Survey of Employees under Federal Jurisdiction (SEFJ)

Detailed information for 2022

Status:

Active

Frequency:

Every 2 years

Record number:

5312

Statistics Canada is conducting the Survey of Employees under Federal Jurisdiction on behalf of Employment and Social Development Canada. This survey collects information from a sample of employees working for businesses under federal jurisdiction, including regular, seasonal, term, and casual or on-call employees.

Information collected is very important for measuring the quality of employees' work conditions, access to benefits and flexible work arrangements, labour relations, work-related wellbeing, and workplace health and safety. The information from this survey will guide research and analysis to update the Canada Labour Code.

Data release - January 13, 2023

Description

Your contact information was either obtained from your tax records or from your employer in accordance with the Statistics Act and the Income Tax Act.

Your participation is important as the information collected will be used to measure the quality of employees' work conditions, access to benefits and flexible work arrangements, labour relations, work-related wellbeing, and workplace health and safety. The information from this survey will guide research and analysis to update the Canada Labour Code.

Statistics Canada will safeguard your identity by grouping your responses with those of other survey participants. Individual responses and results for very small groups will never be shared or published.

Subjects

  • Commuting to work
  • Health
  • Labour
  • Mental health and well-being
  • Non-wage benefits
  • Unionization and industrial relations
  • Wages, salaries and other earnings

Data sources and methodology

Target population

The target population is employees of active establishments under federal jurisdiction (i.e., regulated by the Canada Labour Code) under a predefined list of NAICS covering the following industry groups:

1-Air Transportation
2-Rail Transportation
3-Road Transportation
4-Maritime Transportation
5-Postal Services and Pipelines
6-Banks
7-Feed, Flour, Seed and Grain
8-Telecommunications

The observed population is employees of establishments from the industry groups listed above that were identified by the Survey of Businesses Under Federal Jurisdiction (2021) as being under federal jurisdiction.

Instrument design

The content for the Survey of Employees under Federal Jurisdiction electronic questionnaire was drafted in consultation with Statistics Canada as well as Employment and Social Development Canada.

The questionnaire underwent cognitive testing in the form of in-depth interviews in both of Canada's official languages, conducted by Statistics Canada's Questionnaire Design Resource Centre. The goal of the qualitative study was to test the survey content.

Sampling

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

This survey uses a two-stage sample of one or more employees from each establishment identified as being under federal jurisdiction by the Survey of Businesses under Federal Jurisdiction (2021). The first stage is the sample of establishments taken for the Survey of Businesses under Federal Jurisdiction (SBFJ). For each establishment found by the SBFJ to be under federal jurisdiction, a random sample of employees is taken. The sampling of employees is the second stage.

Sampling unit
The sampling unit for this survey is a person.

Stratification method
Establishments are stratified by eight industry groups and three size groups at the establishment level (1-19 employees, 20-99 employees, 100 employees or more). Within some industry groups, two or three of the size groups are combined due to a small overall number of establishments in that group.

Sampling and sub-sampling
37,500 employees were selected. This sample was allocated among the strata approximately proportionately to their sizes, where a stratum's size is the estimated total number of employees in Canada.

Data sources

Data collection for this reference period: 2022-01-04 to 2022-03-31

Responding to this survey was voluntary.

Data were collected directly from survey respondents.

Data were collected either through an electronic questionnaire (EQ) or through computer assisted telephone interviews (CATI) in both official languages.

Survey invitations were sent by mail to people in the survey sample. The unique Statistics Canada (StatCan) survey access code (SAC) was provided in the invitation. Respondents were able to log in to the StatCan web portal with their SAC to fill out their questionnaire. Three reminders were sent throughout the course of data collection. Reminders were sent via email to those individuals for whom an email address was available.

StatCan conducted follow up phone calls to non-responders in the sampling frame to increase response rates. This was to provide participants with the opportunity to complete the survey over the phone with a trained StatCan interviewer.

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

Error detection

Error detection was done throughout the survey process.

The survey application was developed and tested thoroughly before collection. Edits were programmed into the respondent electronic questionnaire (r-EQ), as well as into the Collection Management Portal (CMP) that was used to conduct interviews by telephone.

The data capture programs in the survey application allow a valid range of codes for each question and built-in edits, and automatically follow the flow of the questionnaire.

Once the questionnaires were submitted and the responses were added to the database, the same edits as the collections systems were performed as well as additional detailed edits. Any duplicate cases were verified and resolved.

The processing team used the Social Survey Processing Environment (SSPE) set of generalized processing steps and utilities to allow the completion of the processing of the survey in a timely fashion and with high quality outputs.

It used a structured environment to monitor the processing of data ensuring best practices and harmonized business processes were followed.

