adults ages 15 and older who have ever been to school.

Adults Ages 25+ Who Have Completed Primary School or Higher

“Adults ages 25+ who have completed primary school or higher”, “At least primary” for short, is the proportion of adults ages 25 and older who at least completed primary school. It is calculated by dividing the number of population ages 25 and older who completed primary education by the total population of the same age group and multiplying by 100.

Adults Ages 25+ Who Have Completed Upper Secondary School or Higher

“Adults ages 25+ who have completed upper secondary school or higher”, “At least secondary” for short, is the proportion of adults ages 25 and older who completed upper secondary school, whether or not they also attended tertiary school. It is calculated by dividing the number of population ages 25 and older who completed upper secondary education by the total population of the same age group and multiplying by 100.

Literacy rate/ Able to Read and Write

“Literacy rate/Able to read and write”, “Literacy” for short, is the proportion of adults 15 and older who can read and write in any language.

Health

Adults in Households Using Safely Managed Drinking Water

“Adults in households using safely managed drinking water”, “Water” for short, is the proportion of adults ages 15 and older who live in households who have safely managed drinking water.

Water sources considered as safely managed include: piped water into dwelling, yard or plot; public taps or standpipes; boreholes or tubewells; protected dug wells; protected springs; packaged water; delivered water and rainwater. Water sources that are not considered as safely managed include: unprotected well, unprotected spring, tanker truck, surface water (river/lake, etc), cart with small tank” (UN Statistics 2017a).

Adults in Households Using Safely Managed Sanitation Services

“Adults in households using safely managed sanitation”, “Sanitation” for short, is the proportion of adults ages 15 and older who live in households who have safely managed sanitation services.

Adults are considered to have safely managed sanitation service if their household’s sanitation facility is improved and is not shared with other households. ‘Improved’ sanitation facilities include: flush or pour flush toilets to sewer systems, septic tanks or pit latrines, ventilated improved pit latrines, pit latrines with a slab, and composting toilets” (UN Statistics 2017b).

Women With Family Planning Needs Met

“Women with family planning needs met”, “Family planning” for short, is the proportion of women who report that they have their family planning needs met, i.e. who want and have access to modern contraceptive methods.

Women Subjected to Violence in The Previous 12 Months

“Women subjected to violence in the previous 12 months”, “Violence” for short, is the proportion of ever-married women who report being subject to domestic violence by their intimate partner in the past 12 months. Domestic violence may be physical, psychological or sexual.

Personal Activities

Adults who Have or Use a Computer

“Adults who have or use a computer”, “Computer” for short, is the proportion of adults ages 15 and older who have or use a computer.

Adults who Have or Use the Internet

“Adults who have or use the internet”, “Internet” for short, is the proportion of adults ages 15 and older who have or use the internet.

Adults who Own a mobile phone

“Adults who own a mobile phone”, “Own mobile” for short, is the proportion of adults ages 15 and older who have their own mobile phone.

Employment Population Ratio (Or Employment Rate)

The “employment population ratio”, also called the “employment rate”, “employment” for short, measures the proportion of the adult population ages 15 and older who work for pay, profit (self-employed) or for a family business/farm (whether paid or unpaid).

Youth Idle Rate

The youth idle rate, also called NEET (Not in Education, Employment or Training) is the proportion of youth ages 15 to 24 who are not enrolled in school and not employed. As information on training was not consistently available, estimates of the youth idle rate do not reflect whether youth might be in training.

Workers in Manufacturing

“Workers in manufacturing”, “manufacturing work” for short, is the proportion of workers ages 15 and older who work in the manufacturing sector.

Women in Managerial Positions

The “Women in managerial positions”, “managerial work” for short, is the proportion of women workers who have managerial positions.

Informal Workers

“Informal workers”, “informal work” for short, is the proportion of workers ages 15 and older who do informal work, i.e. who are self-employed, those who work for a microenterprise of five or fewer employees or in a firm that is unregistered, and those who have no written contract with their employers. Family workers without pay are included as informal workers.

Standard of Living

Adults in Households with Electricity

“Adults in households with electricity”, “Electricity” for short, is the proportion of adults ages 15 and older who live in households with access to electricity (United Nations 2017c).

Access is “only considered if the primary source of lighting is the local electricity provider, solar systems, mini-grids and stand-alone systems. Sources such as generators, candles, batteries, etc., are not considered due to their limited working capacities and since they are usually kept as backup sources for lighting (UN Statistics, 2017c).”

Adults in Households with Clean Cooking Fuel

“Adults in households with clean cooking fuel”, “Clean fuel” for short, is the proportion of adults ages 15 and older who live in households who use clean cooking fuel.

Clean fuel includes electricity, gaseous fuels (e.g. natural gas, biogas). Unclean fuels include kerosene and solid fuels (biomass (wood, crop waste, dung), charcoal, coal) (UN Statistics’ 2017d).

Adults in Households with Adequate Housing

“Adults in households with adequate housing”, “Adequate housing” for short, is the proportion of adults ages 15 and older who live in households with adequate housing.

Adequate housing refers to a household living in a place with quality floor, roof and wall materials. Quality floor conditions include laminates, cement, tiles, bricks, parquet. Poor floor conditions include earth, dung, stone, wood planks. Quality roof conditions include burnt bricks concrete, cement. Poor roof conditions refer to no roof or roofs made of natural or rudimentary materials (e.g. asbestos, thatch, palm leaf, mud, earth, sod, grass, plastic, polythene sheeting, rustic mat, cardboard, canvas, tent, wood planks, reused wood, unburnt bricks). Quality wall conditions include burnt bricks, concrete, cement. Poor wall conditions refer to no walls or walls made of natural or rudimentary materials (e.g. cane, palms, trunk, mud, dirt, grass, reeds, thatch, stone with mud, plywood, cardboard, carton/plastic, canvas, tent, unburnt bricks, reused wood.

Mean Percentage of Assets Owned by Household

The “Mean proportion of assets owned by household”, “Assets” for short, is the proportion of assets owned by an adult’s household among the following assets:  a radio, TV, telephone, mobile phone, bike, motorbike, refrigerator, car (or truck) and computer.

Adults in Households with A Mobile Phone

“Adults in households with a mobile phone”, “Household mobile phone” for short, is the proportion of adults ages 15 and older who live in households with a mobile phone.

Insecurity

Adults Covered by Health Insurance

“Adults covered by health insurance”, “Health insurance” for short, is the proportion of adults ages 15 and older who live in households with health insurance.

Adults in Households Receiving Social Protection

“Adults in households receiving social protection”, “Social protection” for short, is the proportion of adults ages 15 and older who live in households who received social protection benefits in the past year or who currently receive them (e.g. cash benefits, in kind transfers). Benefits may be from Government of Non-Government institutions.

Adults In Food Insecure Households

Proportion of adults ages 15 and older “Adults in food insecure households”, “Food insecurity” for short, is the proportion of adults ages 15 and older who live in households that recently (in the past week, month or 12 months) did not have access to adequate food.  More precisely, (i) the household respondent worried about the household not having enough food or (ii) the household respondent was faced with a situation when they did not have enough food to feed the household; or (iii) there was not enough money to buy food; or (iv) any adult or child in the household went hungry.

Adults In Households That Experienced a Shock Recently

“Adults in households that experienced a shock recently”, “Shock” for short, is the proportion of adults ages 15 and older who live in households that were recently exposed to at least one negative shock. The time frame is usually the past 12 months and shocks include:

• shocks related to the weather (drought, flood, heavy rains),
• negative events affecting household members (death of a household member, illness of a household member),
• economic hardships (loss of a job, crop damage) and disasters (e.g. fire, landslide).

Household Health Expenditures Out of Total Consumption Expenditures

“Household health expenditures out of total consumption expenditures”, “Health expenditures” for short, is the proportion of a household’s total consumption expenditures that are dedicated to health (inpatient care and outpatient care out of pocket expenditures, medicines).

Multidimensional Poverty

Adults who experience Multidimensional Poverty

“Adults who experience multidimensional poverty”, “Multidimensional poverty” for short, is the proportion of adults ages 15 and older who experience more than one deprivation such as not working. This indicator is also called the multidimensional poverty headcount or rate. More details on how it is calculated are in Appendix 5.

References

UN Statistics (2017a). Metadata 06-01-01. Accessed September 30^th 2024 at: 

UN Statistics (2017b). Metadata 06-02-01. Accessed September 30^th 2024 at: 

UN Statistics (2017c). Metadata 07-01-01. Accessed September 30^th 2024 at: 

UN Statistics (2017d). Metadata 07-01-02. Accessed September 30^th 2024 at: 

Go to top

Appendix 5: Multidimensional Poverty

The Disability Statistics – Estimates (DS-E) database uses a multidimensional measure of poverty to investigate the experience of simultaneous deprivations following Alkire and Foster (2011). In brief, this method counts deprivations for a set of dimensions and indicators.

An individual is considered to experience multidimensional poverty if the number of deprivations of the individual exceeds a set threshold. Details on the calculation of this measure are included below. H is the multidimensional poverty headcount (or rate) and gives the percentage of the population who experiences multidimensional poverty or multiple deprivations. Dimensions are weighted and wj is the weight of dimension j. There are different possible methods for setting up weights, for instance, asking people’s opinions or using the observed distribution of successes or deprivations (Decancq and Lugo 2013).

In DS-E, as is often done in multi-dimensional poverty research, all dimensions were considered equally important and were given equal weights (each has a weight of 1) and when more than one indicator was used within a dimension, indicators were equally weighted within the dimension. For instance, for the health dimension with two indicators, each indicator weighs ½.

According to the method laid out in Alkire and Foster (2011), each individual i has a weighted count of dimensions where that person achieves deprivations (c_i) across all measured dimensions: 0≤ c_i ≤ d where d is the number of dimensions; with (c_i) equal to one if individual i has a deprivation in dimension j, and zero otherwise. Let q_i be a binary variable equal to one if the person is identified as deprived, and to zero otherwise. A person is identified as experiencing multidimensional poverty if the person’s count of deprivations is greater than some specified cutoff (k):

if c_i > k, then q_i = 1; if c_i ≤ k, then q_i = 0

In DS-E, k is set at 1. In other words, adults who experience more than one deprivation are considered as multidimensionally poor.

The multidimensional poverty headcount or the proportion of adults experiencing multidimensional poverty H is then the number of persons in multidimensional poverty ( q_i) divided by the total population (n): H=q/n

Dimensions and indicators are laid out in Appendix 3 Table 1 below.  Based on the information available in the datasets under study, four dimensions and eight indicators were selected for the calculation of the multidimensional poverty measure. The four dimensions are: education, personal activities, health, and standard of living. Each dimension has a weight of 1 and when more than one indicator was used within a dimension, indicators were equally weighted within the dimension.

All indicators in the multidimensional poverty measure are defined in Appendix 2, except for assets.  In the multidimensional poverty measure, asset ownership status reflects whether a household owns more than one asset (among radio, TV, telephone, bike, or motorbike or fridge) and if the household does not own a car (or truck).

Education is measured through an indicator of educational attainment for adults ages 15 and older. Personal activities are captured through employment. Health is measured with two indicators and each has a weight of ½: access to safely managed drinking and sanitation services. Standard of living is measured through four indicators with each a weight of ¼: electricity, clean fuel, adequate housing and asset ownership status.

For each indicator, the cutoff for a deprivation is as follows: if a person (1) has less than primary schooling completion; (2) is not working; (3) lives in a household without safely managed drinking water; (4) lives in a household without safely managed sanitation services; (5) lives in a household without clean cooking fuel; (6) lives in a household without adequate housing, i.e. without adequate walls, floor and roof; (7) lives in a household that does not own more than one asset (among radio, TV, telephone, bike, or motorbike or fridge); and the household does not own a car (or truck).

References

Alkire, S. and Foster, J. (2011). Counting and multidimensional poverty measurement. Journal of Public Economics 95(7–8):476–87.

Decancq, K. and Lugo, M. A. (2013). Weights in multidimensional indices of wellbeing: an overview. Econometric Reviews, 32, 7-34.

Go to top

Appendix 6: Data Sources

For the 11 countries with more than one dataset, the default data source for each indicator presented on the DS-E platform is noted in Appendix 6 Table 2.

Go to top

Appendix 7: Datasets and Sampling

Go to top