Health and Activity Limitation Survey : Institutional Component (HALS)

Detailed information for 1991




Every 5 years

Record number:


This was a post-censal disability survey used to identify the numbers and distribution of disabled persons in Canada residing in health related non-penal institutions and the barriers experienced by them.

Data release - August 1994


This was a post-censal disability survey used to identify the numbers and distribution of disabled persons in Canada residing in health related non-penal institutions and the barriers experienced by them.

The Health and Activity Limitation Survey : Institutional Component was discontinued after the 1991 reference period.


  • Disability
  • Equity and inclusion
  • Health
  • Society and community

Data sources and methodology

Target population

The target population of HALS consisted of all persons with a physical or psychological disability who were living in Canada at the time of the Census, including residents of the Yukon and the Northwest Territories, and permanent residents of most collective dwellings and health care institutions. Indian reserves were covered by the Aboriginal Peoples Survey also conducted by Statistics Canada. Persons excluded for operational reasons were residents in penal institutions, correctional facilities, military camps, campgrounds and parks, soup kitchens, merchant and coast guard ships and children's group homes.


This is a sample survey with a cross-sectional design.

The 1991 Census of Population provided the list of institutions, which was used in the first stage of selection for the institutions survey. From this list, six types of institutions were included in HALS. They were:

*nursing homes
*residences for senior citizens
*hospitals: general, maternity, etc.
*chronic care hospitals
*psychiatric institutions
*treatment centres and institutions for the physically handicapped

Institutions were grouped into three categories by size: small, medium and large. These size categories were based on the number of each institution's permanent residents- those who spent a continuous period of six months or longer in an institution. However, the categories "small, medium and large' were determined using different numbers in different provinces.

A sample of institutions was selected based on type and size. All large institutions were included in the survey, while samples of institutions were taken from the small and medium sub-groups.

The administrator of each institution was asked to provide a list of all residents aged 15 and over who on February 1, 1992, had spent a continuous six months or more in an institution. A sample of residents were selected form this list. A maximum sample of 5 residents was taken from small institutions, 10 from medium-sized institutions and 30 from large institutions.

Because of their small size, a census of institutions and their eligible residents was carried out in the Yukon and Northwest Territories.

Data sources

Data collection for this reference period: August 1991 to October 1991

Responding to this survey is mandatory.

Data are collected directly from survey respondents.

All interviews were done in person and, whenever possible, with the selected individual. However, due to their conditions, many residents were unable to answer the questions themselves. In these cases the interviews were conducted with the help of the institutional staff or next-of-kin.

For persons included in the Institutions Survey, there were no corresponding Census data available, because these persons were selected from lists provided by institutions rather than from the Census database. Basic demographic information was collected as part of the Institutions Survey.

Error detection

Data capture for the households and institutional surveys were done in Statistics Canada regional offices. After data capture was completed, the data were transmitted and the questionnaires shipped to Statistics Canada Headquarters in Ottawa for processing and for further reference respectively.


Imputation of age, sex, household size and number of phones in a household, other unknown values converted to "not stated" category.


Statistics from the 1991 Health and Activity Limitation Survey (HALS ) database are estimates based on a sample survey of a portion of the Canadian population (approximately 1 out of 75 persons in the " yes" sample and 1 out of 200 persons in the "no" sample). As a result, the statistics are subject to two types of errors: sampling and non-sampling errors.

In a probability survey such as HALS, each respondent in the sample represents a sub-group of the population observed. Consequently, each database record is assigned a weight corresponding to the number of persons it represents. In addition, the weight is further modified to offset non-responses and discrepancies between the population observed and the target ppopulatioin. The results of the survey are then multiplied by the numerical weight to provide an estimate of what the response would be in the entire population. HALS records were weighted to represent the Canadian population excluding persons not eligible for the survey.

Disclosure control

Statistics Canada is prohibited by law from releasing any information it collects which could identify any person, business, or organization, unless consent has been given by the respondent or as permitted by the Statistics Act. Various confidentiality rules are applied to all data that are released or published to prevent the publication or disclosure of any information deemed confidential. If necessary, data are suppressed to prevent direct or residual disclosure of identifiable data.

Data accuracy

Sampling error is the difference between the estimate derived from a sample and the result that would have been obtained from a population census using the same data collection procedures. For a sample survey such as HALS, this error is estimated from the survey data. The measurement of error used is the standard deviation of the estimate. When a sampling error is more than 33 1/3% of the estimate itself, it is considered to be too unreliable to be published. In such a case, the symbol " -- " appears in statistical tables in place of the estimate. When the sampling error is between 16 2/3% and 33 1/3%, the corresponding estimate is accompanied by the symbol " *" in a table. Such estimates should be used with caution. Finally, all estimates with a sampling error of less than 16 2/3% can be used without restriction.

All other types of errors (coverage, response, processing and non-response) are called non-sampling errors. It is generally difficult to identify and evaluate precisely some of these errors.

Coverage errors arise when there is a difference between the target population and the sample population. Integrating HALS with the census of population has greatly reduced this type of error. Only a certain portion of Indian reserves and collective dwellings were excluded from the sampling process, but because of their small numbers their effect on the total population is negligible. Consequently, coverage errors should not have a significant influence on the HALS data.

A response error occurs when the respondent misunderstands a question, and the interviewer records an incorrect answer.

Processing errors may occur at various stages including: coding, data capture and imputation.

All statistical surveys are susceptible to a certain percentage of non-response among the selected sample. Total non-response occurs when, for one reason or another, a selected respondent cannot be interviewed. Partial non-response occurs if only part of the questionnaire is completed. The impact of non-response errors on estimates depends on the level of non-response and particularly, on any differences between the characteristics of respondents and non-respondents. In principle, the more marked these differences, the greater the impact on the accuracy of the estimates.

With respect to HALS, the response rate of 87% compares favourably with the rate generally observed for this type of survey. In addition, various methods have been used to reduce the bias caused by non-responses (e.g., adjusting the data to reflect the distribution of certain demographic characteristics obtained from the census).


Date modified: