Travel to Work Survey
Detailed information for 1984
Data from this survey is used to assist in the formulation of transportation policies and planning. The survey collects data from individuals traveling to work from randomly chosen households in Canada.
Data release - June 8, 1995
Data from the Travel to Work Survey is used to assist in the formulation of transportation policies and planning. Each Travel to Work supplement provides additional valuable data which is added to the already established data set, thereby increasing the value of this type of series by providing further insight into the transportation and energy problems. The survey collects data from individuals travelling to work from randomly chosen households in Canada.
Reference period: Most of the estimates refer to the week containing the 15th day of the month
- Commuting to work
Data sources and methodology
All persons 15 years of age and over residing in Canada with the exception of inmates of institutions, full-time members of the armed forces, and residents of the Yukon and Northwest Territories, and Indian Reserves. (These exceptions represent less than 3% of the population.)
This is a sample survey with a cross-sectional design.
The survey is based on the multistage stratified, clustered, probability, area sample of the Labour Force Survey.
Responding to this survey is voluntary.
Data are collected directly from survey respondents.
All interviewing is done by personal visit or by telephone. Wherever possible, the interviewer attempts to obtain the information directly from the respondent but failing that, information is accepted from another household member.
Data capture occurs in the regional offices and after the records are transmitted to Ottawa, they are subjected to comprehensive editing, imputation, and tabulation.
The LFS records are weighted using what can be thought of as a three-stage process. The first stage involves the assignment to each record of the inverse of the design sampling ratio applicable to the geographic area where the respondent represented by that record resides. The second stage involves adjustments to the weight assigned in the first stage. These include an adjustment for the rural/urban distribution of the population and an adjustment for non-response (both performed for relatively small sub-provincial areas). It also includes an adjustment for unanticipated population growth in particular small areas selected for the sample (clusters) and an adjustment for the fact that the sample size remains constant (47,500 households) resulting in a slowly declining sampling ratio as the population grows. The third stage involves the comparison of the sum of the weights assigned to the records in the first two stages to population totals derived from sources independent of the LFS. These comparisons are done for 38 age-sex groups for each province. The weights for all records belonging to an age-sex-province group are then adjusted so that their sum is equal to the corresponding independently derived population total. The independently derived population totals are obtained as projections from the annual post-censal estimates of population produced by Demography Division with adjustments to reflect the exclusions described in 'Design and Procedures'.
Statistics Canada is prohibited by law from releasing any information it collects that 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.
The estimates are based on a national sample of slightly less than 1% of the population. The resulting sampling errors, which can be measured, vary according to a number of factors the most important of which is the size of the estimate. Sampling variance indicators are published in 'The Labour Force'.
Errors unrelated to sampling can occur at every stage of a survey. These non-sampling errors range from the respondent misunderstanding the question to errors introduced during processing. Mechanisms to minimize these errors are in place although the final estimates are still affected to some degree.