Monthly Retail Trade Survey (MRTS) Data Quality Statement

Objectives, uses and users
Concepts, variables and classifications
Coverage and frames
Sampling
Questionnaire design
Response and non-response
Data collection and capture operations
Editing
Imputation
Estimation
Revisions and seasonal adjustment
Data quality evaluation
Disclosure control

1. Objectives, uses and users

1.1. Objective

The Monthly Retail Trade Survey (MRTS) provides information on the performance of the retail trade sector on a monthly basis, and when combined with other statistics, represents an important indicator of the state of the Canadian economy.

1.2. Uses

The estimates provide a measure of the health and performance of the retail trade sector. Information collected is used to estimate level and monthly trend for retail sales. At the end of each year, the estimates provide a preliminary look at annual retail sales and performance.

1.3. Users

A variety of organizations, sector associations, and levels of government make use of the information. Retailers rely on the survey results to compare their performance against similar types of businesses, as well as for marketing purposes. Retail associations are able to monitor industry performance and promote their retail industries. Investors can monitor industry growth, which can result in better access to investment capital by retailers. Governments are able to understand the role of retailers in the economy, which aids in the development of policies and tax incentives. As an important industry in the Canadian economy, governments are able to better determine the overall health of the economy through the use of the estimates in the calculation of the nation’s Gross Domestic Product (GDP).

2. Concepts, variables and classifications

2.1. Concepts

The retail trade sector comprises establishments primarily engaged in retailing merchandise, generally without transformation, and rendering services incidental to the sale of merchandise.

The retailing process is the final step in the distribution of merchandise; retailers are therefore organized to sell merchandise in small quantities to the general public. This sector comprises two main types of retailers, that is, store and non-store retailers. The MRTS covers only store retailers. Their main characteristics are described below. Store retailers operate fixed point-of-sale locations, located and designed to attract a high volume of walk-in customers. In general, retail stores have extensive displays of merchandise and use mass-media advertising to attract customers. They typically sell merchandise to the general public for personal or household consumption, but some also serve business and institutional clients. These include establishments such as office supplies stores, computer and software stores, gasoline stations, building material dealers, plumbing supplies stores and electrical supplies stores.

In addition to selling merchandise, some types of store retailers are also engaged in the provision of after-sales services, such as repair and installation. For example, new automobile dealers, electronic and appliance stores and musical instrument and supplies stores often provide repair services, while floor covering stores and window treatment stores often provide installation services. As a general rule, establishments engaged in retailing merchandise and providing after sales services are classified in this sector. Catalogue sales showrooms, gasoline service stations, and mobile home dealers are treated as store retailers.

2.2. Variables

Sales are defined as the sales of all goods purchased for resale, net of returns and discounts. This includes commission revenue and fees earned from selling goods and services on account of others, such as selling lottery tickets, bus tickets, and phone cards. It also includes parts and labour revenue from repair and maintenance; revenue from rental and leasing of goods and equipment; revenues from services, including food services; sales of goods manufactured as a secondary activity; and the proprietor’s withdrawals, at retail, of goods for personal use. Other revenue from rental of real estate, placement fees, operating subsidies, grants, royalties and franchise fees are excluded.

Trading Location is the physical location(s) in which business activity is conducted in each province and territory, and for which sales are credited or recognized in the financial records of the company. For retailers, this would normally be a store.

Constant Dollars: The value of retail trade is measured in two ways; including the effects of price change on sales and net of the effects of price change. The first measure is referred to as retail trade in current dollars and the latter as retail trade in constant dollars. The method of calculating the current dollar estimate is to aggregate the weighted value of sales for all retail outlets. The method of calculating the constant dollar estimate is to first adjust the sales values to a base year, using the Consumer Price Index, and then sum up the resulting values.

2.3. Classification

The Monthly Retail Trade Survey is based on the definition of retail trade under the NAICS (North American Industry Classification System). NAICS is the agreed upon common framework for the production of comparable statistics by the statistical agencies of Canada, Mexico and the United States. The agreement defines the boundaries of twenty sectors. NAICS is based on a production-oriented, or supply based conceptual framework in that establishments are groups into industries according to similarity in production processes used to produce goods and services.

Estimates appear for 21 industries based on special aggregations of the 2012 North American Industry Classification System (NAICS) industries. The 21 industries are further aggregated to 11 sub-sectors.

Geographically, sales estimates are produced for Canada and each province and territory.

3. Coverage and frames

Statistics Canada’s Business Register ( BR) provides the frame for the Monthly Retail Trade Survey. The BR is a structured list of businesses engaged in the production of goods and services in Canada. It is a centrally maintained database containing detailed descriptions of most business entities operating within Canada. The BR includes all incorporated businesses, with or without employees. For unincorporated businesses, the BR includes all employers with businesses, and businesses with no employees with annual sales that have a Goods and Services Tax (GST) or annual revenue that declares individual taxes.  annual sales greater than $30,000 that have a Goods and Services Tax (GST) account (the BR does not include unincorporated businesses with no employees and with annual sales less than $30,000).

The businesses on the BR are represented by a hierarchical structure with four levels, with the statistical enterprise at the top, followed by the statistical company, the statistical establishment and the statistical location. An enterprise can be linked to one or more statistical companies, a statistical company can be linked to one or more statistical establishments, and a statistical establishment to one or more statistical locations.

The target population for the MRTS consists of all statistical establishments on the BR that are classified to the retail sector using the North American Industry Classification System (NAICS) (approximately 200,000 establishments). The NAICS code range for the retail sector is 441100 to 453999. A statistical establishment is the production entity or the smallest grouping of production entities which: produces a homogeneous set of goods or services; does not cross provincial boundaries; and provides data on the value of output, together with the cost of principal intermediate inputs used, along with the cost and quantity of labour used to produce the output. The production entity is the physical unit where the business operations are carried out. It must have a civic address and dedicated labour.

The exclusions to the target population are ancillary establishments (producers of services in support of the activity of producing goods and services for the market of more than one establishment within the enterprise, and serves as a cost centre or a discretionary expense centre for which data on all its costs including labour and depreciation can be reported by the business), future establishments, establishments with a missing or a zero gross business income (GBI) value on the BR and establishments in the following non-covered NAICS:

  • 4541 (electronic shopping and mail-order houses)
  • 4542 (vending machine operators)
  • 45431 (fuel dealers)
  • 45439 (other direct selling establishments)

4. Sampling

The MRTS sample consists of 10,000 groups of establishments (clusters) classified to the Retail Trade sector selected from the Statistics Canada Business Register. A cluster of establishments is defined as all establishments belonging to a statistical enterprise that are in the same industrial group and geographical region. The MRTS uses a stratified design with simple random sample selection in each stratum. The stratification is done by industry groups (the mainly, but not only four digit level NAICS), and the geographical regions consisting of the provinces and territories, as well as three provincial sub-regions. We further stratify the population by size.

The size measure is created using a combination of independent survey data and three administrative variables: the annual profiled revenue, the GST sales expressed on an annual basis, and the declared tax revenue (T1 or T2). The size strata consist of one take-all (census), at most, two take-some (partially sampled) strata, and one take-none (non-sampled) stratum. Take-none strata serve to reduce respondent burden by excluding the smaller businesses from the surveyed population. These businesses should represent at most ten percent of total sales. Instead of sending questionnaires to these businesses, the estimates are produced through the use of administrative data.

The sample was allocated optimally in order to reach target coefficients of variation at the national, provincial/territorial, industrial, and industrial groups by province/territory levels. The sample was also inflated to compensate for dead, non-responding, and misclassified units.

MRTS is a repeated survey with maximisation of monthly sample overlap. The sample is kept month after month, and every month new units are added (births) to the sample.  MRTS births, i.e., new clusters of establishment(s), are identified every month via the BR’s latest universe. They are stratified according to the same criteria as the initial population. A sample of these births is selected according to the sampling fraction of the stratum to which they belong and is added to the monthly sample. Deaths occur on a monthly basis. A death can be a cluster of establishment(s) that have ceased their activities (out-of-business) or whose major activities are no longer in retail trade (out-of-scope). The status of these businesses is updated on the BR using administrative sources and survey feedback, including feedback from the MRTS. Methods to treat dead units and misclassified units are part of the sample and population update procedures.

5. Questionnaire design

The Monthly Retail Trade Survey incorporates the following sub-surveys:

Monthly Retail Trade Survey - R8

Monthly Retail Trade Survey (with inventories) – R8

Survey of Sales and Inventories of Alcoholic Beverages

The questionnaires collect monthly data on retail sales and the number of trading locations by province or territory and inventories of goods owned and intended for resale from a sample of retailers. The items on the questionnaires have remained unchanged for several years. For the 2004 redesign, the general questionnaires were subject to cosmetic changes only. The questionnaire for Sales and Inventories of Alcoholic Beverages underwent more extensive changes. The modifications were discussed withstakeholders and the respondents were given an opportunity to comment before the new questionnaire was finalized. If further changes are needed to any of the questionnaires, proposed changes would go through a review committee and a field test with respondents and data users to ensure its relevancy.

6. Response and non-response

6.1. Response and non-response

Despite the best efforts of survey managers and operations staff to maximize response in the MRTS, some non-response will occur. For statistical establishments to be classified as responding, the degree of partial response (where an accurate response is obtained for only some of the questions asked a respondent) must meet a minimum threshold level below which the response would be rejected and considered a unit non-response.  In such an instance, the business is classified as not having responded at all.

Non-response has two effects on data: first it introduces bias in estimates when non-respondents differ from respondents in the characteristics measured; and second, it contributes to an increase in the sampling variance of estimates because the effective sample size is reduced from that originally sought.

The degree to which efforts are made to get a response from a non-respondent is based on budget and time constraints, its impact on the overall quality and the risk of non-response bias.

The main method to reduce the impact of non-response at sampling is to inflate the sample size through the use of over-sampling rates that have been determined from similar surveys.

Besides the methods to reduce the impact of non-response at sampling and collection, the non-responses to the survey that do occur are treated through imputation. In order to measure the amount of non-response that occurs each month, various response rates are calculated. For a given reference month, the estimation process is run at least twice (a preliminary and a revised run). Between each run, respondent data can be identified as unusable and imputed values can be corrected through respondent data. As a consequence, response rates are computed following each run of the estimation process.

For the MRTS, two types of rates are calculated (un-weighted and weighted). In order to assess the efficiency of the collection process, un-weighted response rates are calculated. Weighted rates, using the estimation weight and the value for the variable of interest, assess the quality of estimation. Within each of these types of rates, there are distinct rates for units that are surveyed and for units that are only modeled from administrative data that has been extracted from GST files.

To get a better picture of the success of the collection process, two un-weighted rates called the ‘collection results rate’ and the ‘extraction results rate’ are computed. They are computed by dividing the number of respondents by the number of units that we tried to contact or tried to receive extracted data for them. Non-monthly reporters (respondents with special reporting arrangements where they do not report every month but for whom actual data is available in subsequent revisions) are excluded from both the numerator and denominator for the months where no contact is performed.

In summary, the various response rates are calculated as follows:

Weighted rates:

Survey Response rate (estimation) =
Sum of weighted sales of units with response status i / Sum of survey weighted sales

where i = units that have either reported data that will be used in estimation or are converted refusals, or have reported data that has not yet been resolved for estimation.

Admin Response rate (estimation) =
Sum of weighted sales of units with response status ii / Sum of administrative weighted sales

where ii = units that have data that was extracted from administrative files and are usable for estimation.

Total Response rate (estimation) =
Sum of weighted sales of units with response status i or response status ii / Sum of all weighted sales

Un-weighted rates:

Survey Response rate (collection) =
Number of questionnaires with response status iii/ Number of questionnaires with response status iv

where iii = units that have either reported data (unresolved, used or not used for estimation) or are converted refusals.

where iv = all of the above plus units that have refused to respond, units that were not contacted and other types of non-respondent units.

Admin Response rate (extraction) =
Number of questionnaires with response status vi/ Number of questionnaires with response status vii

where vi = in-scope units that have data (either usable or non-usable) that was extracted from administrative files

where vii = all of the above plus units that have refused to report to the administrative data source, units that were not contacted and other types of non-respondent units.

(% of questionnaire collected over all in-scope questionnaires)

Collection Results Rate =
Number of questionnaires with response status iii / Number of questionnaires with response status viii

where iii = same as iii defined above

where viii = same as iv except for the exclusion of units that were contacted because their response is unavailable for a particular month since they are non-monthly reporters.

Extraction Results Rate =
Number of questionnaires with response status ix / Number of questionnaires with response status vii

where ix = same as vi with the addition of extracted units that have been imputed or were out of scope

where vii = same as vii defined above

(% of questionnaires collected over all questionnaire in-scope we tried to collect)

All the above weighted and un-weighted rates are provided at the industrial group, geography and size group level or for any combination of these levels.

Use of Administrative Data

Managing response burden is an ongoing challenge for Statistics Canada. In an attempt to alleviate response burden and survey costs, especially for smaller businesses, the MRTS has reduced the number of simple establishments in the sample that are surveyed directly and instead derives sales data for these establishments from Goods and Service Tax (GST) files using a statistical model. The model accounts for differences between sales and revenue (reported for GST purposes) as well as for the time lag between the survey reference period and the reference period of the GST file.

For more information on the methodology used for modeling sales from administrative data sources, refer to ‘Monthly Retail Trade Survey: Use of Administrative Data’ under ‘Documentation’ of the IMDB.

Table 1 contains the weighted response rates for all industry groups as well as for total retail trade for each province and territory. For more detailed weighted response rates, please contact the Marketing and Dissemination Section at (613) 951-3549, toll free: 1-877-421-3067 or by e-mail at retailinfo@statcan.

6.2. Methods used to reduce non-response at collection

Significant effort is spent trying to minimize non-response during collection. Methods used, among others, are interviewer techniques such as probing and persuasion, repeated re-scheduling and call-backs to obtain the information, and procedures dealing with how to handle non-compliant (refusal) respondents.

If data are unavailable at the time of collection, a respondent's best estimates are also accepted, and are subsequently revised once the actual data become available.

To minimize total non-response for all variables, partial responses are accepted. In addition, questionnaires are customized for the collection of certain variables, such as inventory, so that collection is timed for those months when the data are available.

Finally, to build trust and rapport between the interviewers and respondents, cases are generally assigned to the same interviewer each month. This action establishes a personal relationship between interviewer and respondent, and builds respondent trust.

7. Data collection and capture operations

Collection of the data is performed by Statistics Canada’s Regional Offices.

Table 1: Weighted response rates by NAICS, for all provinces and territories: May 2016
Table summary
This table displays the results of Table 1: Weighted response rates by NAICS Weighted Response Rates, calculated using Total, Survey and Administrative units of measure (appearing as column headers).
  Weighted Response Rates
Total Survey Administrative
NAICS - Canada  
Motor Vehicle and Parts Dealers 90.4 91.1 66.7
Automobile Dealers 91.9 92.2 69.0
New Car Dealers 93.3 93.3 Note ...: not applicable
Used Car Dealers 69.7 69.9 69.0
Other Motor Vehicle Dealers 76.7 78.7 65.7
Automotive Parts, Accessories and Tire Stores 84.1 86.6 64.7
Furniture and Home Furnishings Stores 87.8 90.8 59.7
Furniture Stores 91.4 92.3 75.0
Home Furnishings Stores 81.2 87.7 52.9
Electronics and Appliance Stores 82.9 83.3 65.6
Building Material and Garden Equipment Dealers 87.5 90.0 48.6
Food and Beverage Stores 83.8 86.1 51.4
Grocery Stores 88.2 90.8 54.3
Grocery (except Convenience) Stores 91.1 93.6 57.7
Convenience Stores 48.4 51.2 29.4
Specialty Food Stores 59.0 63.1 38.0
Beer, Wine and Liquor Stores 73.8 74.5 38.6
Health and Personal Care Stores 88.4 88.4 87.1
Gasoline Stations 76.9 78.6 52.7
Clothing and Clothing Accessories Stores 88.1 89.3 42.8
Clothing Stores 88.9 89.9 46.2
Shoe Stores 88.2 89.3 13.6
Jewellery, Luggage and Leather Goods Stores 80.7 83.2 42.0
Sporting Goods, Hobby, Book and Music Stores 87.9 92.6 31.1
General Merchandise Stores 99.3 99.4 90.2
Department Stores 100.0 100.0 Note ...: not applicable
Other general merchandise stores 98.9 99.0 90.2
Miscellaneous Store Retailers 77.9 82.9 28.5
Total 88.0 89.4 56.5
Regions  
Newfoundland and Labrador 82.0 82.9 48.2
Prince Edward Island 82.3 83.3 7.0
Nova Scotia 92.7 93.8 54.9
New Brunswick 84.3 85.5 52.8
Québec 89.7 91.5 57.7
Ontario 89.2 91.0 54.3
Manitoba 81.8 82.1 64.1
Saskatchewan 88.6 89.6 65.2
Alberta 86.9 88.1 64.4
British Columbia 85.1 86.4 48.9
Yukon Territory 78.4 78.4 Note ...: not applicable
Northwest Territories 82.5 82.5 Note ...: not applicable
Nunavut 94.0 94.0 Note ...: not applicable


Weighted Response Rates

Respondents are sent a questionnaire or are contacted by telephone to obtain their sales and inventory values, as well as to confirm the opening or closing of business trading locations. Collection of the data begins approximately 7 working days after the end of the reference month and continues for the duration of that month.

New entrants to the survey are introduced to the survey via an introductory letter that informs the respondent that a representative of Statistics Canada will be calling. This call is to introduce the respondent to the survey, confirm the respondent's business activity, establish and begin data collection, as well as to answer any questions that the respondent may have.

8. Editing

Data editing is the application of checks to detect missing, invalid or inconsistent entries or to point to data records that are potentially in error. In the survey process for the MRTS, data editing is done at two different time periods.

First of all, editing is done during data collection. Once data are collected via the telephone, or via the receipt of completed mail-in questionnaires, the data are captured using customized data capture applications. All data are subjected to data editing. Edits during data collection are referred to as field edits and generally consist of validity and some simple consistency edits. They are used to detect mistakes made during the interview by the respondent or the interviewer and to identify missing information during collection in order to reduce the need for follow-up later on. Another purpose of the field edits is to clean up responses. In the MRTS, the current month’s responses are edited against the respondent’s previous month’s responses and/or the previous year’s responses for the current month. Field edits are also used to identify problems with data collection procedures and the design of the questionnaire, as well as the need for more interviewer training.

Follow-up with respondents occurs to validate potential erroneous data following any failed preliminary edit check of the data. Once validated, the collected data is regularly transmitted to the head office in Ottawa.

Secondly, editing known as statistical editing is also done after data collection and this is more empirical in nature. Statistical editing is run prior to imputation in order to identify the data that will be used as a basis to impute non-respondents. Large outliers that could disrupt a monthly trend are excluded from trend calculations by the statistical edits. It should be noted that adjustments are not made at this stage to correct the reported outliers.

The first step in the statistical editing is to identify which responses will be subjected to the statistical edit rules. Reported data for the current reference month will go through various edit checks.

The first set of edit checks is based on the Hidiriglou-Berthelot method whereby a ratio of the respondent’s current month data over historical (last month, same month last year) or auxiliary data is analyzed. When the respondent’s ratio differs significantly from ratios of respondents who are similar in terms of industry and/or geography group, the response is deemed an outlier.

The second set of edits consists of an edit known as the share of market edit. With this method, one is able to edit all respondents, even those where historical and auxiliary data is unavailable. The method relies on current month data only. Therefore, within a group of respondents, that are similar in terms of industrial group and/or geography, if the weighted contribution of a respondent to the group’s total is too large, it will be flagged as an outlier.

For edit checks based on the Hidiriglou-Berthelot method, data that are flagged as an outlier will not be included in the imputation models (those based on ratios). Also, data that are flagged as outliers in the share of market edit will not be included in the imputation models where means and medians are calculated to impute for responses that have no historical responses.

In conjunction with the statistical editing after data collection of reported data, there is also error detection done on the extracted GST data. Modeled data based on the GST are also subject to an extensive series of processing steps which thoroughly verify each record that is the basis for the model as well as the record being modeled. Edits are performed at a more aggregate level (industry by geography level) to detect records which deviate from the expected range, either by exhibiting large month-to-month change, or differing significantly from the remaining units. All data which fail these edits are subject to manual inspection and possible corrective action.

9. Imputation

Imputation in the MRTS is the process used to assign replacement values for missing data. This is done by assigning values when they are missing on the record being edited to ensure that estimates are of high quality and that a plausible, internal consistency is created. Due to concerns of response burden, cost and timeliness, it is generally impossible to do all follow-ups with the respondents in order to resolve missing responses. Since it is desirable to produce a complete and consistent microdata file, imputation is used to handle the remaining missing cases.

In the MRTS, imputation is based on historical data or administrative data (GST sales). The appropriate method is selected according to a strategy that is based on whether historical data is available, auxiliary data is available and/or which reference month is being processed.

There are three types of historical imputation methods. The first type is a general trend that uses one historical data source (previous month, data from next month or data from same month previous year). The second type is a regression model where data from previous month and same month, previous year are used simultaneously. The third type uses the historical data as a direct replacement value for a non-respondent. Depending upon the particular reference month, there is an order of preference that exists so that top quality imputation can result. The historical imputation method that was labelled as the third type above is always the last option in the order for each reference month.

The imputation method using administrative data is automatically selected when historical information is unavailable for a non-respondent. Trends are then applied to the administrative data source (monthly size) depending on whether the structure is simple, e.g. enterprises with only one establishment, or the unit has a more complex structure.

10. Estimation

Estimation is a process that approximates unknown population parameters using only part of the population that is included in a sample. Inferences about these unknown parameters are then made, using the sample data and associated survey design. This stage uses Statistics Canada's Generalized Estimation System (GES).

For retail sales, the population is divided into a survey portion (take-all and take-some strata) and a non-survey portion (take-none stratum). From the sample that is drawn from the survey portion, an estimate for the population is determined through the use of a Horvitz-Thompson estimator where responses for sales are weighted by using the inverses of the inclusion probabilities of the sampled units. Such weights (called sampling weights) can be interpreted as the number of times that each sampled unit should be replicated to represent the entire population. The calculated weighted sales values are summed by domain, to produce the total sales estimates by each industrial group / geographic area combination. A domain is defined as the most recent classification values available from the BR for the unit and the survey reference period. These domains may differ from the original sampling strata because units may have changed size, industry or location. Changes in classification are reflected immediately in the estimates and do not accumulate over time. For the non-survey portion, the sales are estimated with statistical models using monthly GST sales.

For more information on the methodology for modeling sales from administrative data sources which also contributes to the estimates of the survey portion, refer to ‘Monthly Retail Survey: Use of Administrative Data’ under ‘Documentation’ of the IMDB.

The measure of precision used for the MRTS to evaluate the quality of a population parameter estimate and to obtain valid inferences is the variance. The variance from the survey portion is derived directly from a stratified simple random sample without replacement.

Sample estimates may differ from the expected value of the estimates. However, since the estimate is based on a probability sample, the variability of the sample estimate with respect to its expected value can be measured. The variance of an estimate is a measure of the precision of the sample estimate and is defined as the average, over all possible samples, of the squared difference of the estimate from its expected value.

11. Revisions and seasonal adjustment

Revisions in the raw data are required to correct known non-sampling errors. These normally include replacing imputed data with reported data, corrections to previously reported data, and estimates for new births that were not known at the time of the original estimates.

Raw data are revised, on a monthly basis, for the month immediately prior to the current reference month being published. That is, when data for December are being published for the first time, there will also be revisions, if necessary, to the raw data for November. In addition, revisions are made once a year, with the initial release of the February data, for all months in the previous years. The purpose is to correct any significant problems that have been found that apply for an extended period. The actual period of revision depends on the nature of the problem identified, but rarely exceeds three years. The revision period can be extended when historical revisions or restratitfication are done.

Retail trade data are seasonally adjusted using the X12-ARIMA method. This consists of extrapolating a year's worth of raw data with the ARIMA model (auto-regressive integrated moving average model), and of seasonally adjusting the raw time series. Finally, the annual totals of the seasonally adjusted series are forced to the annual totals of the original series.

The seasonally adjusted data also need to be revised. In part, they need to reflect the revisions identified for the raw data. Also, the seasonally adjusted estimates are calculated using X-12-ARIMA, and are sensitive to the most recent values reported in the raw data. For this reason, with the release of each month of new data, the seasonally adjusted values for the previous three months are revised.  A seasonally adjusted time series is a time series that has been modified to eliminate the effect of seasonal and calendar influences. For this reason, the seasonally adjusted data allows for more meaningful comparisons of economic conditions from month to month.

Once a year, seasonal adjustments options are reviewed to take into account the most recent data. Revised seasonally adjusted estimates for each month in the previous years are released at the same time as the annual revision to the raw data. The actual period of revision depends on the number years the raw data was revised.

12. Data quality evaluation

The methodology of this survey has been designed to control errors and to reduce their potential effects on estimates. However, the survey results remain subject to errors, of which sampling error is only one component of the total survey error. Sampling error results when observations are made only on a sample and not on the entire population. All other errors arising from the various phases of a survey are referred to as nonsampling errors. For example, these types of errors can occur when a respondent provides incorrect information or does not answer certain questions; when a unit in the target population is omitted or covered more than once; when GST data for records being modeled for a particular month are not representative of the actual record for various reasons; when a unit that is out of scope for the survey is included by mistake or when errors occur in data processing, such as coding or capture errors.

Prior to publication, combined survey results are analyzed for comparability; in general, this includes a detailed review of individual responses (especially for large businesses), general economic conditions and historical trends.

A common measure of data quality for surveys is the coefficient of variation (CV). The coefficient of variation, defined as the standard error divided by the sample estimate, is a measure of precision in relative terms. Since the coefficient of variation is calculated from responses of individual units, it also measures some non-sampling errors.

The formula used to calculate coefficients of variation (CV) as percentages is:

CV (X) = S(X) * 100% / X
where X denotes the estimate and S(X) denotes the standard error of X.

Confidence intervals can be constructed around the estimates using the estimate and the CV. Thus, for our sample, it is possible to state with a given level of confidence that the expected value will fall within the confidence interval constructed around the estimate. For example, if an estimate of $12,000,000 has a CV of 2%, the standard error will be $240,000 (the estimate multiplied by the CV). It can be stated with 68% confidence that the expected values will fall within the interval whose length equals the standard deviation about the estimate, i.e. between $11,760,000 and $12,240,000.

Alternatively, it can be stated with 95% confidence that the expected value will fall within the interval whose length equals two standard deviations about the estimate, i.e. between $11,520,000 and $12,480,000.

Finally, due to the small contribution of the non-survey portion to the total estimates, bias in the non-survey portion has a negligible impact on the CVs. Therefore, the CV from the survey portion is used for the total estimate that is the summation of estimates from the surveyed and non-surveyed portions.

13. 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.

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.

Monthly Retail Trade Survey (MRTS) Data Quality Statement

Objectives, uses and users
Concepts, variables and classifications
Coverage and frames
Sampling
Questionnaire design
Response and non-response
Data collection and capture operations
Editing
Imputation
Estimation
Revisions and seasonal adjustment
Data quality evaluation
Disclosure control

1. Objectives, uses and users

1.1. Objective

The Monthly Retail Trade Survey (MRTS) provides information on the performance of the retail trade sector on a monthly basis, and when combined with other statistics, represents an important indicator of the state of the Canadian economy.

1.2. Uses

The estimates provide a measure of the health and performance of the retail trade sector. Information collected is used to estimate level and monthly trend for retail sales. At the end of each year, the estimates provide a preliminary look at annual retail sales and performance.

1.3. Users

A variety of organizations, sector associations, and levels of government make use of the information. Retailers rely on the survey results to compare their performance against similar types of businesses, as well as for marketing purposes. Retail associations are able to monitor industry performance and promote their retail industries. Investors can monitor industry growth, which can result in better access to investment capital by retailers. Governments are able to understand the role of retailers in the economy, which aids in the development of policies and tax incentives. As an important industry in the Canadian economy, governments are able to better determine the overall health of the economy through the use of the estimates in the calculation of the nation’s Gross Domestic Product (GDP).

2. Concepts, variables and classifications

2.1. Concepts

The retail trade sector comprises establishments primarily engaged in retailing merchandise, generally without transformation, and rendering services incidental to the sale of merchandise.

The retailing process is the final step in the distribution of merchandise; retailers are therefore organized to sell merchandise in small quantities to the general public. This sector comprises two main types of retailers, that is, store and non-store retailers. The MRTS covers only store retailers. Their main characteristics are described below. Store retailers operate fixed point-of-sale locations, located and designed to attract a high volume of walk-in customers. In general, retail stores have extensive displays of merchandise and use mass-media advertising to attract customers. They typically sell merchandise to the general public for personal or household consumption, but some also serve business and institutional clients. These include establishments such as office supplies stores, computer and software stores, gasoline stations, building material dealers, plumbing supplies stores and electrical supplies stores.

In addition to selling merchandise, some types of store retailers are also engaged in the provision of after-sales services, such as repair and installation. For example, new automobile dealers, electronic and appliance stores and musical instrument and supplies stores often provide repair services, while floor covering stores and window treatment stores often provide installation services. As a general rule, establishments engaged in retailing merchandise and providing after sales services are classified in this sector. Catalogue sales showrooms, gasoline service stations, and mobile home dealers are treated as store retailers.

2.2. Variables

Sales are defined as the sales of all goods purchased for resale, net of returns and discounts. This includes commission revenue and fees earned from selling goods and services on account of others, such as selling lottery tickets, bus tickets, and phone cards. It also includes parts and labour revenue from repair and maintenance; revenue from rental and leasing of goods and equipment; revenues from services, including food services; sales of goods manufactured as a secondary activity; and the proprietor’s withdrawals, at retail, of goods for personal use. Other revenue from rental of real estate, placement fees, operating subsidies, grants, royalties and franchise fees are excluded.

Trading Location is the physical location(s) in which business activity is conducted in each province and territory, and for which sales are credited or recognized in the financial records of the company. For retailers, this would normally be a store.

Constant Dollars: The value of retail trade is measured in two ways; including the effects of price change on sales and net of the effects of price change. The first measure is referred to as retail trade in current dollars and the latter as retail trade in constant dollars. The method of calculating the current dollar estimate is to aggregate the weighted value of sales for all retail outlets. The method of calculating the constant dollar estimate is to first adjust the sales values to a base year, using the Consumer Price Index, and then sum up the resulting values.

2.3. Classification

The Monthly Retail Trade Survey is based on the definition of retail trade under the NAICS (North American Industry Classification System). NAICS is the agreed upon common framework for the production of comparable statistics by the statistical agencies of Canada, Mexico and the United States. The agreement defines the boundaries of twenty sectors. NAICS is based on a production-oriented, or supply based conceptual framework in that establishments are groups into industries according to similarity in production processes used to produce goods and services.

Estimates appear for 21 industries based on special aggregations of the 2012 North American Industry Classification System (NAICS) industries. The 21 industries are further aggregated to 11 sub-sectors.

Geographically, sales estimates are produced for Canada and each province and territory.

3. Coverage and frames

Statistics Canada’s Business Register ( BR) provides the frame for the Monthly Retail Trade Survey. The BR is a structured list of businesses engaged in the production of goods and services in Canada. It is a centrally maintained database containing detailed descriptions of most business entities operating within Canada. The BR includes all incorporated businesses, with or without employees. For unincorporated businesses, the BR includes all employers with businesses, and businesses with no employees with annual sales that have a Goods and Services Tax (GST) or annual revenue that declares individual taxes.  annual sales greater than $30,000 that have a Goods and Services Tax (GST) account (the BR does not include unincorporated businesses with no employees and with annual sales less than $30,000).

The businesses on the BR are represented by a hierarchical structure with four levels, with the statistical enterprise at the top, followed by the statistical company, the statistical establishment and the statistical location. An enterprise can be linked to one or more statistical companies, a statistical company can be linked to one or more statistical establishments, and a statistical establishment to one or more statistical locations.

The target population for the MRTS consists of all statistical establishments on the BR that are classified to the retail sector using the North American Industry Classification System (NAICS) (approximately 200,000 establishments). The NAICS code range for the retail sector is 441100 to 453999. A statistical establishment is the production entity or the smallest grouping of production entities which: produces a homogeneous set of goods or services; does not cross provincial boundaries; and provides data on the value of output, together with the cost of principal intermediate inputs used, along with the cost and quantity of labour used to produce the output. The production entity is the physical unit where the business operations are carried out. It must have a civic address and dedicated labour.

The exclusions to the target population are ancillary establishments (producers of services in support of the activity of producing goods and services for the market of more than one establishment within the enterprise, and serves as a cost centre or a discretionary expense centre for which data on all its costs including labour and depreciation can be reported by the business), future establishments, establishments with a missing or a zero gross business income (GBI) value on the BR and establishments in the following non-covered NAICS:

  • 4541 (electronic shopping and mail-order houses)
  • 4542 (vending machine operators)
  • 45431 (fuel dealers)
  • 45439 (other direct selling establishments)

4. Sampling

The MRTS sample consists of 10,000 groups of establishments (clusters) classified to the Retail Trade sector selected from the Statistics Canada Business Register. A cluster of establishments is defined as all establishments belonging to a statistical enterprise that are in the same industrial group and geographical region. The MRTS uses a stratified design with simple random sample selection in each stratum. The stratification is done by industry groups (the mainly, but not only four digit level NAICS), and the geographical regions consisting of the provinces and territories, as well as three provincial sub-regions. We further stratify the population by size.

The size measure is created using a combination of independent survey data and three administrative variables: the annual profiled revenue, the GST sales expressed on an annual basis, and the declared tax revenue (T1 or T2). The size strata consist of one take-all (census), at most, two take-some (partially sampled) strata, and one take-none (non-sampled) stratum. Take-none strata serve to reduce respondent burden by excluding the smaller businesses from the surveyed population. These businesses should represent at most ten percent of total sales. Instead of sending questionnaires to these businesses, the estimates are produced through the use of administrative data.

The sample was allocated optimally in order to reach target coefficients of variation at the national, provincial/territorial, industrial, and industrial groups by province/territory levels. The sample was also inflated to compensate for dead, non-responding, and misclassified units.

MRTS is a repeated survey with maximisation of monthly sample overlap. The sample is kept month after month, and every month new units are added (births) to the sample.  MRTS births, i.e., new clusters of establishment(s), are identified every month via the BR’s latest universe. They are stratified according to the same criteria as the initial population. A sample of these births is selected according to the sampling fraction of the stratum to which they belong and is added to the monthly sample. Deaths occur on a monthly basis. A death can be a cluster of establishment(s) that have ceased their activities (out-of-business) or whose major activities are no longer in retail trade (out-of-scope). The status of these businesses is updated on the BR using administrative sources and survey feedback, including feedback from the MRTS. Methods to treat dead units and misclassified units are part of the sample and population update procedures.

5. Questionnaire design

The Monthly Retail Trade Survey incorporates the following sub-surveys:

Monthly Retail Trade Survey - R8

Monthly Retail Trade Survey (with inventories) – R8

Survey of Sales and Inventories of Alcoholic Beverages

The questionnaires collect monthly data on retail sales and the number of trading locations by province or territory and inventories of goods owned and intended for resale from a sample of retailers. The items on the questionnaires have remained unchanged for several years. For the 2004 redesign, the general questionnaires were subject to cosmetic changes only. The questionnaire for Sales and Inventories of Alcoholic Beverages underwent more extensive changes. The modifications were discussed withstakeholders and the respondents were given an opportunity to comment before the new questionnaire was finalized. If further changes are needed to any of the questionnaires, proposed changes would go through a review committee and a field test with respondents and data users to ensure its relevancy.

6. Response and non-response

6.1. Response and non-response

Despite the best efforts of survey managers and operations staff to maximize response in the MRTS, some non-response will occur. For statistical establishments to be classified as responding, the degree of partial response (where an accurate response is obtained for only some of the questions asked a respondent) must meet a minimum threshold level below which the response would be rejected and considered a unit non-response.  In such an instance, the business is classified as not having responded at all.

Non-response has two effects on data: first it introduces bias in estimates when non-respondents differ from respondents in the characteristics measured; and second, it contributes to an increase in the sampling variance of estimates because the effective sample size is reduced from that originally sought.

The degree to which efforts are made to get a response from a non-respondent is based on budget and time constraints, its impact on the overall quality and the risk of non-response bias.

The main method to reduce the impact of non-response at sampling is to inflate the sample size through the use of over-sampling rates that have been determined from similar surveys.

Besides the methods to reduce the impact of non-response at sampling and collection, the non-responses to the survey that do occur are treated through imputation. In order to measure the amount of non-response that occurs each month, various response rates are calculated. For a given reference month, the estimation process is run at least twice (a preliminary and a revised run). Between each run, respondent data can be identified as unusable and imputed values can be corrected through respondent data. As a consequence, response rates are computed following each run of the estimation process.

For the MRTS, two types of rates are calculated (un-weighted and weighted). In order to assess the efficiency of the collection process, un-weighted response rates are calculated. Weighted rates, using the estimation weight and the value for the variable of interest, assess the quality of estimation. Within each of these types of rates, there are distinct rates for units that are surveyed and for units that are only modeled from administrative data that has been extracted from GST files.

To get a better picture of the success of the collection process, two un-weighted rates called the ‘collection results rate’ and the ‘extraction results rate’ are computed. They are computed by dividing the number of respondents by the number of units that we tried to contact or tried to receive extracted data for them. Non-monthly reporters (respondents with special reporting arrangements where they do not report every month but for whom actual data is available in subsequent revisions) are excluded from both the numerator and denominator for the months where no contact is performed.

In summary, the various response rates are calculated as follows:

Weighted rates:

Survey Response rate (estimation) =
Sum of weighted sales of units with response status i / Sum of survey weighted sales

where i = units that have either reported data that will be used in estimation or are converted refusals, or have reported data that has not yet been resolved for estimation.

Admin Response rate (estimation) =
Sum of weighted sales of units with response status ii / Sum of administrative weighted sales

where ii = units that have data that was extracted from administrative files and are usable for estimation.

Total Response rate (estimation) =
Sum of weighted sales of units with response status i or response status ii / Sum of all weighted sales

Un-weighted rates:

Survey Response rate (collection) =
Number of questionnaires with response status iii/ Number of questionnaires with response status iv

where iii = units that have either reported data (unresolved, used or not used for estimation) or are converted refusals.

where iv = all of the above plus units that have refused to respond, units that were not contacted and other types of non-respondent units.

Admin Response rate (extraction) =
Number of questionnaires with response status vi/ Number of questionnaires with response status vii

where vi = in-scope units that have data (either usable or non-usable) that was extracted from administrative files

where vii = all of the above plus units that have refused to report to the administrative data source, units that were not contacted and other types of non-respondent units.

(% of questionnaire collected over all in-scope questionnaires)

Collection Results Rate =
Number of questionnaires with response status iii / Number of questionnaires with response status viii

where iii = same as iii defined above

where viii = same as iv except for the exclusion of units that were contacted because their response is unavailable for a particular month since they are non-monthly reporters.

Extraction Results Rate =
Number of questionnaires with response status ix / Number of questionnaires with response status vii

where ix = same as vi with the addition of extracted units that have been imputed or were out of scope

where vii = same as vii defined above

(% of questionnaires collected over all questionnaire in-scope we tried to collect)

All the above weighted and un-weighted rates are provided at the industrial group, geography and size group level or for any combination of these levels.

Use of Administrative Data

Managing response burden is an ongoing challenge for Statistics Canada. In an attempt to alleviate response burden and survey costs, especially for smaller businesses, the MRTS has reduced the number of simple establishments in the sample that are surveyed directly and instead derives sales data for these establishments from Goods and Service Tax (GST) files using a statistical model. The model accounts for differences between sales and revenue (reported for GST purposes) as well as for the time lag between the survey reference period and the reference period of the GST file.

For more information on the methodology used for modeling sales from administrative data sources, refer to ‘Monthly Retail Trade Survey: Use of Administrative Data’ under ‘Documentation’ of the IMDB.

Table 1 contains the weighted response rates for all industry groups as well as for total retail trade for each province and territory. For more detailed weighted response rates, please contact the Marketing and Dissemination Section at (613) 951-3549, toll free: 1-877-421-3067 or by e-mail at retailinfo@statcan.

6.2. Methods used to reduce non-response at collection

Significant effort is spent trying to minimize non-response during collection. Methods used, among others, are interviewer techniques such as probing and persuasion, repeated re-scheduling and call-backs to obtain the information, and procedures dealing with how to handle non-compliant (refusal) respondents.

If data are unavailable at the time of collection, a respondent's best estimates are also accepted, and are subsequently revised once the actual data become available.

To minimize total non-response for all variables, partial responses are accepted. In addition, questionnaires are customized for the collection of certain variables, such as inventory, so that collection is timed for those months when the data are available.

Finally, to build trust and rapport between the interviewers and respondents, cases are generally assigned to the same interviewer each month. This action establishes a personal relationship between interviewer and respondent, and builds respondent trust.

7. Data collection and capture operations

Collection of the data is performed by Statistics Canada’s Regional Offices.

Table 1: Weighted response rates by NAICS, for all provinces and territories: April 2016
Table summary
This table displays the results of Table 1: Weighted response rates by NAICS Weighted Response Rates, calculated using Total, Survey and Administrative units of measure (appearing as column headers).
  Weighted Response Rates
Total Survey Administrative
NAICS - Canada  
Motor Vehicle and Parts Dealers 91.1 91.5 62.9
Automobile Dealers 92.9 92.9 90.6
New Car Dealers 94.5 94.5 Note ...: not applicable
Used Car Dealers 67.4 64.6 90.6
Other Motor Vehicle Dealers 70.4 73.0 37.3
Automotive Parts, Accessories and Tire Stores 81.9 85.6 53.7
Furniture and Home Furnishings Stores 81.0 83.4 56.7
Furniture Stores 80.0 80.6 67.9
Home Furnishings Stores 82.9 89.4 50.2
Electronics and Appliance Stores 81.4 82.7 5.1
Building Material and Garden Equipment Dealers 90.2 90.4 86.3
Food and Beverage Stores 84.2 86.1 57.9
Grocery Stores 89.7 92.0 58.9
Grocery (except Convenience) Stores 90.9 93.2 58.8
Convenience Stores 73.1 75.1 59.5
Specialty Food Stores 58.7 61.1 44.9
Beer, Wine and Liquor Stores 69.4 69.4 67.2
Health and Personal Care Stores 86.7 87.4 77.2
Gasoline Stations 77.2 77.8 68.7
Clothing and Clothing Accessories Stores 87.1 88.2 36.6
Clothing Stores 88.0 89.1 27.5
Shoe Stores 82.4 82.4 80.3
Jewellery, Luggage and Leather Goods Stores 86.3 88.7 47.0
Sporting Goods, Hobby, Book and Music Stores 86.4 91.9 25.0
General Merchandise Stores 98.2 98.3 89.5
Department Stores 100.0 100.0 Note ...: not applicable
Other general merchandise stores 97.1 97.2 89.5
Miscellaneous Store Retailers 76.8 79.6 52.9
Total 87.9 89.0 62.7
Regions  
Newfoundland and Labrador 81.9 82.1 74.8
Prince Edward Island 81.9 82.7 23.4
Nova Scotia 91.8 92.2 79.2
New Brunswick 89.4 90.6 62.4
Québec 89.1 90.5 64.6
Ontario 89.2 90.7 53.9
Manitoba 79.8 80.2 57.4
Saskatchewan 86.7 86.9 82.1
Alberta 86.8 87.6 68.8
British Columbia 86.8 87.4 71.4
Yukon Territory 72.8 72.8 Note ...: not applicable
Northwest Territories 52.2 52.2 Note ...: not applicable
Nunavut 35.2 35.2 Note ...: not applicable


Weighted Response Rates

Respondents are sent a questionnaire or are contacted by telephone to obtain their sales and inventory values, as well as to confirm the opening or closing of business trading locations. Collection of the data begins approximately 7 working days after the end of the reference month and continues for the duration of that month.

New entrants to the survey are introduced to the survey via an introductory letter that informs the respondent that a representative of Statistics Canada will be calling. This call is to introduce the respondent to the survey, confirm the respondent's business activity, establish and begin data collection, as well as to answer any questions that the respondent may have.

8. Editing

Data editing is the application of checks to detect missing, invalid or inconsistent entries or to point to data records that are potentially in error. In the survey process for the MRTS, data editing is done at two different time periods.

First of all, editing is done during data collection. Once data are collected via the telephone, or via the receipt of completed mail-in questionnaires, the data are captured using customized data capture applications. All data are subjected to data editing. Edits during data collection are referred to as field edits and generally consist of validity and some simple consistency edits. They are used to detect mistakes made during the interview by the respondent or the interviewer and to identify missing information during collection in order to reduce the need for follow-up later on. Another purpose of the field edits is to clean up responses. In the MRTS, the current month’s responses are edited against the respondent’s previous month’s responses and/or the previous year’s responses for the current month. Field edits are also used to identify problems with data collection procedures and the design of the questionnaire, as well as the need for more interviewer training.

Follow-up with respondents occurs to validate potential erroneous data following any failed preliminary edit check of the data. Once validated, the collected data is regularly transmitted to the head office in Ottawa.

Secondly, editing known as statistical editing is also done after data collection and this is more empirical in nature. Statistical editing is run prior to imputation in order to identify the data that will be used as a basis to impute non-respondents. Large outliers that could disrupt a monthly trend are excluded from trend calculations by the statistical edits. It should be noted that adjustments are not made at this stage to correct the reported outliers.

The first step in the statistical editing is to identify which responses will be subjected to the statistical edit rules. Reported data for the current reference month will go through various edit checks.

The first set of edit checks is based on the Hidiriglou-Berthelot method whereby a ratio of the respondent’s current month data over historical (last month, same month last year) or auxiliary data is analyzed. When the respondent’s ratio differs significantly from ratios of respondents who are similar in terms of industry and/or geography group, the response is deemed an outlier.

The second set of edits consists of an edit known as the share of market edit. With this method, one is able to edit all respondents, even those where historical and auxiliary data is unavailable. The method relies on current month data only. Therefore, within a group of respondents, that are similar in terms of industrial group and/or geography, if the weighted contribution of a respondent to the group’s total is too large, it will be flagged as an outlier.

For edit checks based on the Hidiriglou-Berthelot method, data that are flagged as an outlier will not be included in the imputation models (those based on ratios). Also, data that are flagged as outliers in the share of market edit will not be included in the imputation models where means and medians are calculated to impute for responses that have no historical responses.

In conjunction with the statistical editing after data collection of reported data, there is also error detection done on the extracted GST data. Modeled data based on the GST are also subject to an extensive series of processing steps which thoroughly verify each record that is the basis for the model as well as the record being modeled. Edits are performed at a more aggregate level (industry by geography level) to detect records which deviate from the expected range, either by exhibiting large month-to-month change, or differing significantly from the remaining units. All data which fail these edits are subject to manual inspection and possible corrective action.

9. Imputation

Imputation in the MRTS is the process used to assign replacement values for missing data. This is done by assigning values when they are missing on the record being edited to ensure that estimates are of high quality and that a plausible, internal consistency is created. Due to concerns of response burden, cost and timeliness, it is generally impossible to do all follow-ups with the respondents in order to resolve missing responses. Since it is desirable to produce a complete and consistent microdata file, imputation is used to handle the remaining missing cases.

In the MRTS, imputation is based on historical data or administrative data (GST sales). The appropriate method is selected according to a strategy that is based on whether historical data is available, auxiliary data is available and/or which reference month is being processed.

There are three types of historical imputation methods. The first type is a general trend that uses one historical data source (previous month, data from next month or data from same month previous year). The second type is a regression model where data from previous month and same month, previous year are used simultaneously. The third type uses the historical data as a direct replacement value for a non-respondent. Depending upon the particular reference month, there is an order of preference that exists so that top quality imputation can result. The historical imputation method that was labelled as the third type above is always the last option in the order for each reference month.

The imputation method using administrative data is automatically selected when historical information is unavailable for a non-respondent. Trends are then applied to the administrative data source (monthly size) depending on whether the structure is simple, e.g. enterprises with only one establishment, or the unit has a more complex structure.

10. Estimation

Estimation is a process that approximates unknown population parameters using only part of the population that is included in a sample. Inferences about these unknown parameters are then made, using the sample data and associated survey design. This stage uses Statistics Canada's Generalized Estimation System (GES).

For retail sales, the population is divided into a survey portion (take-all and take-some strata) and a non-survey portion (take-none stratum). From the sample that is drawn from the survey portion, an estimate for the population is determined through the use of a Horvitz-Thompson estimator where responses for sales are weighted by using the inverses of the inclusion probabilities of the sampled units. Such weights (called sampling weights) can be interpreted as the number of times that each sampled unit should be replicated to represent the entire population. The calculated weighted sales values are summed by domain, to produce the total sales estimates by each industrial group / geographic area combination. A domain is defined as the most recent classification values available from the BR for the unit and the survey reference period. These domains may differ from the original sampling strata because units may have changed size, industry or location. Changes in classification are reflected immediately in the estimates and do not accumulate over time. For the non-survey portion, the sales are estimated with statistical models using monthly GST sales.

For more information on the methodology for modeling sales from administrative data sources which also contributes to the estimates of the survey portion, refer to ‘Monthly Retail Survey: Use of Administrative Data’ under ‘Documentation’ of the IMDB.

The measure of precision used for the MRTS to evaluate the quality of a population parameter estimate and to obtain valid inferences is the variance. The variance from the survey portion is derived directly from a stratified simple random sample without replacement.

Sample estimates may differ from the expected value of the estimates. However, since the estimate is based on a probability sample, the variability of the sample estimate with respect to its expected value can be measured. The variance of an estimate is a measure of the precision of the sample estimate and is defined as the average, over all possible samples, of the squared difference of the estimate from its expected value.

11. Revisions and seasonal adjustment

Revisions in the raw data are required to correct known non-sampling errors. These normally include replacing imputed data with reported data, corrections to previously reported data, and estimates for new births that were not known at the time of the original estimates.

Raw data are revised, on a monthly basis, for the month immediately prior to the current reference month being published. That is, when data for December are being published for the first time, there will also be revisions, if necessary, to the raw data for November. In addition, revisions are made once a year, with the initial release of the February data, for all months in the previous years. The purpose is to correct any significant problems that have been found that apply for an extended period. The actual period of revision depends on the nature of the problem identified, but rarely exceeds three years. The revision period can be extended when historical revisions or restratitfication are done.

Retail trade data are seasonally adjusted using the X12-ARIMA method. This consists of extrapolating a year's worth of raw data with the ARIMA model (auto-regressive integrated moving average model), and of seasonally adjusting the raw time series. Finally, the annual totals of the seasonally adjusted series are forced to the annual totals of the original series.

The seasonally adjusted data also need to be revised. In part, they need to reflect the revisions identified for the raw data. Also, the seasonally adjusted estimates are calculated using X-12-ARIMA, and are sensitive to the most recent values reported in the raw data. For this reason, with the release of each month of new data, the seasonally adjusted values for the previous three months are revised.  A seasonally adjusted time series is a time series that has been modified to eliminate the effect of seasonal and calendar influences. For this reason, the seasonally adjusted data allows for more meaningful comparisons of economic conditions from month to month.

Once a year, seasonal adjustments options are reviewed to take into account the most recent data. Revised seasonally adjusted estimates for each month in the previous years are released at the same time as the annual revision to the raw data. The actual period of revision depends on the number years the raw data was revised.

12. Data quality evaluation

The methodology of this survey has been designed to control errors and to reduce their potential effects on estimates. However, the survey results remain subject to errors, of which sampling error is only one component of the total survey error. Sampling error results when observations are made only on a sample and not on the entire population. All other errors arising from the various phases of a survey are referred to as nonsampling errors. For example, these types of errors can occur when a respondent provides incorrect information or does not answer certain questions; when a unit in the target population is omitted or covered more than once; when GST data for records being modeled for a particular month are not representative of the actual record for various reasons; when a unit that is out of scope for the survey is included by mistake or when errors occur in data processing, such as coding or capture errors.

Prior to publication, combined survey results are analyzed for comparability; in general, this includes a detailed review of individual responses (especially for large businesses), general economic conditions and historical trends.

A common measure of data quality for surveys is the coefficient of variation (CV). The coefficient of variation, defined as the standard error divided by the sample estimate, is a measure of precision in relative terms. Since the coefficient of variation is calculated from responses of individual units, it also measures some non-sampling errors.

The formula used to calculate coefficients of variation (CV) as percentages is:

CV (X) = S(X) * 100% / X
where X denotes the estimate and S(X) denotes the standard error of X.

Confidence intervals can be constructed around the estimates using the estimate and the CV. Thus, for our sample, it is possible to state with a given level of confidence that the expected value will fall within the confidence interval constructed around the estimate. For example, if an estimate of $12,000,000 has a CV of 2%, the standard error will be $240,000 (the estimate multiplied by the CV). It can be stated with 68% confidence that the expected values will fall within the interval whose length equals the standard deviation about the estimate, i.e. between $11,760,000 and $12,240,000.

Alternatively, it can be stated with 95% confidence that the expected value will fall within the interval whose length equals two standard deviations about the estimate, i.e. between $11,520,000 and $12,480,000.

Finally, due to the small contribution of the non-survey portion to the total estimates, bias in the non-survey portion has a negligible impact on the CVs. Therefore, the CV from the survey portion is used for the total estimate that is the summation of estimates from the surveyed and non-surveyed portions.

13. 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.

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.

Monthly Retail Trade Survey (MRTS) Data Quality Statement

Objectives, uses and users
Concepts, variables and classifications
Coverage and frames
Sampling
Questionnaire design
Response and non-response
Data collection and capture operations
Editing
Imputation
Estimation
Revisions and seasonal adjustment
Data quality evaluation
Disclosure control

1. Objectives, uses and users

1.1. Objective

The Monthly Retail Trade Survey (MRTS) provides information on the performance of the retail trade sector on a monthly basis, and when combined with other statistics, represents an important indicator of the state of the Canadian economy.

1.2. Uses

The estimates provide a measure of the health and performance of the retail trade sector. Information collected is used to estimate level and monthly trend for retail sales. At the end of each year, the estimates provide a preliminary look at annual retail sales and performance.

1.3. Users

A variety of organizations, sector associations, and levels of government make use of the information. Retailers rely on the survey results to compare their performance against similar types of businesses, as well as for marketing purposes. Retail associations are able to monitor industry performance and promote their retail industries. Investors can monitor industry growth, which can result in better access to investment capital by retailers. Governments are able to understand the role of retailers in the economy, which aids in the development of policies and tax incentives. As an important industry in the Canadian economy, governments are able to better determine the overall health of the economy through the use of the estimates in the calculation of the nation’s Gross Domestic Product (GDP).

2. Concepts, variables and classifications

2.1. Concepts

The retail trade sector comprises establishments primarily engaged in retailing merchandise, generally without transformation, and rendering services incidental to the sale of merchandise.

The retailing process is the final step in the distribution of merchandise; retailers are therefore organized to sell merchandise in small quantities to the general public. This sector comprises two main types of retailers, that is, store and non-store retailers. The MRTS covers only store retailers. Their main characteristics are described below. Store retailers operate fixed point-of-sale locations, located and designed to attract a high volume of walk-in customers. In general, retail stores have extensive displays of merchandise and use mass-media advertising to attract customers. They typically sell merchandise to the general public for personal or household consumption, but some also serve business and institutional clients. These include establishments such as office supplies stores, computer and software stores, gasoline stations, building material dealers, plumbing supplies stores and electrical supplies stores.

In addition to selling merchandise, some types of store retailers are also engaged in the provision of after-sales services, such as repair and installation. For example, new automobile dealers, electronic and appliance stores and musical instrument and supplies stores often provide repair services, while floor covering stores and window treatment stores often provide installation services. As a general rule, establishments engaged in retailing merchandise and providing after sales services are classified in this sector. Catalogue sales showrooms, gasoline service stations, and mobile home dealers are treated as store retailers.

2.2. Variables

Sales are defined as the sales of all goods purchased for resale, net of returns and discounts. This includes commission revenue and fees earned from selling goods and services on account of others, such as selling lottery tickets, bus tickets, and phone cards. It also includes parts and labour revenue from repair and maintenance; revenue from rental and leasing of goods and equipment; revenues from services, including food services; sales of goods manufactured as a secondary activity; and the proprietor’s withdrawals, at retail, of goods for personal use. Other revenue from rental of real estate, placement fees, operating subsidies, grants, royalties and franchise fees are excluded.

Trading Location is the physical location(s) in which business activity is conducted in each province and territory, and for which sales are credited or recognized in the financial records of the company. For retailers, this would normally be a store.

Constant Dollars: The value of retail trade is measured in two ways; including the effects of price change on sales and net of the effects of price change. The first measure is referred to as retail trade in current dollars and the latter as retail trade in constant dollars. The method of calculating the current dollar estimate is to aggregate the weighted value of sales for all retail outlets. The method of calculating the constant dollar estimate is to first adjust the sales values to a base year, using the Consumer Price Index, and then sum up the resulting values.

2.3. Classification

The Monthly Retail Trade Survey is based on the definition of retail trade under the NAICS (North American Industry Classification System). NAICS is the agreed upon common framework for the production of comparable statistics by the statistical agencies of Canada, Mexico and the United States. The agreement defines the boundaries of twenty sectors. NAICS is based on a production-oriented, or supply based conceptual framework in that establishments are groups into industries according to similarity in production processes used to produce goods and services.

Estimates appear for 21 industries based on special aggregations of the 2012 North American Industry Classification System (NAICS) industries. The 21 industries are further aggregated to 11 sub-sectors.

Geographically, sales estimates are produced for Canada and each province and territory.

3. Coverage and frames

Statistics Canada’s Business Register ( BR) provides the frame for the Monthly Retail Trade Survey. The BR is a structured list of businesses engaged in the production of goods and services in Canada. It is a centrally maintained database containing detailed descriptions of most business entities operating within Canada. The BR includes all incorporated businesses, with or without employees. For unincorporated businesses, the BR includes all employers with businesses, and businesses with no employees with annual sales that have a Goods and Services Tax (GST) or annual revenue that declares individual taxes.  annual sales greater than $30,000 that have a Goods and Services Tax (GST) account (the BR does not include unincorporated businesses with no employees and with annual sales less than $30,000).

The businesses on the BR are represented by a hierarchical structure with four levels, with the statistical enterprise at the top, followed by the statistical company, the statistical establishment and the statistical location. An enterprise can be linked to one or more statistical companies, a statistical company can be linked to one or more statistical establishments, and a statistical establishment to one or more statistical locations.

The target population for the MRTS consists of all statistical establishments on the BR that are classified to the retail sector using the North American Industry Classification System (NAICS) (approximately 200,000 establishments). The NAICS code range for the retail sector is 441100 to 453999. A statistical establishment is the production entity or the smallest grouping of production entities which: produces a homogeneous set of goods or services; does not cross provincial boundaries; and provides data on the value of output, together with the cost of principal intermediate inputs used, along with the cost and quantity of labour used to produce the output. The production entity is the physical unit where the business operations are carried out. It must have a civic address and dedicated labour.

The exclusions to the target population are ancillary establishments (producers of services in support of the activity of producing goods and services for the market of more than one establishment within the enterprise, and serves as a cost centre or a discretionary expense centre for which data on all its costs including labour and depreciation can be reported by the business), future establishments, establishments with a missing or a zero gross business income (GBI) value on the BR and establishments in the following non-covered NAICS:

  • 4541 (electronic shopping and mail-order houses)
  • 4542 (vending machine operators)
  • 45431 (fuel dealers)
  • 45439 (other direct selling establishments)

4. Sampling

The MRTS sample consists of 10,000 groups of establishments (clusters) classified to the Retail Trade sector selected from the Statistics Canada Business Register. A cluster of establishments is defined as all establishments belonging to a statistical enterprise that are in the same industrial group and geographical region. The MRTS uses a stratified design with simple random sample selection in each stratum. The stratification is done by industry groups (the mainly, but not only four digit level NAICS), and the geographical regions consisting of the provinces and territories, as well as three provincial sub-regions. We further stratify the population by size.

The size measure is created using a combination of independent survey data and three administrative variables: the annual profiled revenue, the GST sales expressed on an annual basis, and the declared tax revenue (T1 or T2). The size strata consist of one take-all (census), at most, two take-some (partially sampled) strata, and one take-none (non-sampled) stratum. Take-none strata serve to reduce respondent burden by excluding the smaller businesses from the surveyed population. These businesses should represent at most ten percent of total sales. Instead of sending questionnaires to these businesses, the estimates are produced through the use of administrative data.

The sample was allocated optimally in order to reach target coefficients of variation at the national, provincial/territorial, industrial, and industrial groups by province/territory levels. The sample was also inflated to compensate for dead, non-responding, and misclassified units.

MRTS is a repeated survey with maximisation of monthly sample overlap. The sample is kept month after month, and every month new units are added (births) to the sample.  MRTS births, i.e., new clusters of establishment(s), are identified every month via the BR’s latest universe. They are stratified according to the same criteria as the initial population. A sample of these births is selected according to the sampling fraction of the stratum to which they belong and is added to the monthly sample. Deaths occur on a monthly basis. A death can be a cluster of establishment(s) that have ceased their activities (out-of-business) or whose major activities are no longer in retail trade (out-of-scope). The status of these businesses is updated on the BR using administrative sources and survey feedback, including feedback from the MRTS. Methods to treat dead units and misclassified units are part of the sample and population update procedures.

5. Questionnaire design

The Monthly Retail Trade Survey incorporates the following sub-surveys:

Monthly Retail Trade Survey - R8

Monthly Retail Trade Survey (with inventories) – R8

Survey of Sales and Inventories of Alcoholic Beverages

The questionnaires collect monthly data on retail sales and the number of trading locations by province or territory and inventories of goods owned and intended for resale from a sample of retailers. The items on the questionnaires have remained unchanged for several years. For the 2004 redesign, the general questionnaires were subject to cosmetic changes only. The questionnaire for Sales and Inventories of Alcoholic Beverages underwent more extensive changes. The modifications were discussed withstakeholders and the respondents were given an opportunity to comment before the new questionnaire was finalized. If further changes are needed to any of the questionnaires, proposed changes would go through a review committee and a field test with respondents and data users to ensure its relevancy.

6. Response and non-response

6.1. Response and non-response

Despite the best efforts of survey managers and operations staff to maximize response in the MRTS, some non-response will occur. For statistical establishments to be classified as responding, the degree of partial response (where an accurate response is obtained for only some of the questions asked a respondent) must meet a minimum threshold level below which the response would be rejected and considered a unit non-response.  In such an instance, the business is classified as not having responded at all.

Non-response has two effects on data: first it introduces bias in estimates when non-respondents differ from respondents in the characteristics measured; and second, it contributes to an increase in the sampling variance of estimates because the effective sample size is reduced from that originally sought.

The degree to which efforts are made to get a response from a non-respondent is based on budget and time constraints, its impact on the overall quality and the risk of non-response bias.

The main method to reduce the impact of non-response at sampling is to inflate the sample size through the use of over-sampling rates that have been determined from similar surveys.

Besides the methods to reduce the impact of non-response at sampling and collection, the non-responses to the survey that do occur are treated through imputation. In order to measure the amount of non-response that occurs each month, various response rates are calculated. For a given reference month, the estimation process is run at least twice (a preliminary and a revised run). Between each run, respondent data can be identified as unusable and imputed values can be corrected through respondent data. As a consequence, response rates are computed following each run of the estimation process.

For the MRTS, two types of rates are calculated (un-weighted and weighted). In order to assess the efficiency of the collection process, un-weighted response rates are calculated. Weighted rates, using the estimation weight and the value for the variable of interest, assess the quality of estimation. Within each of these types of rates, there are distinct rates for units that are surveyed and for units that are only modeled from administrative data that has been extracted from GST files.

To get a better picture of the success of the collection process, two un-weighted rates called the ‘collection results rate’ and the ‘extraction results rate’ are computed. They are computed by dividing the number of respondents by the number of units that we tried to contact or tried to receive extracted data for them. Non-monthly reporters (respondents with special reporting arrangements where they do not report every month but for whom actual data is available in subsequent revisions) are excluded from both the numerator and denominator for the months where no contact is performed.

In summary, the various response rates are calculated as follows:

Weighted rates:

Survey Response rate (estimation) =
Sum of weighted sales of units with response status i / Sum of survey weighted sales

where i = units that have either reported data that will be used in estimation or are converted refusals, or have reported data that has not yet been resolved for estimation.

Admin Response rate (estimation) =
Sum of weighted sales of units with response status ii / Sum of administrative weighted sales

where ii = units that have data that was extracted from administrative files and are usable for estimation.

Total Response rate (estimation) =
Sum of weighted sales of units with response status i or response status ii / Sum of all weighted sales

Un-weighted rates:

Survey Response rate (collection) =
Number of questionnaires with response status iii/ Number of questionnaires with response status iv

where iii = units that have either reported data (unresolved, used or not used for estimation) or are converted refusals.

where iv = all of the above plus units that have refused to respond, units that were not contacted and other types of non-respondent units.

Admin Response rate (extraction) =
Number of questionnaires with response status vi/ Number of questionnaires with response status vii

where vi = in-scope units that have data (either usable or non-usable) that was extracted from administrative files

where vii = all of the above plus units that have refused to report to the administrative data source, units that were not contacted and other types of non-respondent units.

(% of questionnaire collected over all in-scope questionnaires)

Collection Results Rate =
Number of questionnaires with response status iii / Number of questionnaires with response status viii

where iii = same as iii defined above

where viii = same as iv except for the exclusion of units that were contacted because their response is unavailable for a particular month since they are non-monthly reporters.

Extraction Results Rate =
Number of questionnaires with response status ix / Number of questionnaires with response status vii

where ix = same as vi with the addition of extracted units that have been imputed or were out of scope

where vii = same as vii defined above

(% of questionnaires collected over all questionnaire in-scope we tried to collect)

All the above weighted and un-weighted rates are provided at the industrial group, geography and size group level or for any combination of these levels.

Use of Administrative Data

Managing response burden is an ongoing challenge for Statistics Canada. In an attempt to alleviate response burden and survey costs, especially for smaller businesses, the MRTS has reduced the number of simple establishments in the sample that are surveyed directly and instead derives sales data for these establishments from Goods and Service Tax (GST) files using a statistical model. The model accounts for differences between sales and revenue (reported for GST purposes) as well as for the time lag between the survey reference period and the reference period of the GST file.

For more information on the methodology used for modeling sales from administrative data sources, refer to ‘Monthly Retail Trade Survey: Use of Administrative Data’ under ‘Documentation’ of the IMDB.

Table 1 contains the weighted response rates for all industry groups as well as for total retail trade for each province and territory. For more detailed weighted response rates, please contact the Marketing and Dissemination Section at (613) 951-3549, toll free: 1-877-421-3067 or by e-mail at retailinfo@statcan.

6.2. Methods used to reduce non-response at collection

Significant effort is spent trying to minimize non-response during collection. Methods used, among others, are interviewer techniques such as probing and persuasion, repeated re-scheduling and call-backs to obtain the information, and procedures dealing with how to handle non-compliant (refusal) respondents.

If data are unavailable at the time of collection, a respondent's best estimates are also accepted, and are subsequently revised once the actual data become available.

To minimize total non-response for all variables, partial responses are accepted. In addition, questionnaires are customized for the collection of certain variables, such as inventory, so that collection is timed for those months when the data are available.

Finally, to build trust and rapport between the interviewers and respondents, cases are generally assigned to the same interviewer each month. This action establishes a personal relationship between interviewer and respondent, and builds respondent trust.

7. Data collection and capture operations

Collection of the data is performed by Statistics Canada’s Regional Offices.

Table 1: Weighted response rates by NAICS, for all provinces and territories: March 2016 Table summary
This table displays the results of Table 1: Weighted response rates by NAICS Weighted Response Rates, calculated using Total, Survey and Administrative units of measure (appearing as column headers).
  Weighted Response Rates
  Total Survey Administrative
NAICS - Canada  
Motor Vehicle and Parts Dealers 92.3 93.0 52.4
Automobile Dealers 93.6 93.9 49.6
New Car Dealers 95.1 95.1 Note ...: not applicable
Used Car Dealers 71.0 74.0 49.6
Other Motor Vehicle Dealers 73.5 74.1 67.4
Automotive Parts, Accessories and Tire Stores 85.8 91.6 43.5
Furniture and Home Furnishings Stores 87.1 91.6 40.3
Furniture Stores 92.4 93.7 64.3
Home Furnishings Stores 77.9 87.3 29.3
Electronics and Appliance Stores 88.4 89.1 51.3
Building Material and Garden Equipment Dealers 86.2 89.6 42.5
Food and Beverage Stores 85.0 87.6 48.8
Grocery Stores 89.5 92.4 52.4
Grocery (except Convenience) Stores 91.8 94.5 55.8
Convenience Stores 55.3 59.6 26.2
Specialty Food Stores 62.7 67.9 35.7
Beer, Wine and Liquor Stores 73.0 74.1 31.5
Health and Personal Care Stores 90.9 91.4 84.1
Gasoline Stations 81.5 83.2 57.2
Clothing and Clothing Accessories Stores 84.3 85.0 55.8
Clothing Stores 84.8 85.5 55.4
Shoe Stores 80.8 81.6 20.7
Jewellery, Luggage and Leather Goods Stores 84.2 85.3 68.2
Sporting Goods, Hobby, Book and Music Stores 89.2 92.5 50.0
General Merchandise Stores 98.8 99.4 34.9
Department Stores 100.0 100.0 Note ...: not applicable
Other general merchandise stores 97.9 98.9 34.9
Miscellaneous Store Retailers 73.7 77.6 37.2
Total 89.0 90.6 52.3
Regions  
Newfoundland and Labrador 89.2 90.9 30.2
Prince Edward Island 83.8 84.5 32.8
Nova Scotia 87.9 89.1 47.0
New Brunswick 86.7 88.7 45.0
Québec 90.9 92.9 57.0
Ontario 89.7 91.6 47.0
Manitoba 82.8 83.0 65.1
Saskatchewan 90.7 92.6 47.0
Alberta 89.4 90.9 57.4
British Columbia 86.0 87.2 54.9
Yukon Territory 79.5 79.5 Note ...: not applicable
Northwest Territories 75.0 75.0 Note ...: not applicable
Nunavut 87.8 87.8 Note ...: not applicable


Weighted Response Rates

Respondents are sent a questionnaire or are contacted by telephone to obtain their sales and inventory values, as well as to confirm the opening or closing of business trading locations. Collection of the data begins approximately 7 working days after the end of the reference month and continues for the duration of that month.

New entrants to the survey are introduced to the survey via an introductory letter that informs the respondent that a representative of Statistics Canada will be calling. This call is to introduce the respondent to the survey, confirm the respondent's business activity, establish and begin data collection, as well as to answer any questions that the respondent may have.

8. Editing

Data editing is the application of checks to detect missing, invalid or inconsistent entries or to point to data records that are potentially in error. In the survey process for the MRTS, data editing is done at two different time periods.

First of all, editing is done during data collection. Once data are collected via the telephone, or via the receipt of completed mail-in questionnaires, the data are captured using customized data capture applications. All data are subjected to data editing. Edits during data collection are referred to as field edits and generally consist of validity and some simple consistency edits. They are used to detect mistakes made during the interview by the respondent or the interviewer and to identify missing information during collection in order to reduce the need for follow-up later on. Another purpose of the field edits is to clean up responses. In the MRTS, the current month’s responses are edited against the respondent’s previous month’s responses and/or the previous year’s responses for the current month. Field edits are also used to identify problems with data collection procedures and the design of the questionnaire, as well as the need for more interviewer training.

Follow-up with respondents occurs to validate potential erroneous data following any failed preliminary edit check of the data. Once validated, the collected data is regularly transmitted to the head office in Ottawa.

Secondly, editing known as statistical editing is also done after data collection and this is more empirical in nature. Statistical editing is run prior to imputation in order to identify the data that will be used as a basis to impute non-respondents. Large outliers that could disrupt a monthly trend are excluded from trend calculations by the statistical edits. It should be noted that adjustments are not made at this stage to correct the reported outliers.

The first step in the statistical editing is to identify which responses will be subjected to the statistical edit rules. Reported data for the current reference month will go through various edit checks.

The first set of edit checks is based on the Hidiriglou-Berthelot method whereby a ratio of the respondent’s current month data over historical (last month, same month last year) or auxiliary data is analyzed. When the respondent’s ratio differs significantly from ratios of respondents who are similar in terms of industry and/or geography group, the response is deemed an outlier.

The second set of edits consists of an edit known as the share of market edit. With this method, one is able to edit all respondents, even those where historical and auxiliary data is unavailable. The method relies on current month data only. Therefore, within a group of respondents, that are similar in terms of industrial group and/or geography, if the weighted contribution of a respondent to the group’s total is too large, it will be flagged as an outlier.

For edit checks based on the Hidiriglou-Berthelot method, data that are flagged as an outlier will not be included in the imputation models (those based on ratios). Also, data that are flagged as outliers in the share of market edit will not be included in the imputation models where means and medians are calculated to impute for responses that have no historical responses.

In conjunction with the statistical editing after data collection of reported data, there is also error detection done on the extracted GST data. Modeled data based on the GST are also subject to an extensive series of processing steps which thoroughly verify each record that is the basis for the model as well as the record being modeled. Edits are performed at a more aggregate level (industry by geography level) to detect records which deviate from the expected range, either by exhibiting large month-to-month change, or differing significantly from the remaining units. All data which fail these edits are subject to manual inspection and possible corrective action.

9. Imputation

Imputation in the MRTS is the process used to assign replacement values for missing data. This is done by assigning values when they are missing on the record being edited to ensure that estimates are of high quality and that a plausible, internal consistency is created. Due to concerns of response burden, cost and timeliness, it is generally impossible to do all follow-ups with the respondents in order to resolve missing responses. Since it is desirable to produce a complete and consistent microdata file, imputation is used to handle the remaining missing cases.

In the MRTS, imputation is based on historical data or administrative data (GST sales). The appropriate method is selected according to a strategy that is based on whether historical data is available, auxiliary data is available and/or which reference month is being processed.

There are three types of historical imputation methods. The first type is a general trend that uses one historical data source (previous month, data from next month or data from same month previous year). The second type is a regression model where data from previous month and same month, previous year are used simultaneously. The third type uses the historical data as a direct replacement value for a non-respondent. Depending upon the particular reference month, there is an order of preference that exists so that top quality imputation can result. The historical imputation method that was labelled as the third type above is always the last option in the order for each reference month.

The imputation method using administrative data is automatically selected when historical information is unavailable for a non-respondent. Trends are then applied to the administrative data source (monthly size) depending on whether the structure is simple, e.g. enterprises with only one establishment, or the unit has a more complex structure.

10. Estimation

Estimation is a process that approximates unknown population parameters using only part of the population that is included in a sample. Inferences about these unknown parameters are then made, using the sample data and associated survey design. This stage uses Statistics Canada's Generalized Estimation System (GES).

For retail sales, the population is divided into a survey portion (take-all and take-some strata) and a non-survey portion (take-none stratum). From the sample that is drawn from the survey portion, an estimate for the population is determined through the use of a Horvitz-Thompson estimator where responses for sales are weighted by using the inverses of the inclusion probabilities of the sampled units. Such weights (called sampling weights) can be interpreted as the number of times that each sampled unit should be replicated to represent the entire population. The calculated weighted sales values are summed by domain, to produce the total sales estimates by each industrial group / geographic area combination. A domain is defined as the most recent classification values available from the BR for the unit and the survey reference period. These domains may differ from the original sampling strata because units may have changed size, industry or location. Changes in classification are reflected immediately in the estimates and do not accumulate over time. For the non-survey portion, the sales are estimated with statistical models using monthly GST sales.

For more information on the methodology for modeling sales from administrative data sources which also contributes to the estimates of the survey portion, refer to ‘Monthly Retail Survey: Use of Administrative Data’ under ‘Documentation’ of the IMDB.

The measure of precision used for the MRTS to evaluate the quality of a population parameter estimate and to obtain valid inferences is the variance. The variance from the survey portion is derived directly from a stratified simple random sample without replacement.

Sample estimates may differ from the expected value of the estimates. However, since the estimate is based on a probability sample, the variability of the sample estimate with respect to its expected value can be measured. The variance of an estimate is a measure of the precision of the sample estimate and is defined as the average, over all possible samples, of the squared difference of the estimate from its expected value.

11. Revisions and seasonal adjustment

Revisions in the raw data are required to correct known non-sampling errors. These normally include replacing imputed data with reported data, corrections to previously reported data, and estimates for new births that were not known at the time of the original estimates.

Raw data are revised, on a monthly basis, for the month immediately prior to the current reference month being published. That is, when data for December are being published for the first time, there will also be revisions, if necessary, to the raw data for November. In addition, revisions are made once a year, with the initial release of the February data, for all months in the previous years. The purpose is to correct any significant problems that have been found that apply for an extended period. The actual period of revision depends on the nature of the problem identified, but rarely exceeds three years. The revision period can be extended when historical revisions or restratitfication are done.

Retail trade data are seasonally adjusted using the X12-ARIMA method. This consists of extrapolating a year's worth of raw data with the ARIMA model (auto-regressive integrated moving average model), and of seasonally adjusting the raw time series. Finally, the annual totals of the seasonally adjusted series are forced to the annual totals of the original series.

The seasonally adjusted data also need to be revised. In part, they need to reflect the revisions identified for the raw data. Also, the seasonally adjusted estimates are calculated using X-12-ARIMA, and are sensitive to the most recent values reported in the raw data. For this reason, with the release of each month of new data, the seasonally adjusted values for the previous three months are revised.  A seasonally adjusted time series is a time series that has been modified to eliminate the effect of seasonal and calendar influences. For this reason, the seasonally adjusted data allows for more meaningful comparisons of economic conditions from month to month.

Once a year, seasonal adjustments options are reviewed to take into account the most recent data. Revised seasonally adjusted estimates for each month in the previous years are released at the same time as the annual revision to the raw data. The actual period of revision depends on the number years the raw data was revised.

12. Data quality evaluation

The methodology of this survey has been designed to control errors and to reduce their potential effects on estimates. However, the survey results remain subject to errors, of which sampling error is only one component of the total survey error. Sampling error results when observations are made only on a sample and not on the entire population. All other errors arising from the various phases of a survey are referred to as nonsampling errors. For example, these types of errors can occur when a respondent provides incorrect information or does not answer certain questions; when a unit in the target population is omitted or covered more than once; when GST data for records being modeled for a particular month are not representative of the actual record for various reasons; when a unit that is out of scope for the survey is included by mistake or when errors occur in data processing, such as coding or capture errors.

Prior to publication, combined survey results are analyzed for comparability; in general, this includes a detailed review of individual responses (especially for large businesses), general economic conditions and historical trends.

A common measure of data quality for surveys is the coefficient of variation (CV). The coefficient of variation, defined as the standard error divided by the sample estimate, is a measure of precision in relative terms. Since the coefficient of variation is calculated from responses of individual units, it also measures some non-sampling errors.

The formula used to calculate coefficients of variation (CV) as percentages is:

CV (X) = S(X) * 100% / X
where X denotes the estimate and S(X) denotes the standard error of X.

Confidence intervals can be constructed around the estimates using the estimate and the CV. Thus, for our sample, it is possible to state with a given level of confidence that the expected value will fall within the confidence interval constructed around the estimate. For example, if an estimate of $12,000,000 has a CV of 2%, the standard error will be $240,000 (the estimate multiplied by the CV). It can be stated with 68% confidence that the expected values will fall within the interval whose length equals the standard deviation about the estimate, i.e. between $11,760,000 and $12,240,000.

Alternatively, it can be stated with 95% confidence that the expected value will fall within the interval whose length equals two standard deviations about the estimate, i.e. between $11,520,000 and $12,480,000.

Finally, due to the small contribution of the non-survey portion to the total estimates, bias in the non-survey portion has a negligible impact on the CVs. Therefore, the CV from the survey portion is used for the total estimate that is the summation of estimates from the surveyed and non-surveyed portions.

13. 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.

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.

Monthly Retail Trade Survey (MRTS) Data Quality Statement

Objectives, uses and users
Concepts, variables and classifications
Coverage and frames
Sampling
Questionnaire design
Response and non-response
Data collection and capture operations
Editing
Imputation
Estimation
Revisions and seasonal adjustment
Data quality evaluation
Disclosure control

1. Objectives, uses and users

1.1. Objective

The Monthly Retail Trade Survey (MRTS) provides information on the performance of the retail trade sector on a monthly basis, and when combined with other statistics, represents an important indicator of the state of the Canadian economy.

1.2. Uses

The estimates provide a measure of the health and performance of the retail trade sector. Information collected is used to estimate level and monthly trend for retail sales. At the end of each year, the estimates provide a preliminary look at annual retail sales and performance.

1.3. Users

A variety of organizations, sector associations, and levels of government make use of the information. Retailers rely on the survey results to compare their performance against similar types of businesses, as well as for marketing purposes. Retail associations are able to monitor industry performance and promote their retail industries. Investors can monitor industry growth, which can result in better access to investment capital by retailers. Governments are able to understand the role of retailers in the economy, which aids in the development of policies and tax incentives. As an important industry in the Canadian economy, governments are able to better determine the overall health of the economy through the use of the estimates in the calculation of the nation’s Gross Domestic Product (GDP).

2. Concepts, variables and classifications

2.1. Concepts

The retail trade sector comprises establishments primarily engaged in retailing merchandise, generally without transformation, and rendering services incidental to the sale of merchandise.

The retailing process is the final step in the distribution of merchandise; retailers are therefore organized to sell merchandise in small quantities to the general public. This sector comprises two main types of retailers, that is, store and non-store retailers. The MRTS covers only store retailers. Their main characteristics are described below. Store retailers operate fixed point-of-sale locations, located and designed to attract a high volume of walk-in customers. In general, retail stores have extensive displays of merchandise and use mass-media advertising to attract customers. They typically sell merchandise to the general public for personal or household consumption, but some also serve business and institutional clients. These include establishments such as office supplies stores, computer and software stores, gasoline stations, building material dealers, plumbing supplies stores and electrical supplies stores.

