Provincial Wage and Salary Survey

Detailed information for 2009 (British Columbia Wage and Salary Survey)

Status:

Inactive

Frequency:

Occasional

Record number:

2920

The objective of this survey is to produce statistical information on wages and salaries paid for various occupations classified to the National Occupation Classification (NOC).

Data release - February 8, 2010

Description

This is a client-sponsored special survey that has been conducted in various provinces at different times. The objective of this survey is to produce statistical information on wages and salaries paid for various occupations classified to the National Occupation Classification (NOC).

The results of the Provincial Wage and Salary Survey help governments and businesses by providing accurate and up-to-date information on the wages paid by employers for workers in different occupations and industries.

Subjects

  • Labour
  • Occupations
  • Wages, salaries and other earnings

Data sources and methodology

Target population

The survey population comprises of establishments with high concentrations of occupations targeted based on the 2006 Census in the province of British Columbia. Establishments with 10 or more employees based on Statistics Canada's Business Register are selected.

Instrument design

The questionnaire is based on previous general wage rate surveys undertaken by Small Business and Special Surveys Division of Statistics Canada. Questionnaire testing was conducted on location by a consultant from the Questionnaire Design Resource Center and an observer from the Small Business and Special Surveys project team with recruited participants.

Sampling

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

We are interested in producing estimates by occupation and in order to obtain this information we need to contact an establishment. The establishment records are classified based on the North American Industry Classification System (NAICS) but do not have occupation information for employees working in the establishment. Targeting for occupations is done using Census tables of people employed by 4-digit level NAICS, by 4-digit National Occupation Classification (NOC), and by economic region. Fifteen occupations are associated with each 4-digit level NAICS in a way that optimizes coverage of all the occupations selected for the survey.

Although 15 occupations are assigned to each unit in the sample, to keep the response burden low the respondents are asked to provide wage information for a maximum of 10 of the 15. To preserve optimal coverage, the occupations associated with a given 4-digit level NAICS are ordered randomly for each establishment in that NAICS. This is done so that each occupation has an equal probability of being among the first ten.

Data sources

Data collection for this reference period: 2009-03-01 to 2009-05-30

Responding to this survey is voluntary.

Data are collected directly from survey respondents.

An introductory letter is sent to all establishments in the sample.

The data are collected directly from respondents through a telephone interview. The data are captured by the interviewers during the interview via the Computer Assisted Telephone Interview (CATI) system.

Telephone follow-up is conducted for partially completed and for problematic cases such as extreme wage values. Respondents are asked to provide missing information or to confirm or correct the extreme values.

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

Error detection

Error detection is applied at both the collection and the processing stages of the survey. A series of validity and consistency edits is built into the CATI system. The edits trigger a message to the interviewers to allow corrections to be made during the collection phase. The edits ensure that the relationships between wage and other variables are legitimate. These include, for example, flagging reported wages which are below the provincial or territorial legal minimum wage.

Outlier detection is performed on the wage variables to detect values that are improbable or influential. If a declared wage rate is further than three standard deviations from its mean it is flagged as an outlier. All unusual values and outliers are resolved using computer programs or manually on a case-by-case basis.

Imputation

Imputation is used to replace partially missing data and invalid or inconsistent entries using a combination of mean, ratio and hot deck methods. For the wage variables, the starting wage is imputed using the mean imputation method. The most frequently paid wage and maximum wage are imputed with the ratio method using the starting wage for the base. The total number of employees working on commercial or institutional and the number by occupation are imputed using the mean method.

Estimation

Before estimation and weighting, sample counts are adjusted for non-response. The adjustment of counts is based on the assumptions that the non-response is random and that a non-respondent is not characteristically different from a respondent. Estimates are produced using the Generalized Estimation System (GES).

Records are assigned weights based on the number of employees in the occupations; establishments with more employees for an occupation make a larger contribution to the estimates for that occupation. A weighted-average estimate is calculated for every occupation and for three different hourly wage rates: starting wage, most frequently paid wage and maximum wage. Estimates are also produced by sub-provincial area. Domain estimation is applied. For this survey, the domain regroups all records for a given occupation for the purposes of calculating statistics, regardless of industry and, in some cases, region.

Quality evaluation

Quality control for this survey includes outlier detection during data collection, data validation following capture, and telephone follow-up for partially completed or problematic cases. Data validation is conducted by Small Business and Special Surveys Division using both comparison with published data from both Statistics Canada and other sources and historical trend analysis to gauge the consistency of reporting. Statistics Canada data used include the Labour Force Survey (record number 3701), the Survey of Employment Payroll and Hours (record number 2612).

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.

Confidentiality analysis includes the detection of possible "direct disclosure", which occurs when the value in a tabulation cell is composed of a few respondents or when the cell is dominated by a few companies.

In the BC Wage and Salary Survey, all estimates created with four or fewer occupational records have not been released for confidentiality reasons.

Revisions and seasonal adjustment

This methodology does not apply to this survey.

Data accuracy

Since the BC Wage and Salary Survey estimates are based on sample results, they are subject to sampling error. This error can be expressed as a coefficient of variation (CV). The CV is the standard error expressed as a percentage of the estimate. In the output tables, the CVs are converted to a code. Code A indicates a coefficient of variation in the range 0 to .05 (very good quality), B is .051 to .15 (good quality), C is .151 to .33 (good to poor quality - use with caution) and D is .331 or higher (very poor quality).

Survey estimates may also contain non-sampling error, for example population coverage errors, differences in the interpretation of questions, incorrect information from respondents, and mistakes in recording, coding and processing data. Non-response is an important source of non sampling error. While the impact of non-sampling errors is difficult to evaluate, measures such as response rates and imputation rates can be used as indicators of the potential level of non-sampling error.

Report a problem on this page

Is something not working? Is there information outdated? Can't find what you're looking for?

Please contact us and let us know how we can help you.

Privacy notice

Date modified: