Labour Market Indicators

Detailed information for June 2024

The purpose of this survey is to identify emerging trends in the Canadian labour market. Gathering data on topics such as work locations, types of employment and quality of employment will allow policy makers to address these trends.

Data release - July 5, 2024

• Questionnaire(s) and reporting guide(s)
• Description
• Data sources and methodology
• Data accuracy

Description

Objectives:
- Results from this Supplement survey will provide a better understanding of labour market indicators on key trends for diverse groups of Canadians, and will inform government as it reacts to these trends.

Topics Studied:
- Topics will be modified on a regular basis to provide timely insights on targeted concepts related to the state of the labour force.
- Content on working from home will be included in the Supplement on a monthly basis, giving a better picture of the current structure of the workforce.

Reference period: Monthly

Collection period: Data is collected for nine days following the reference period each month.

Subjects

• Labour

Data sources and methodology

Target population

The survey is conducted in the provinces only, and excludes respondents in their first or last rotation of the Labour Force Survey (LFS).

Instrument design

All questions have been reviewed by the Questionnaire Design Resource Centre (QDRC).

Sampling

This is a sample survey with a longitudinal design.

The frame for this supplement survey comes from the Labour Force Survey (LFS).

The sample is made up of Labour Force Survey (LFS) respondents from the provinces only, and excludes respondents in their first or last rotation of LFS. This is a longitudinal random sample of households from dwellings across Canada (more details available on the LFS webpage record number 3701).

Sampling unit
This is a dwelling based survey.

Stratification method
The stratification methodology is the same as the one used for the Canadian Labour Force Survey (Catalogue No. 71-526-X, 2.5 Stratification, available online).

Sampling and sub-sampling
Sample consists of households in their 2nd, 3rd, 4th, or 5th month of LFS. Estimated sample size is 40,000 dwellings.

Data sources

Data collection for this reference period: 2024-06-09 to 2024-06-15

Responding to this survey is voluntary.

Data are collected directly from survey respondents.

- Collection method: CMP EQ, CATI & CAPI
- Capture method: CMP rEQ/iEQ
- Method of initial contact: mail-out letter
- Follow-up method: CATI
- Use of proxy reporting: proxy reporting is used for this survey
- Language(s) offered to potential respondents: English & French
- Average time required to complete the interview/survey: 5 minutes.

For imputation and validation purposes, the data collected is merged with the main LFS questionnaire. No other source is used in data integration.

View the Questionnaire(s) and reporting guide(s) .

Error detection

A series of verification steps are performed to identify and eliminate potential duplicate records and to exclude non-response and out-of-scope records.

Any record that is on the main LFS file, that is not a birth (respondents that are in their first month of the LFS) or a rotate-out (respondents that are in their last month of the LFS), should also exist on the LMI file. This step adds any missing records to the LMI file (with missing values for all variables).

Editing is also performed at this step by modifying the data at the individual variable level, to correct for inconsistent or invalid responses.

Imputation

Imputation was done using the nearest neighbour donor imputation method. Persons that responded to all of the questions are considered "donor" candidates, while those with missing or invalid data are considered "recipients". A distance measure between donors and recipients is calculated, based on their similarities in terms of labour force status (employed, absent, unemployed, not in the labour force), industry (NAICS) and occupation (NOC), province, and other variables describing demographic and economic characteristics. Data from the best possible donor are used to fill in the missing values for the recipient.

Imputation was carried out when individual items were missing or invalid, and also for entire records that were deemed eligible for LMI based on their TABS data. As such, all records from in-scope rotations on TABS have a corresponding record on LMI. For records that were out of scope for LMI based on other reasons (e.g. Age>69), all variables on the LMI file will have a value of "Valid Skip".

Estimation

For information on the calculation of the LFS weights, refer to the Methodology of the Canadian Labour Force Survey (Catalogue No. 71-526-X, available online).

After the calculation of the LFS final weights, further adjustments are made to derive a weight for the individual records on the LMI Supplement microdata file:
• an adjustment to account for using only four of the six rotations in the LFS sample;
• an adjustment to calibrate the weights to 18 age-sex groups, labour force status (employed, unemployed, not in the labour force) and rotation group.

The calibration totals are based on final LFS estimates, which themselves are based on population projections from Statistics Canada's Demography division.

Bootstrap Weights
The bootstrap weights file provided contains 1,000 bootstrap weight replicates. In general, variance estimation using bootstrap weights is done by computing the estimate of a parameter using each set of bootstrap weights and then computing the variance of the resulting estimates. This method is valid not only for totals, means, and ratios, but also for non-linear parameters such as medians and other quantiles. These bootstrap weights can be used to compute a variety of variance estimates based on the LMI Supplement data. The bootstrap weights for the LMI Supplement were calculated starting from the LFS bootstrap weights, and applying the same non-response adjustment and calibration that were applied to create the final LMI Supplement weights. For more information on the calculation and use of the LFS bootstrap weights, refer to the Methodology of the Canadian Labour Force Survey (Catalogue No. 71-526-X, available online).

Quality evaluation

Selected totals and their coefficients of variation were estimated with the LMI weight and compared to the estimates using the final LFS weight. The estimates were:
1. Employment, Unemployment, and Not in Labour Force, by province
2. Population by province

The validation step confirmed that totals were the same, and that their coefficients of variation were within 0.1%.

Disclosure control

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.

Revisions and seasonal adjustment

This methodology type does not apply to this statistical program.

Data accuracy

The quality of estimates calculated from the Supplement file are similar to those obtained from the main LFS. The content of the Supplement changes from month to month, but one recurrent theme is the proportion of employed people that work from home. The coefficient of variation associated with that estimate is typically around 7% for the smaller provinces, and around 1% at the Canada level.