In addition to selling merchandise, some types of store retailers are also engaged in the provision of after-sales services, such as repair and installation. For example, new automobile dealers, electronic and appliance stores and musical instrument and supplies stores often provide repair services, while floor covering stores and window treatment stores often provide installation services. As a general rule, establishments engaged in retailing merchandise and providing after sales services are classified in this sector. Catalogue sales showrooms, gasoline service stations, and mobile home dealers are treated as store retailers.

2.2. Variables

Sales are defined as the sales of all goods purchased for resale, net of returns and discounts. This includes commission revenue and fees earned from selling goods and services on account of others, such as selling lottery tickets, bus tickets, and phone cards. It also includes parts and labour revenue from repair and maintenance; revenue from rental and leasing of goods and equipment; revenues from services, including food services; sales of goods manufactured as a secondary activity; and the proprietor’s withdrawals, at retail, of goods for personal use. Other revenue from rental of real estate, placement fees, operating subsidies, grants, royalties and franchise fees are excluded.

Trading Location is the physical location(s) in which business activity is conducted in each province and territory, and for which sales are credited or recognized in the financial records of the company. For retailers, this would normally be a store.

Constant Dollars: The value of retail trade is measured in two ways; including the effects of price change on sales and net of the effects of price change. The first measure is referred to as retail trade in current dollars and the latter as retail trade in constant dollars. The method of calculating the current dollar estimate is to aggregate the weighted value of sales for all retail outlets. The method of calculating the constant dollar estimate is to first adjust the sales values to a base year, using the Consumer Price Index, and then sum up the resulting values.

2.3. Classification

The Monthly Retail Trade Survey is based on the definition of retail trade under the NAICS (North American Industry Classification System). NAICS is the agreed upon common framework for the production of comparable statistics by the statistical agencies of Canada, Mexico and the United States. The agreement defines the boundaries of twenty sectors. NAICS is based on a production-oriented, or supply based conceptual framework in that establishments are groups into industries according to similarity in production processes used to produce goods and services.

Estimates appear for 21 industries based on special aggregations of the 2012 North American Industry Classification System (NAICS) industries. The 21 industries are further aggregated to 11 sub-sectors.

Geographically, sales estimates are produced for Canada and each province and territory.

3. Coverage and frames

Statistics Canada’s Business Register ( BR) provides the frame for the Monthly Retail Trade Survey. The BR is a structured list of businesses engaged in the production of goods and services in Canada. It is a centrally maintained database containing detailed descriptions of most business entities operating within Canada. The BR includes all incorporated businesses, with or without employees. For unincorporated businesses, the BR includes all employers with businesses, and businesses with no employees with annual sales that have a Goods and Services Tax (GST) or annual revenue that declares individual taxes.  annual sales greater than $30,000 that have a Goods and Services Tax (GST) account (the BR does not include unincorporated businesses with no employees and with annual sales less than $30,000).

The businesses on the BR are represented by a hierarchical structure with four levels, with the statistical enterprise at the top, followed by the statistical company, the statistical establishment and the statistical location. An enterprise can be linked to one or more statistical companies, a statistical company can be linked to one or more statistical establishments, and a statistical establishment to one or more statistical locations.

The target population for the MRTS consists of all statistical establishments on the BR that are classified to the retail sector using the North American Industry Classification System (NAICS) (approximately 200,000 establishments). The NAICS code range for the retail sector is 441100 to 453999. A statistical establishment is the production entity or the smallest grouping of production entities which: produces a homogeneous set of goods or services; does not cross provincial boundaries; and provides data on the value of output, together with the cost of principal intermediate inputs used, along with the cost and quantity of labour used to produce the output. The production entity is the physical unit where the business operations are carried out. It must have a civic address and dedicated labour.

The exclusions to the target population are ancillary establishments (producers of services in support of the activity of producing goods and services for the market of more than one establishment within the enterprise, and serves as a cost centre or a discretionary expense centre for which data on all its costs including labour and depreciation can be reported by the business), future establishments, establishments with a missing or a zero gross business income (GBI) value on the BR and establishments in the following non-covered NAICS:

  • 4541 (electronic shopping and mail-order houses)
  • 4542 (vending machine operators)
  • 45431 (fuel dealers)
  • 45439 (other direct selling establishments)

4. Sampling

The MRTS sample consists of 10,000 groups of establishments (clusters) classified to the Retail Trade sector selected from the Statistics Canada Business Register. A cluster of establishments is defined as all establishments belonging to a statistical enterprise that are in the same industrial group and geographical region. The MRTS uses a stratified design with simple random sample selection in each stratum. The stratification is done by industry groups (the mainly, but not only four digit level NAICS), and the geographical regions consisting of the provinces and territories, as well as three provincial sub-regions. We further stratify the population by size.

The size measure is created using a combination of independent survey data and three administrative variables: the annual profiled revenue, the GST sales expressed on an annual basis, and the declared tax revenue (T1 or T2). The size strata consist of one take-all (census), at most, two take-some (partially sampled) strata, and one take-none (non-sampled) stratum. Take-none strata serve to reduce respondent burden by excluding the smaller businesses from the surveyed population. These businesses should represent at most ten percent of total sales. Instead of sending questionnaires to these businesses, the estimates are produced through the use of administrative data.

The sample was allocated optimally in order to reach target coefficients of variation at the national, provincial/territorial, industrial, and industrial groups by province/territory levels. The sample was also inflated to compensate for dead, non-responding, and misclassified units.

MRTS is a repeated survey with maximisation of monthly sample overlap. The sample is kept month after month, and every month new units are added (births) to the sample.  MRTS births, i.e., new clusters of establishment(s), are identified every month via the BR’s latest universe. They are stratified according to the same criteria as the initial population. A sample of these births is selected according to the sampling fraction of the stratum to which they belong and is added to the monthly sample. Deaths occur on a monthly basis. A death can be a cluster of establishment(s) that have ceased their activities (out-of-business) or whose major activities are no longer in retail trade (out-of-scope). The status of these businesses is updated on the BR using administrative sources and survey feedback, including feedback from the MRTS. Methods to treat dead units and misclassified units are part of the sample and population update procedures.

5. Questionnaire design

The Monthly Retail Trade Survey incorporates the following sub-surveys:

Monthly Retail Trade Survey - R8

Monthly Retail Trade Survey (with inventories) – R8

Survey of Sales and Inventories of Alcoholic Beverages

The questionnaires collect monthly data on retail sales and the number of trading locations by province or territory and inventories of goods owned and intended for resale from a sample of retailers. The items on the questionnaires have remained unchanged for several years. For the 2004 redesign, the general questionnaires were subject to cosmetic changes only. The questionnaire for Sales and Inventories of Alcoholic Beverages underwent more extensive changes. The modifications were discussed withstakeholders and the respondents were given an opportunity to comment before the new questionnaire was finalized. If further changes are needed to any of the questionnaires, proposed changes would go through a review committee and a field test with respondents and data users to ensure its relevancy.

6. Response and non-response

6.1. Response and non-response

Despite the best efforts of survey managers and operations staff to maximize response in the MRTS, some non-response will occur. For statistical establishments to be classified as responding, the degree of partial response (where an accurate response is obtained for only some of the questions asked a respondent) must meet a minimum threshold level below which the response would be rejected and considered a unit non-response.  In such an instance, the business is classified as not having responded at all.

Non-response has two effects on data: first it introduces bias in estimates when non-respondents differ from respondents in the characteristics measured; and second, it contributes to an increase in the sampling variance of estimates because the effective sample size is reduced from that originally sought.

The degree to which efforts are made to get a response from a non-respondent is based on budget and time constraints, its impact on the overall quality and the risk of non-response bias.

The main method to reduce the impact of non-response at sampling is to inflate the sample size through the use of over-sampling rates that have been determined from similar surveys.

Besides the methods to reduce the impact of non-response at sampling and collection, the non-responses to the survey that do occur are treated through imputation. In order to measure the amount of non-response that occurs each month, various response rates are calculated. For a given reference month, the estimation process is run at least twice (a preliminary and a revised run). Between each run, respondent data can be identified as unusable and imputed values can be corrected through respondent data. As a consequence, response rates are computed following each run of the estimation process.

For the MRTS, two types of rates are calculated (un-weighted and weighted). In order to assess the efficiency of the collection process, un-weighted response rates are calculated. Weighted rates, using the estimation weight and the value for the variable of interest, assess the quality of estimation. Within each of these types of rates, there are distinct rates for units that are surveyed and for units that are only modeled from administrative data that has been extracted from GST files.

To get a better picture of the success of the collection process, two un-weighted rates called the ‘collection results rate’ and the ‘extraction results rate’ are computed. They are computed by dividing the number of respondents by the number of units that we tried to contact or tried to receive extracted data for them. Non-monthly reporters (respondents with special reporting arrangements where they do not report every month but for whom actual data is available in subsequent revisions) are excluded from both the numerator and denominator for the months where no contact is performed.

In summary, the various response rates are calculated as follows:

Weighted rates:

Survey Response rate (estimation) =
Sum of weighted sales of units with response status i / Sum of survey weighted sales

where i = units that have either reported data that will be used in estimation or are converted refusals, or have reported data that has not yet been resolved for estimation.

Admin Response rate (estimation) =
Sum of weighted sales of units with response status ii / Sum of administrative weighted sales

where ii = units that have data that was extracted from administrative files and are usable for estimation.

Total Response rate (estimation) =
Sum of weighted sales of units with response status i or response status ii / Sum of all weighted sales

Un-weighted rates:

Survey Response rate (collection) =
Number of questionnaires with response status iii/ Number of questionnaires with response status iv

where iii = units that have either reported data (unresolved, used or not used for estimation) or are converted refusals.

where iv = all of the above plus units that have refused to respond, units that were not contacted and other types of non-respondent units.

Admin Response rate (extraction) =
Number of questionnaires with response status vi/ Number of questionnaires with response status vii

where vi = in-scope units that have data (either usable or non-usable) that was extracted from administrative files

where vii = all of the above plus units that have refused to report to the administrative data source, units that were not contacted and other types of non-respondent units.

(% of questionnaire collected over all in-scope questionnaires)

Collection Results Rate =
Number of questionnaires with response status iii / Number of questionnaires with response status viii

where iii = same as iii defined above

where viii = same as iv except for the exclusion of units that were contacted because their response is unavailable for a particular month since they are non-monthly reporters.

Extraction Results Rate =
Number of questionnaires with response status ix / Number of questionnaires with response status vii

where ix = same as vi with the addition of extracted units that have been imputed or were out of scope

where vii = same as vii defined above

(% of questionnaires collected over all questionnaire in-scope we tried to collect)

All the above weighted and un-weighted rates are provided at the industrial group, geography and size group level or for any combination of these levels.

Use of Administrative Data

Managing response burden is an ongoing challenge for Statistics Canada. In an attempt to alleviate response burden and survey costs, especially for smaller businesses, the MRTS has reduced the number of simple establishments in the sample that are surveyed directly and instead derives sales data for these establishments from Goods and Service Tax (GST) files using a statistical model. The model accounts for differences between sales and revenue (reported for GST purposes) as well as for the time lag between the survey reference period and the reference period of the GST file.

For more information on the methodology used for modeling sales from administrative data sources, refer to ‘Monthly Retail Trade Survey: Use of Administrative Data’ under ‘Documentation’ of the IMDB.

Table 1 contains the weighted response rates for all industry groups as well as for total retail trade for each province and territory. For more detailed weighted response rates, please contact the Marketing and Dissemination Section at (613) 951-3549, toll free: 1-877-421-3067 or by e-mail at retailinfo@statcan.

6.2. Methods used to reduce non-response at collection

Significant effort is spent trying to minimize non-response during collection. Methods used, among others, are interviewer techniques such as probing and persuasion, repeated re-scheduling and call-backs to obtain the information, and procedures dealing with how to handle non-compliant (refusal) respondents.

If data are unavailable at the time of collection, a respondent's best estimates are also accepted, and are subsequently revised once the actual data become available.

To minimize total non-response for all variables, partial responses are accepted. In addition, questionnaires are customized for the collection of certain variables, such as inventory, so that collection is timed for those months when the data are available.

Finally, to build trust and rapport between the interviewers and respondents, cases are generally assigned to the same interviewer each month. This action establishes a personal relationship between interviewer and respondent, and builds respondent trust.

7. Data collection and capture operations

Collection of the data is performed by Statistics Canada’s Regional Offices.

Table 1: Weighted response rates by NAICS, for all provinces and territories: November 2015
Table summary
This table displays the results of Weighted response rates by NAICS Weighted Response Rates, calculated using Total, Survey and Administrative units of measure (appearing as column headers).
  Weighted Response Rates
Total Survey Administrative
NAICS - Canada  
Motor Vehicle and Parts Dealers 92.1 92.8 56.2
Automobile Dealers 93.9 94.3 47.2
New Car Dealers 95.6 95.6 Note ...: not applicable
Used Car Dealers 69.0 72.5 47.2
Other Motor Vehicle Dealers 67.5 68.9 51.7
Automotive Parts, Accessories and Tire Stores 83.0 84.6 68.3
Furniture and Home Furnishings Stores 77.9 80.4 53.9
Furniture Stores 77.9 78.9 59.4
Home Furnishings Stores 78.0 82.9 50.7
Electronics and Appliance Stores 86.0 86.7 32.7
Building Material and Garden Equipment Dealers 88.7 92.1 51.7
Food and Beverage Stores 89.8 91.5 66.2
Grocery Stores 90.3 91.9 70.5
Grocery (except Convenience) Stores 92.9 94.1 77.1
Convenience Stores 54.5 59.5 22.8
Specialty Food Stores 69.0 72.6 50.9
Beer, Wine and Liquor Stores 93.1 94.3 38.2
Health and Personal Care Stores 85.3 84.9 91.7
Gasoline Stations 72.4 73.2 59.4
Clothing and Clothing Accessories Stores 82.5 83.3 50.1
Clothing Stores 82.4 83.0 53.6
Shoe Stores 82.5 83.3 19.3
Jewellery, Luggage and Leather Goods Stores 83.9 86.0 46.3
Sporting Goods, Hobby, Book and Music Stores 88.0 91.1 28.8
General Merchandise Stores 99.2 99.4 74.3
Department Stores 100.0 100.0 Note ...: not applicable
Other general merchandise stores 98.7 99.0 74.3
Miscellaneous Store Retailers 77.6 83.5 23.0
Total 88.4 89.6 60.1
Regions  
Newfoundland and Labrador 83.5 84.7 39.9
Prince Edward Island 80.4 81.3 9.2
Nova Scotia 89.3 90.4 55.0
New Brunswick 83.0 84.2 58.2
Québec 87.9 89.3 64.2
Ontario 89.3 90.7 58.4
Manitoba 87.5 87.9 68.7
Saskatchewan 91.0 92.4 54.6
Alberta 87.6 88.7 61.6
British Columbia 88.8 89.8 58.4
Yukon Territory 85.2 85.2 Note ...: not applicable
Northwest Territories 61.7 61.7 Note ...: not applicable
Nunavut 69.5 69.5 Note ...: not applicable


Weighted Response Rates

Respondents are sent a questionnaire or are contacted by telephone to obtain their sales and inventory values, as well as to confirm the opening or closing of business trading locations. Collection of the data begins approximately 7 working days after the end of the reference month and continues for the duration of that month.

New entrants to the survey are introduced to the survey via an introductory letter that informs the respondent that a representative of Statistics Canada will be calling. This call is to introduce the respondent to the survey, confirm the respondent's business activity, establish and begin data collection, as well as to answer any questions that the respondent may have.

8. Editing

Data editing is the application of checks to detect missing, invalid or inconsistent entries or to point to data records that are potentially in error. In the survey process for the MRTS, data editing is done at two different time periods.

First of all, editing is done during data collection. Once data are collected via the telephone, or via the receipt of completed mail-in questionnaires, the data are captured using customized data capture applications. All data are subjected to data editing. Edits during data collection are referred to as field edits and generally consist of validity and some simple consistency edits. They are used to detect mistakes made during the interview by the respondent or the interviewer and to identify missing information during collection in order to reduce the need for follow-up later on. Another purpose of the field edits is to clean up responses. In the MRTS, the current month’s responses are edited against the respondent’s previous month’s responses and/or the previous year’s responses for the current month. Field edits are also used to identify problems with data collection procedures and the design of the questionnaire, as well as the need for more interviewer training.

Follow-up with respondents occurs to validate potential erroneous data following any failed preliminary edit check of the data. Once validated, the collected data is regularly transmitted to the head office in Ottawa.

Secondly, editing known as statistical editing is also done after data collection and this is more empirical in nature. Statistical editing is run prior to imputation in order to identify the data that will be used as a basis to impute non-respondents. Large outliers that could disrupt a monthly trend are excluded from trend calculations by the statistical edits. It should be noted that adjustments are not made at this stage to correct the reported outliers.

The first step in the statistical editing is to identify which responses will be subjected to the statistical edit rules. Reported data for the current reference month will go through various edit checks.

The first set of edit checks is based on the Hidiriglou-Berthelot method whereby a ratio of the respondent’s current month data over historical (last month, same month last year) or auxiliary data is analyzed. When the respondent’s ratio differs significantly from ratios of respondents who are similar in terms of industry and/or geography group, the response is deemed an outlier.

The second set of edits consists of an edit known as the share of market edit. With this method, one is able to edit all respondents, even those where historical and auxiliary data is unavailable. The method relies on current month data only. Therefore, within a group of respondents, that are similar in terms of industrial group and/or geography, if the weighted contribution of a respondent to the group’s total is too large, it will be flagged as an outlier.

For edit checks based on the Hidiriglou-Berthelot method, data that are flagged as an outlier will not be included in the imputation models (those based on ratios). Also, data that are flagged as outliers in the share of market edit will not be included in the imputation models where means and medians are calculated to impute for responses that have no historical responses.

In conjunction with the statistical editing after data collection of reported data, there is also error detection done on the extracted GST data. Modeled data based on the GST are also subject to an extensive series of processing steps which thoroughly verify each record that is the basis for the model as well as the record being modeled. Edits are performed at a more aggregate level (industry by geography level) to detect records which deviate from the expected range, either by exhibiting large month-to-month change, or differing significantly from the remaining units. All data which fail these edits are subject to manual inspection and possible corrective action.

9. Imputation

Imputation in the MRTS is the process used to assign replacement values for missing data. This is done by assigning values when they are missing on the record being edited to ensure that estimates are of high quality and that a plausible, internal consistency is created. Due to concerns of response burden, cost and timeliness, it is generally impossible to do all follow-ups with the respondents in order to resolve missing responses. Since it is desirable to produce a complete and consistent microdata file, imputation is used to handle the remaining missing cases.

In the MRTS, imputation is based on historical data or administrative data (GST sales). The appropriate method is selected according to a strategy that is based on whether historical data is available, auxiliary data is available and/or which reference month is being processed.

There are three types of historical imputation methods. The first type is a general trend that uses one historical data source (previous month, data from next month or data from same month previous year). The second type is a regression model where data from previous month and same month, previous year are used simultaneously. The third type uses the historical data as a direct replacement value for a non-respondent. Depending upon the particular reference month, there is an order of preference that exists so that top quality imputation can result. The historical imputation method that was labelled as the third type above is always the last option in the order for each reference month.

The imputation method using administrative data is automatically selected when historical information is unavailable for a non-respondent. Trends are then applied to the administrative data source (monthly size) depending on whether the structure is simple, e.g. enterprises with only one establishment, or the unit has a more complex structure.

10. Estimation

Estimation is a process that approximates unknown population parameters using only part of the population that is included in a sample. Inferences about these unknown parameters are then made, using the sample data and associated survey design. This stage uses Statistics Canada's Generalized Estimation System (GES).

For retail sales, the population is divided into a survey portion (take-all and take-some strata) and a non-survey portion (take-none stratum). From the sample that is drawn from the survey portion, an estimate for the population is determined through the use of a Horvitz-Thompson estimator where responses for sales are weighted by using the inverses of the inclusion probabilities of the sampled units. Such weights (called sampling weights) can be interpreted as the number of times that each sampled unit should be replicated to represent the entire population. The calculated weighted sales values are summed by domain, to produce the total sales estimates by each industrial group / geographic area combination. A domain is defined as the most recent classification values available from the BR for the unit and the survey reference period. These domains may differ from the original sampling strata because units may have changed size, industry or location. Changes in classification are reflected immediately in the estimates and do not accumulate over time. For the non-survey portion, the sales are estimated with statistical models using monthly GST sales.

For more information on the methodology for modeling sales from administrative data sources which also contributes to the estimates of the survey portion, refer to ‘Monthly Retail Survey: Use of Administrative Data’ under ‘Documentation’ of the IMDB.

The measure of precision used for the MRTS to evaluate the quality of a population parameter estimate and to obtain valid inferences is the variance. The variance from the survey portion is derived directly from a stratified simple random sample without replacement.

Sample estimates may differ from the expected value of the estimates. However, since the estimate is based on a probability sample, the variability of the sample estimate with respect to its expected value can be measured. The variance of an estimate is a measure of the precision of the sample estimate and is defined as the average, over all possible samples, of the squared difference of the estimate from its expected value.

11. Revisions and seasonal adjustment

Revisions in the raw data are required to correct known non-sampling errors. These normally include replacing imputed data with reported data, corrections to previously reported data, and estimates for new births that were not known at the time of the original estimates.

Raw data are revised, on a monthly basis, for the month immediately prior to the current reference month being published. That is, when data for December are being published for the first time, there will also be revisions, if necessary, to the raw data for November. In addition, revisions are made once a year, with the initial release of the February data, for all months in the previous years. The purpose is to correct any significant problems that have been found that apply for an extended period. The actual period of revision depends on the nature of the problem identified, but rarely exceeds three years. The revision period can be extended when historical revisions or restratitfication are done.

Retail trade data are seasonally adjusted using the X12-ARIMA method. This consists of extrapolating a year's worth of raw data with the ARIMA model (auto-regressive integrated moving average model), and of seasonally adjusting the raw time series. Finally, the annual totals of the seasonally adjusted series are forced to the annual totals of the original series.

The seasonally adjusted data also need to be revised. In part, they need to reflect the revisions identified for the raw data. Also, the seasonally adjusted estimates are calculated using X-12-ARIMA, and are sensitive to the most recent values reported in the raw data. For this reason, with the release of each month of new data, the seasonally adjusted values for the previous three months are revised.  A seasonally adjusted time series is a time series that has been modified to eliminate the effect of seasonal and calendar influences. For this reason, the seasonally adjusted data allows for more meaningful comparisons of economic conditions from month to month.

Once a year, seasonal adjustments options are reviewed to take into account the most recent data. Revised seasonally adjusted estimates for each month in the previous years are released at the same time as the annual revision to the raw data. The actual period of revision depends on the number years the raw data was revised.

12. Data quality evaluation

The methodology of this survey has been designed to control errors and to reduce their potential effects on estimates. However, the survey results remain subject to errors, of which sampling error is only one component of the total survey error. Sampling error results when observations are made only on a sample and not on the entire population. All other errors arising from the various phases of a survey are referred to as nonsampling errors. For example, these types of errors can occur when a respondent provides incorrect information or does not answer certain questions; when a unit in the target population is omitted or covered more than once; when GST data for records being modeled for a particular month are not representative of the actual record for various reasons; when a unit that is out of scope for the survey is included by mistake or when errors occur in data processing, such as coding or capture errors.

Prior to publication, combined survey results are analyzed for comparability; in general, this includes a detailed review of individual responses (especially for large businesses), general economic conditions and historical trends.

A common measure of data quality for surveys is the coefficient of variation (CV). The coefficient of variation, defined as the standard error divided by the sample estimate, is a measure of precision in relative terms. Since the coefficient of variation is calculated from responses of individual units, it also measures some non-sampling errors.

The formula used to calculate coefficients of variation (CV) as percentages is:

CV (X) = S(X) * 100% / X
where X denotes the estimate and S(X) denotes the standard error of X.

Confidence intervals can be constructed around the estimates using the estimate and the CV. Thus, for our sample, it is possible to state with a given level of confidence that the expected value will fall within the confidence interval constructed around the estimate. For example, if an estimate of $12,000,000 has a CV of 2%, the standard error will be $240,000 (the estimate multiplied by the CV). It can be stated with 68% confidence that the expected values will fall within the interval whose length equals the standard deviation about the estimate, i.e. between $11,760,000 and $12,240,000.

Alternatively, it can be stated with 95% confidence that the expected value will fall within the interval whose length equals two standard deviations about the estimate, i.e. between $11,520,000 and $12,480,000.

Finally, due to the small contribution of the non-survey portion to the total estimates, bias in the non-survey portion has a negligible impact on the CVs. Therefore, the CV from the survey portion is used for the total estimate that is the summation of estimates from the surveyed and non-surveyed portions.

13. 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.

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.

Monthly Retail Trade Survey (MRTS) Data Quality Statement

Objectives, uses and users
Concepts, variables and classifications
Coverage and frames
Sampling
Questionnaire design
Response and non-response
Data collection and capture operations
Editing
Imputation
Estimation
Revisions and seasonal adjustment
Data quality evaluation
Disclosure control

1. Objectives, uses and users

1.1. Objective

The Monthly Retail Trade Survey (MRTS) provides information on the performance of the retail trade sector on a monthly basis, and when combined with other statistics, represents an important indicator of the state of the Canadian economy.

1.2. Uses

The estimates provide a measure of the health and performance of the retail trade sector. Information collected is used to estimate level and monthly trend for retail sales. At the end of each year, the estimates provide a preliminary look at annual retail sales and performance.

1.3. Users

A variety of organizations, sector associations, and levels of government make use of the information. Retailers rely on the survey results to compare their performance against similar types of businesses, as well as for marketing purposes. Retail associations are able to monitor industry performance and promote their retail industries. Investors can monitor industry growth, which can result in better access to investment capital by retailers. Governments are able to understand the role of retailers in the economy, which aids in the development of policies and tax incentives. As an important industry in the Canadian economy, governments are able to better determine the overall health of the economy through the use of the estimates in the calculation of the nation’s Gross Domestic Product (GDP).

2. Concepts, variables and classifications

2.1. Concepts

The retail trade sector comprises establishments primarily engaged in retailing merchandise, generally without transformation, and rendering services incidental to the sale of merchandise.

The retailing process is the final step in the distribution of merchandise; retailers are therefore organized to sell merchandise in small quantities to the general public. This sector comprises two main types of retailers, that is, store and non-store retailers. The MRTS covers only store retailers. Their main characteristics are described below. Store retailers operate fixed point-of-sale locations, located and designed to attract a high volume of walk-in customers. In general, retail stores have extensive displays of merchandise and use mass-media advertising to attract customers. They typically sell merchandise to the general public for personal or household consumption, but some also serve business and institutional clients. These include establishments such as office supplies stores, computer and software stores, gasoline stations, building material dealers, plumbing supplies stores and electrical supplies stores.

In addition to selling merchandise, some types of store retailers are also engaged in the provision of after-sales services, such as repair and installation. For example, new automobile dealers, electronic and appliance stores and musical instrument and supplies stores often provide repair services, while floor covering stores and window treatment stores often provide installation services. As a general rule, establishments engaged in retailing merchandise and providing after sales services are classified in this sector. Catalogue sales showrooms, gasoline service stations, and mobile home dealers are treated as store retailers.

2.2. Variables

Sales are defined as the sales of all goods purchased for resale, net of returns and discounts. This includes commission revenue and fees earned from selling goods and services on account of others, such as selling lottery tickets, bus tickets, and phone cards. It also includes parts and labour revenue from repair and maintenance; revenue from rental and leasing of goods and equipment; revenues from services, including food services; sales of goods manufactured as a secondary activity; and the proprietor’s withdrawals, at retail, of goods for personal use. Other revenue from rental of real estate, placement fees, operating subsidies, grants, royalties and franchise fees are excluded.

Trading Location is the physical location(s) in which business activity is conducted in each province and territory, and for which sales are credited or recognized in the financial records of the company. For retailers, this would normally be a store.

Constant Dollars: The value of retail trade is measured in two ways; including the effects of price change on sales and net of the effects of price change. The first measure is referred to as retail trade in current dollars and the latter as retail trade in constant dollars. The method of calculating the current dollar estimate is to aggregate the weighted value of sales for all retail outlets. The method of calculating the constant dollar estimate is to first adjust the sales values to a base year, using the Consumer Price Index, and then sum up the resulting values.

2.3. Classification

The Monthly Retail Trade Survey is based on the definition of retail trade under the NAICS (North American Industry Classification System). NAICS is the agreed upon common framework for the production of comparable statistics by the statistical agencies of Canada, Mexico and the United States. The agreement defines the boundaries of twenty sectors. NAICS is based on a production-oriented, or supply based conceptual framework in that establishments are groups into industries according to similarity in production processes used to produce goods and services.

Estimates appear for 21 industries based on special aggregations of the 2012 North American Industry Classification System (NAICS) industries. The 21 industries are further aggregated to 11 sub-sectors.

Geographically, sales estimates are produced for Canada and each province and territory.

3. Coverage and frames

Statistics Canada’s Business Register ( BR) provides the frame for the Monthly Retail Trade Survey. The BR is a structured list of businesses engaged in the production of goods and services in Canada. It is a centrally maintained database containing detailed descriptions of most business entities operating within Canada. The BR includes all incorporated businesses, with or without employees. For unincorporated businesses, the BR includes all employers with businesses, and businesses with no employees with annual sales that have a Goods and Services Tax (GST) or annual revenue that declares individual taxes.  annual sales greater than $30,000 that have a Goods and Services Tax (GST) account (the BR does not include unincorporated businesses with no employees and with annual sales less than $30,000).

The businesses on the BR are represented by a hierarchical structure with four levels, with the statistical enterprise at the top, followed by the statistical company, the statistical establishment and the statistical location. An enterprise can be linked to one or more statistical companies, a statistical company can be linked to one or more statistical establishments, and a statistical establishment to one or more statistical locations.

The target population for the MRTS consists of all statistical establishments on the BR that are classified to the retail sector using the North American Industry Classification System (NAICS) (approximately 200,000 establishments). The NAICS code range for the retail sector is 441100 to 453999. A statistical establishment is the production entity or the smallest grouping of production entities which: produces a homogeneous set of goods or services; does not cross provincial boundaries; and provides data on the value of output, together with the cost of principal intermediate inputs used, along with the cost and quantity of labour used to produce the output. The production entity is the physical unit where the business operations are carried out. It must have a civic address and dedicated labour.

The exclusions to the target population are ancillary establishments (producers of services in support of the activity of producing goods and services for the market of more than one establishment within the enterprise, and serves as a cost centre or a discretionary expense centre for which data on all its costs including labour and depreciation can be reported by the business), future establishments, establishments with a missing or a zero gross business income (GBI) value on the BR and establishments in the following non-covered NAICS:

  • 4541 (electronic shopping and mail-order houses)
  • 4542 (vending machine operators)
  • 45431 (fuel dealers)
  • 45439 (other direct selling establishments)

4. Sampling

The MRTS sample consists of 10,000 groups of establishments (clusters) classified to the Retail Trade sector selected from the Statistics Canada Business Register. A cluster of establishments is defined as all establishments belonging to a statistical enterprise that are in the same industrial group and geographical region. The MRTS uses a stratified design with simple random sample selection in each stratum. The stratification is done by industry groups (the mainly, but not only four digit level NAICS), and the geographical regions consisting of the provinces and territories, as well as three provincial sub-regions. We further stratify the population by size.

The size measure is created using a combination of independent survey data and three administrative variables: the annual profiled revenue, the GST sales expressed on an annual basis, and the declared tax revenue (T1 or T2). The size strata consist of one take-all (census), at most, two take-some (partially sampled) strata, and one take-none (non-sampled) stratum. Take-none strata serve to reduce respondent burden by excluding the smaller businesses from the surveyed population. These businesses should represent at most ten percent of total sales. Instead of sending questionnaires to these businesses, the estimates are produced through the use of administrative data.

The sample was allocated optimally in order to reach target coefficients of variation at the national, provincial/territorial, industrial, and industrial groups by province/territory levels. The sample was also inflated to compensate for dead, non-responding, and misclassified units.

MRTS is a repeated survey with maximisation of monthly sample overlap. The sample is kept month after month, and every month new units are added (births) to the sample.  MRTS births, i.e., new clusters of establishment(s), are identified every month via the BR’s latest universe. They are stratified according to the same criteria as the initial population. A sample of these births is selected according to the sampling fraction of the stratum to which they belong and is added to the monthly sample. Deaths occur on a monthly basis. A death can be a cluster of establishment(s) that have ceased their activities (out-of-business) or whose major activities are no longer in retail trade (out-of-scope). The status of these businesses is updated on the BR using administrative sources and survey feedback, including feedback from the MRTS. Methods to treat dead units and misclassified units are part of the sample and population update procedures.

