Definition of Disability
Statistics are based on a set of statistical indicators that are designed to measure a specific issue. Critical to the design of statistical indicators is the definition of disability. In the region, the definition of disability is mainly based on the medical model. This means that the definition has a narrow scope, i.e. it is restrictive and exclusionary, leaving out whole segments of population with particular disabilities such as people with intellectual disabilities. Furthermore, the definition can often be contradictory depending on the different laws in which it is specifically defined. For example, in Albania, the National Disability Strategy gives a wide definition of disability but the specific laws on the status of people with disabilities are very narrow and do not include people with intellectual or psycho-social disabilities.
Depending on the choice of definition, statistics will cover a wide or a narrow group of people and this greatly impacts the data sets and the analysis of information. For example, according to the UN Compendium of Disability Statistics (DISTAT) compiled in 1992, the percentage of people who are disabled ranges from 0.2 to as far as 20.9 % of the total population in different countries under survey. A large part of this variation is due to the choice of definition of disability.
Methods for Collecting Data
A country can collect data on disability using three national data collection systems: surveys, population censuses and registries (or administrative records).

1. Surveys
Sample surveys are not intended to enumerate and poll every household or individual in the country. They are designed to be representative of specific portion of population under study so they could be as small as 500 households or as large as 15,000. Surveys should be determined randomly to be representative of the population.
Surveys can be used to collect data on disability either by conducting a special survey on disability or by including questions or a module (a set of questions) on disability in another survey such as Labour Force Surveys, Household Budget Surveys or Living Standards Measurement Surveys.
Surveys cover many different and often specialized topics such as health, welfare, labour force, agriculture and other socioeconomic issues often through a series of detailed questions. The majority of surveys are household-based.
However, in surveys of the population with disabilities it is also important to include the general population for comparative purposes.

2. Population censuses
In most national statistical systems, population censuses are the principal source of statistical data on the population and its characteristics. A census is a nationwide study with every person enumerated separately and their characteristics recorded separately.
Universal enumeration, an essential feature, permits population censuses to provide comprehensive demographic, economic and social data for small geographical areas, which would not be possible with a sample survey. A population census is a complex and costly undertaking that the majority of countries are able to conduct at ten-year intervals only.
Using the census to collect information about a certain segment of the population is not a new concept, although census organizers typically try to avoid using the census to collect anything other than basic demographic information. However, with the high cost of data collection and the increasing need for socio-economic and demographic information, many countries are exploring the census as a way to obtain other types of data.
Although a broad range of topics can be included in a census, most can be covered only briefly because of budgetary, personnel and time constraints.
Disability is increasingly a topic investigated in a population census. For the first time, the United Nations Principles and Recommendations for Population and Housing Censuses, Revision 1, includes disability as a topic that may be covered in censuses.

3. Administrative records and registers
Information on disability can also be collected in administrative recording and registration systems. These include population registers, disability pension registration systems, social security systems, registries of occupational injuries, employment registries, rehabilitation programmes and other services for persons with disabilities. Usually the information in these systems has been collected for reasons other than statistical and is usually related to the administration of a particular programme or service for persons with disabilities who meet its specific criteria.
Registries can be restrictive sources of data firstly because there may be legal problems, such as privacy issues or confidentiality of personal information, in the use of administrative service records or registers for statistical purposes other than in support of the programme or service. Secondly, they only show the people with disabilities who have access to a particular programme or service. Thirdly, in many cases, the data in registries overlap or have duplicate information that cannot be disaggregated. However, in some cases, these barriers can be overcome and, particularly in conjunction with other data sources, administrative records can enrich our knowledge about trends concerning persons with disability.