Edits were performed automatically and manually at various stages of processing at macro and micro levels. They included validity, consistency and flow edits. A series of checks were done to ensure the consistency of survey data. An example of such a consistency check was to ensure that respondents who reported working in 2020 did not report having had 0 employers. Flow edits were used to ensure respondents had followed the correct path and fix off-path situations.

The flow edits carried out ensured that the data set followed the flow of questions in the questionnaire, using a 'top-down' strategy.

A "Valid Skip" code is used to indicate that a question was not asked to a respondent based on an answer to a previous question. For example, if a respondent responded "No" to the question asking if they had experienced harassment or violence at their workplace, they were not asked the follow up questions on harassment or violence. In processing, all of these follow up questions received a "Valid Skip" value (6, 96, 996, etc.).

A "Not Stated" code is used to indicate that the question was shown to the respondent, and they did not respond. It differs from a "Valid Skip" in that the question was asked or shown to the respondent and they did not respond, rather than the respondent not having been shown the question at all. In addition, if no answer was provided for a question, the subsequent follow up questions are also set to "Not Stated". For example, if a respondent did not respond to the question that asks if they had experienced harassment or violence at their workplace, all follow-up questions pertaining to harassment and violence were set to "Not Stated" (9, 99, 999, etc.).

Imputation

A statistical algorithm was used to identify extremely low and extremely high values for hourly wages. A small number of extreme values were then randomly imputed from an appropriate statistical distribution.

Estimation

The principle behind estimation in a probability sample is that each unit in the sample "represents", besides itself, several other units not in the sample. For example, in a simple random 2% sample of the population, each unit in the sample represents 50 units in the population.

Weighting is a process which calculates, for each record, what this number is. This weight appears on the microdata file and must be used to derive meaningful estimates from the survey.

The following steps were performed to calculate sampling weights for the SEFJ:
1) Design weights were generated by computing the inverse of the probability of selection, taking into account the likelihood that the establishment the selected person worked at in 2020 responded to the 2021 SBFJ.

2) Persons who did not work at an establishment targeted in the SBFJ were out of scope for the SEFJ, and they were removed from the weighting process.

3) The weights of the persons that responded to the survey were inflated to account for the persons that did not respond to the survey. Auxiliary information from the sample frame was used to guide the inflation of the respondents' weights.

4) Influential weights were identified and treated using the Sigma Gap method, and a small number of extreme observations had their weights reduced. The persons whose weights were reduced had the amount by which their weights were reduced redistributed among the other respondents in their industry, so the total number of employees estimated to be in each industry did not change.

Variance was estimated using 1,000 bootstrap replicate weights that were generated using a generalized bootstrap weight process. For each replicate, the same adjustments were applied as those used for the survey weights.

Quality evaluation

Quality assurance measures were implemented at every collection and processing step. These measures included recruitment of qualified interviewers; training provided to interviewers for specific survey concepts and procedures; procedures to ensure that data captures are minimized; and edit quality checks to verify the processing logics.

While rigorous quality assurance mechanisms are applied across all steps of the statistical process, validation and scrutiny of the data by statisticians are the ultimate quality checks prior to dissemination. Many validation measures were implemented. They included the following measures (among others):
- Verification of Estimates through Cross-tabulation,
- Coherence Analysis Based on Known Current Events,
- Consultation with Stakeholders Internal to Statistics Canada,
- Consultation with External Stakeholders,
- Review of external production processes.

Disclosure control

Statistics Canada is prohibited by law from releasing any data which would divulge information obtained under the Statistics Act that relates to any identifiable person, business, or organization without the prior knowledge or the consent in writing of that person, business, or organization. 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

Since the SEFJ is a sample survey, all estimates are subject to both sampling and non-sampling errors.

Non-sampling errors can be defined as errors arising during the course of virtually all survey activities, apart from sampling. These include coverage errors, non-response errors, response errors, interviewer errors, coding errors, and other types of processing errors.

The response rate for the SEFJ is calculated as the total number of respondents divided by the size of the in-scope population. The size of the in-scope population is not known, and it is estimated based on the observed in-scope rate among survey respondents. The response rate for the SEFJ is 56.4%.

Non-respondents often have different characteristics from respondents, which can result in bias. Attempts were made to reduce the potential non-response bias as much as possible through weighting adjustments.

Sampling error is defined as the error that arises because an estimate is based on a sample rather than the entire population. The sampling error for the SEFJ is reported through 95% confidence intervals. The 95% confidence interval of an estimate means that if the survey were repeated over and over again, then 95% of the time (or 19 times out of 20), the confidence interval would cover the true population value.

Date modified: