Provincial Wage and Salary Survey
Detailed information for 2013-2014 (New Brunswick Wage Rate Survey (NBWRS))
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 - July 14, 2014
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.
- Wages, salaries and other earnings
Data sources and methodology
The target population for the 2013-14 New Brunswick Wage Rate Survey (NBWRS) included establishments in New Brunswick in the 3-digit North American Industry Classification System (NAICS) industries with employees in the occupations of interest to the survey sponsor and classified by 4-digit National Occupational Classification (NOC). The NAICS industries were identified using data on employment by occupation and industry from the 2011 Census. Establishments with less than 6 employees were excluded.
The NBWRS questionnaire was designed by Statistics Canada in collaboration with the New Brunswick Department of Post-Secondary Education, Training and Labour. The questionnaire asked respondent establishments to report their total number of employees and the number of full-time employees for each of up to 12 occupations. In addition, they were asked to provide information on the union status, usual hours worked and five different wage rates for full-time employees in up to 4 occupations.
Questionnaire testing in both English and French was conducted in Moncton by the Questionnaire Design Resource Centre and the Centre for Special Business Projects of Statistics Canada. The results of the testing were used to further improve the questionnaire.
This is a sample survey with a cross-sectional design.
The NBWRS used the September 2013 version of Statistics Canada's Business Register (BR) as its sampling frame. The BR contains the universe of establishments in Canada. It was constructed using various types of tax records from the Canada Revenue Agency and is updated regularly based on feedback from various sources including business surveys. Establishments in industries that had workers in the occupations specified by the sponsor and that had 6 or more employees were in-scope for this survey. The final frame included 9,587 units.
The frame was stratified by region (including Moncton, Saint John, Fredericton and the rest of the New Brunswick) and by industry based on the 3-digit level of the 2012 North American Industry Classification System (NAICS).
The occupation component of the stratification was based on a list of occupations defined to balance the needs of the survey sponsor with the feasibility of finding enough information about these occupations in the survey. The list included 140 occupations.
The initial size of the NBWRS sample was established to ensure that a minimum number of establishments would be selected to produce quality estimates for each of the 4 Economic Region (ER) and 140 occupation combinations. It was assumed that the survey response rate and the number of occupations reported by each establishment would be similar to those observed for the 2003 iteration of the survey. The survey sample was increased until every occupation of interest would have the greatest possibility of having at least 5 respondents at provincial level. Finally, additional units were selected to compensate for in-sample units which would not be contacted for data for various reasons. The final sample size was 6,341 establishments.
Targeting for occupations was done using tabulations from the 2011 Census of employed individuals by 3-digit level NAICS, by 4-digit National Occupation Classification (NOC), and by Economic Region (ER) in New Brunswick. These tabulations were used to identify and select the top 12 occupations in terms of employment for each 3-digit NAICS and Economic Region stratum. The order of these 12 occupations was randomized for each establishment in the same 3-digit NAICS and region. Respondents were asked to report information for the first 4 relevant occupations in the randomized list assigned to them to keep respondent burden low. For the 3-digit NAICS and region combinations with less than 12 occupations, all the occupations were kept and the order was randomized. Additional occupations were selected based on employment for the 1 and 2-digit levels of NAICS in the same region. These additional occupations were randomized separately and added to the end of the first randomized list. It is possible that a final list contained less than 12 occupations after performing this extra step.
The final sample of establishments was allocated among strata using the square-root proportional allocation method. Once the total sample size was allocated to each stratum, stratified simple random sampling was used to select the final sample of units.
Data collection for this reference period: 2013-11-04 to 2014-02-07
Responding to this survey is voluntary.
Data are collected directly from survey respondents.
The NBWRS was collected based on computer-assisted telephone interviews. Each establishment in the sample received an introductory letter prior to data collection which described the purpose of the survey, indicated what questions were going to be asked and listed the occupations that respondents could be asked about so that they could prepare for the telephone interview.
Units with no valid telephone number were traced so that they could eventually be contacted. In some cases, establishments could not provide the information requested and referred us to their head offices. In other cases, special arrangements were made to collect the survey data using spreadsheets and secure electronic file transfer.
View the Questionnaire(s) and reporting guide(s) .
Error detection is an integral part of both collection and data processing activities. Automated edits were applied to data records during collection to identify capture and reporting errors. Respondents were asked to validate their reported data when these collection edits failed.
Prior to imputation, a series of edits were applied to the collected data to identify errors and inconsistencies. Outlier detection was also performed on most variables to identify improbable or influential values. If the value of a variable was further than a certain distance from the mean value for a specific domain, it was flagged for review as an outlier.
Errors and inconsistencies in the data were reviewed and resolved by referring to data for similar units in the survey, data from other surveys and information from external sources. If a record could not be resolved, it was flagged for imputation. Also, all reported wages were converted to hourly rates during editing.
Finally, edit rules were incorporated into the imputation system to detect and resolve any remaining errors, as well as to ensure that the imputed data were consistent.
Imputation is used to determine plausible values for missing, inconsistent or outlier values in the collected data which have not been resolved through editing.
A number of different approaches were used to impute missing or inconsistent data for the NBWRS. The simplest technique involved using deterministic and coherence rules that dictate acceptable relationships among variables and derive missing values residually. Missing variables were often imputed by applying the ratios between variables to data records with partial information. Ratio imputation was used frequently in the case of missing wage data because of the strong linear relationship between wage variables. Donor imputation was also used for non-wage variables and when ratio imputation for wages was not possible. Donor imputation involved identifying a respondent record (donor) that was similar to the record which required imputation (recipient) based on information which was available for both businesses. The data available for the respondent was then used to derive that for the record requiring imputation. If there was no wage information for a respondent, all wages were imputed from a donor record. In addition, donor imputation was used exclusively for the imputation of missing maximum wages since this variable was not well correlated with other collected wage data.
The imputation and donor classes used for the imputation of wage rates were defined based on occupation, business size and industry. These were applied during imputation in the following order from most to least detailed :
NOC, business size, NAICS4
NOC, business size, NAICS3
NOC, business size, NAICS2
NOC, business size
Only in cases when there were not enough units in a given class would a more aggregated imputation class be used.
Most imputation of survey variables was performed in an automated way using BANFF, a generalized system designed by Statistics Canada. No imputation was done for complete non-response.
A weighted-average estimate was calculated for every occupation and for five different hourly wage rates for full-time employees. These wage rates included the starting wage with no experience, the starting wage with one year of previous experience, the wage after one year on the job, the average wage and the maximum wage. Estimates are also produced by sub-provincial area and industry.
The weights assigned to each record included in the calculation of a weighted-average have two components. The first is the adjusted sampling weight assigned to the establishment included in the estimate. The sampling strata for the NBWRS were defined by Economic Region and 3-digit NAICS. The initial sampling weights were calculated for each establishment as the ratio of the total number of units in the stratum to which the establishment was assigned, divided by the number of sample units in the stratum. These were adjusted before estimation to account for non-responding and out-of-scope units. These adjustments were based on the assumptions that non-response is random and that a non-respondent is not characteristically different from a respondent. The second weight component reflects the number of employees reported by the establishment in the occupation for which an average is being calculated. Establishments with more employees for an occupation make a larger contribution to the estimates for that occupation.
Estimates were produced using Statistics Canada's Generalised Estimation System (GES).
Where possible, estimates of average hourly wages from the NBWRS were compared with similar data from the other Statistics Canada surveys including the Census, the Labour Force Survey and the Survey of Employment, Payrolls and Hours, as well as with external sources. In addition, subject matter experts from outside Statistics Canada were given an opportunity to review the estimates and provide feedback on their quality prior to their official release.
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.
In order to prevent any data disclosure, confidentiality analysis was done using the G-CONFID system. G-CONFID was used for primary confidentiality as well as for the secondary suppression (residual disclosure). Direct disclosure or primary confidentiality occurs when the value in a tabulation cell is composed or dominated by few enterprises. Residual disclosure occurs when confidential information can be derived indirectly by piecing together information from different sources or data series.
Revisions and seasonal adjustment
This methodology does not apply to this survey.
Data quality for the NBWRS has been assessed based on measures of sampling error and non- sampling error. Sampling errors occur as a result of taking a sample of the population of interest. Non-sampling error is not related to sampling and may occur for various reasons during the collection and processing of data. For example, non-response is an important source of non-sampling error. Under or over-coverage of the population, differences in the interpretations of questions and mistakes in recording, coding and processing data are other examples of non-sampling errors. To the maximum extent possible, these errors are minimized through careful design of the survey questionnaire, verification of the survey data, and follow-up with delinquent respondents to maximize response rates.
Sampling error for the NBWRS is measured based on coefficients of variation (CV). The CV is a percentage that expresses the size of the standard error as a proportion of the estimate to which it is related. For example, a CV of 10% indicates that the standard error is 10% of the estimate. If a wage rate estimate is $15.00 per hour, with a CV of 10%, then the standard error is $1.50.
A weighted imputation rate provides one measure of non-sampling error. It is defined as the percentage of a final estimate resulting from imputation for an occupation in the population.
The quality indicators in the published tables for the NBWRS reflect a combination of CVs and weighted imputation rates.