5. Questionnaire design

The Monthly Retail Trade Survey incorporates the following sub-surveys:

Monthly Retail Trade Survey - R8

Monthly Retail Trade Survey (with inventories) – R8

Survey of Sales and Inventories of Alcoholic Beverages

The questionnaires collect monthly data on retail sales and the number of trading locations by province or territory and inventories of goods owned and intended for resale from a sample of retailers. The items on the questionnaires have remained unchanged for several years. For the 2004 redesign, the general questionnaires were subject to cosmetic changes only. The questionnaire for Sales and Inventories of Alcoholic Beverages underwent more extensive changes. The modifications were discussed withstakeholders and the respondents were given an opportunity to comment before the new questionnaire was finalized. If further changes are needed to any of the questionnaires, proposed changes would go through a review committee and a field test with respondents and data users to ensure its relevancy.

6. Response and non-response

6.1. Response and non-response

Despite the best efforts of survey managers and operations staff to maximize response in the MRTS, some non-response will occur. For statistical establishments to be classified as responding, the degree of partial response (where an accurate response is obtained for only some of the questions asked a respondent) must meet a minimum threshold level below which the response would be rejected and considered a unit non-response.  In such an instance, the business is classified as not having responded at all.

Non-response has two effects on data: first it introduces bias in estimates when non-respondents differ from respondents in the characteristics measured; and second, it contributes to an increase in the sampling variance of estimates because the effective sample size is reduced from that originally sought.

The degree to which efforts are made to get a response from a non-respondent is based on budget and time constraints, its impact on the overall quality and the risk of non-response bias.

The main method to reduce the impact of non-response at sampling is to inflate the sample size through the use of over-sampling rates that have been determined from similar surveys.

Besides the methods to reduce the impact of non-response at sampling and collection, the non-responses to the survey that do occur are treated through imputation. In order to measure the amount of non-response that occurs each month, various response rates are calculated. For a given reference month, the estimation process is run at least twice (a preliminary and a revised run). Between each run, respondent data can be identified as unusable and imputed values can be corrected through respondent data. As a consequence, response rates are computed following each run of the estimation process.

For the MRTS, two types of rates are calculated (un-weighted and weighted). In order to assess the efficiency of the collection process, un-weighted response rates are calculated. Weighted rates, using the estimation weight and the value for the variable of interest, assess the quality of estimation. Within each of these types of rates, there are distinct rates for units that are surveyed and for units that are only modeled from administrative data that has been extracted from GST files.

To get a better picture of the success of the collection process, two un-weighted rates called the ‘collection results rate’ and the ‘extraction results rate’ are computed. They are computed by dividing the number of respondents by the number of units that we tried to contact or tried to receive extracted data for them. Non-monthly reporters (respondents with special reporting arrangements where they do not report every month but for whom actual data is available in subsequent revisions) are excluded from both the numerator and denominator for the months where no contact is performed.

In summary, the various response rates are calculated as follows:

Weighted rates:

Survey Response rate (estimation) =
Sum of weighted sales of units with response status i / Sum of survey weighted sales

where i = units that have either reported data that will be used in estimation or are converted refusals, or have reported data that has not yet been resolved for estimation.

Admin Response rate (estimation) =
Sum of weighted sales of units with response status ii / Sum of administrative weighted sales

where ii = units that have data that was extracted from administrative files and are usable for estimation.

Total Response rate (estimation) =
Sum of weighted sales of units with response status i or response status ii / Sum of all weighted sales

Un-weighted rates:

Survey Response rate (collection) =
Number of questionnaires with response status iii/ Number of questionnaires with response status iv

where iii = units that have either reported data (unresolved, used or not used for estimation) or are converted refusals.

where iv = all of the above plus units that have refused to respond, units that were not contacted and other types of non-respondent units.

Admin Response rate (extraction) =
Number of questionnaires with response status vi/ Number of questionnaires with response status vii

where vi = in-scope units that have data (either usable or non-usable) that was extracted from administrative files

where vii = all of the above plus units that have refused to report to the administrative data source, units that were not contacted and other types of non-respondent units.

(% of questionnaire collected over all in-scope questionnaires)

Collection Results Rate =
Number of questionnaires with response status iii / Number of questionnaires with response status viii

where iii = same as iii defined above

where viii = same as iv except for the exclusion of units that were contacted because their response is unavailable for a particular month since they are non-monthly reporters.

Extraction Results Rate =
Number of questionnaires with response status ix / Number of questionnaires with response status vii

where ix = same as vi with the addition of extracted units that have been imputed or were out of scope

where vii = same as vii defined above

(% of questionnaires collected over all questionnaire in-scope we tried to collect)

All the above weighted and un-weighted rates are provided at the industrial group, geography and size group level or for any combination of these levels.

Use of Administrative Data

Managing response burden is an ongoing challenge for Statistics Canada. In an attempt to alleviate response burden and survey costs, especially for smaller businesses, the MRTS has reduced the number of simple establishments in the sample that are surveyed directly and instead derives sales data for these establishments from Goods and Service Tax (GST) files using a statistical model. The model accounts for differences between sales and revenue (reported for GST purposes) as well as for the time lag between the survey reference period and the reference period of the GST file.

For more information on the methodology used for modeling sales from administrative data sources, refer to ‘Monthly Retail Trade Survey: Use of Administrative Data’ under ‘Documentation’ of the IMDB.

Table 1 contains the weighted response rates for all industry groups as well as for total retail trade for each province and territory. For more detailed weighted response rates, please contact the Marketing and Dissemination Section at (613) 951-3549, toll free: 1-877-421-3067 or by e-mail at retailinfo@statcan.

6.2. Methods used to reduce non-response at collection

Significant effort is spent trying to minimize non-response during collection. Methods used, among others, are interviewer techniques such as probing and persuasion, repeated re-scheduling and call-backs to obtain the information, and procedures dealing with how to handle non-compliant (refusal) respondents.

If data are unavailable at the time of collection, a respondent's best estimates are also accepted, and are subsequently revised once the actual data become available.

To minimize total non-response for all variables, partial responses are accepted. In addition, questionnaires are customized for the collection of certain variables, such as inventory, so that collection is timed for those months when the data are available.

Finally, to build trust and rapport between the interviewers and respondents, cases are generally assigned to the same interviewer each month. This action establishes a personal relationship between interviewer and respondent, and builds respondent trust.

7. Data collection and capture operations

Collection of the data is performed by Statistics Canada’s Regional Offices.

Table 1: Weighted response rates by NAICS, for all provinces and territories: October 2015
Table summary
This table displays the results of Table 1: Weighted response rates by NAICS Weighted Response Rates (appearing as column headers).
  Weighted Response Rates
NAICS - Canada      
Motor Vehicle and Parts Dealers 92.4 92.7 77.2
Automobile Dealers 94.1 94.3 80.0
New Car Dealers 95.5 95.5 ...
Used Car Dealers 74.3 73.5 80.0
Other Motor Vehicle Dealers 72.5 71.0 79.9
Automotive Parts, Accessories and Tire Stores 82.6 83.9 71.4
Furniture and Home Furnishings Stores 79.7 81.9 60.5
Furniture Stores 77.4 78.5 60.0
Home Furnishings Stores 83.8 89.2 60.8
Electronics and Appliance Stores 81.8 82.2 64.2
Building Material and Garden Equipment Dealers 91.9 92.9 83.6
Food and Beverage Stores 87.1 88.4 69.3
Grocery Stores 90.5 92.0 72.1
Grocery (except Convenience) Stores 92.4 93.7 74.4
Convenience Stores 63.6 64.8 56.6
Specialty Food Stores 64.0 66.7 51.6
Beer, Wine and Liquor Stores 79.8 80.0 69.0
Health and Personal Care Stores 84.6 84.5 86.0
Gasoline Stations 72.7 72.2 82.1
Clothing and Clothing Accessories Stores 84.1 85.5 32.5
Clothing Stores 83.0 84.5 24.9
Shoe Stores 88.3 88.7 54.6
Jewellery, Luggage and Leather Goods Stores 88.0 89.9 57.5
Sporting Goods, Hobby, Book and Music Stores 86.1 90.9 28.5
General Merchandise Stores 99.0 99.2 72.2
Department Stores 100.0 100.0 ...
Other general merchandise stores 98.3 98.6 72.2
Miscellaneous Store Retailers 80.4 82.7 55.0
Total 88.0 88.8 71.2
Regions      
Newfoundland and Labrador 87.3 88.0 64.2
Prince Edward Island 80.5 81.1 40.6
Nova Scotia 90.8 90.8 90.9
New Brunswick 87.6 88.9 62.2
Québec 88.3 89.6 70.2
Ontario 89.4 90.1 73.1
Manitoba 86.5 87.0 58.2
Saskatchewan 88.8 89.7 69.8
Alberta 87.4 88.2 69.3
British Columbia 85.0 85.5 71.8
Yukon Territory 85.6 85.6 ...
Northwest Territories 64.8 64.8 ...
Nunavut 77.2 77.2 ...


Weighted Response Rates

Respondents are sent a questionnaire or are contacted by telephone to obtain their sales and inventory values, as well as to confirm the opening or closing of business trading locations. Collection of the data begins approximately 7 working days after the end of the reference month and continues for the duration of that month.

New entrants to the survey are introduced to the survey via an introductory letter that informs the respondent that a representative of Statistics Canada will be calling. This call is to introduce the respondent to the survey, confirm the respondent's business activity, establish and begin data collection, as well as to answer any questions that the respondent may have.

8. Editing

Data editing is the application of checks to detect missing, invalid or inconsistent entries or to point to data records that are potentially in error. In the survey process for the MRTS, data editing is done at two different time periods.

First of all, editing is done during data collection. Once data are collected via the telephone, or via the receipt of completed mail-in questionnaires, the data are captured using customized data capture applications. All data are subjected to data editing. Edits during data collection are referred to as field edits and generally consist of validity and some simple consistency edits. They are used to detect mistakes made during the interview by the respondent or the interviewer and to identify missing information during collection in order to reduce the need for follow-up later on. Another purpose of the field edits is to clean up responses. In the MRTS, the current month’s responses are edited against the respondent’s previous month’s responses and/or the previous year’s responses for the current month. Field edits are also used to identify problems with data collection procedures and the design of the questionnaire, as well as the need for more interviewer training.

Follow-up with respondents occurs to validate potential erroneous data following any failed preliminary edit check of the data. Once validated, the collected data is regularly transmitted to the head office in Ottawa.

Secondly, editing known as statistical editing is also done after data collection and this is more empirical in nature. Statistical editing is run prior to imputation in order to identify the data that will be used as a basis to impute non-respondents. Large outliers that could disrupt a monthly trend are excluded from trend calculations by the statistical edits. It should be noted that adjustments are not made at this stage to correct the reported outliers.

The first step in the statistical editing is to identify which responses will be subjected to the statistical edit rules. Reported data for the current reference month will go through various edit checks.

The first set of edit checks is based on the Hidiriglou-Berthelot method whereby a ratio of the respondent’s current month data over historical (last month, same month last year) or auxiliary data is analyzed. When the respondent’s ratio differs significantly from ratios of respondents who are similar in terms of industry and/or geography group, the response is deemed an outlier.

The second set of edits consists of an edit known as the share of market edit. With this method, one is able to edit all respondents, even those where historical and auxiliary data is unavailable. The method relies on current month data only. Therefore, within a group of respondents, that are similar in terms of industrial group and/or geography, if the weighted contribution of a respondent to the group’s total is too large, it will be flagged as an outlier.

For edit checks based on the Hidiriglou-Berthelot method, data that are flagged as an outlier will not be included in the imputation models (those based on ratios). Also, data that are flagged as outliers in the share of market edit will not be included in the imputation models where means and medians are calculated to impute for responses that have no historical responses.

In conjunction with the statistical editing after data collection of reported data, there is also error detection done on the extracted GST data. Modeled data based on the GST are also subject to an extensive series of processing steps which thoroughly verify each record that is the basis for the model as well as the record being modeled. Edits are performed at a more aggregate level (industry by geography level) to detect records which deviate from the expected range, either by exhibiting large month-to-month change, or differing significantly from the remaining units. All data which fail these edits are subject to manual inspection and possible corrective action.

9. Imputation

Imputation in the MRTS is the process used to assign replacement values for missing data. This is done by assigning values when they are missing on the record being edited to ensure that estimates are of high quality and that a plausible, internal consistency is created. Due to concerns of response burden, cost and timeliness, it is generally impossible to do all follow-ups with the respondents in order to resolve missing responses. Since it is desirable to produce a complete and consistent microdata file, imputation is used to handle the remaining missing cases.

In the MRTS, imputation is based on historical data or administrative data (GST sales). The appropriate method is selected according to a strategy that is based on whether historical data is available, auxiliary data is available and/or which reference month is being processed.

There are three types of historical imputation methods. The first type is a general trend that uses one historical data source (previous month, data from next month or data from same month previous year). The second type is a regression model where data from previous month and same month, previous year are used simultaneously. The third type uses the historical data as a direct replacement value for a non-respondent. Depending upon the particular reference month, there is an order of preference that exists so that top quality imputation can result. The historical imputation method that was labelled as the third type above is always the last option in the order for each reference month.

The imputation method using administrative data is automatically selected when historical information is unavailable for a non-respondent. Trends are then applied to the administrative data source (monthly size) depending on whether the structure is simple, e.g. enterprises with only one establishment, or the unit has a more complex structure.

10. Estimation

Estimation is a process that approximates unknown population parameters using only part of the population that is included in a sample. Inferences about these unknown parameters are then made, using the sample data and associated survey design. This stage uses Statistics Canada's Generalized Estimation System (GES).

For retail sales, the population is divided into a survey portion (take-all and take-some strata) and a non-survey portion (take-none stratum). From the sample that is drawn from the survey portion, an estimate for the population is determined through the use of a Horvitz-Thompson estimator where responses for sales are weighted by using the inverses of the inclusion probabilities of the sampled units. Such weights (called sampling weights) can be interpreted as the number of times that each sampled unit should be replicated to represent the entire population. The calculated weighted sales values are summed by domain, to produce the total sales estimates by each industrial group / geographic area combination. A domain is defined as the most recent classification values available from the BR for the unit and the survey reference period. These domains may differ from the original sampling strata because units may have changed size, industry or location. Changes in classification are reflected immediately in the estimates and do not accumulate over time. For the non-survey portion, the sales are estimated with statistical models using monthly GST sales.

For more information on the methodology for modeling sales from administrative data sources which also contributes to the estimates of the survey portion, refer to ‘Monthly Retail Survey: Use of Administrative Data’ under ‘Documentation’ of the IMDB.

The measure of precision used for the MRTS to evaluate the quality of a population parameter estimate and to obtain valid inferences is the variance. The variance from the survey portion is derived directly from a stratified simple random sample without replacement.

Sample estimates may differ from the expected value of the estimates. However, since the estimate is based on a probability sample, the variability of the sample estimate with respect to its expected value can be measured. The variance of an estimate is a measure of the precision of the sample estimate and is defined as the average, over all possible samples, of the squared difference of the estimate from its expected value.

11. Revisions and seasonal adjustment

Revisions in the raw data are required to correct known non-sampling errors. These normally include replacing imputed data with reported data, corrections to previously reported data, and estimates for new births that were not known at the time of the original estimates.

Raw data are revised, on a monthly basis, for the month immediately prior to the current reference month being published. That is, when data for December are being published for the first time, there will also be revisions, if necessary, to the raw data for November. In addition, revisions are made once a year, with the initial release of the February data, for all months in the previous years. The purpose is to correct any significant problems that have been found that apply for an extended period. The actual period of revision depends on the nature of the problem identified, but rarely exceeds three years. The revision period can be extended when historical revisions or restratitfication are done.

Retail trade data are seasonally adjusted using the X12-ARIMA method. This consists of extrapolating a year's worth of raw data with the ARIMA model (auto-regressive integrated moving average model), and of seasonally adjusting the raw time series. Finally, the annual totals of the seasonally adjusted series are forced to the annual totals of the original series.

The seasonally adjusted data also need to be revised. In part, they need to reflect the revisions identified for the raw data. Also, the seasonally adjusted estimates are calculated using X-12-ARIMA, and are sensitive to the most recent values reported in the raw data. For this reason, with the release of each month of new data, the seasonally adjusted values for the previous three months are revised.  A seasonally adjusted time series is a time series that has been modified to eliminate the effect of seasonal and calendar influences. For this reason, the seasonally adjusted data allows for more meaningful comparisons of economic conditions from month to month.

Once a year, seasonal adjustments options are reviewed to take into account the most recent data. Revised seasonally adjusted estimates for each month in the previous years are released at the same time as the annual revision to the raw data. The actual period of revision depends on the number years the raw data was revised.

12. Data quality evaluation

The methodology of this survey has been designed to control errors and to reduce their potential effects on estimates. However, the survey results remain subject to errors, of which sampling error is only one component of the total survey error. Sampling error results when observations are made only on a sample and not on the entire population. All other errors arising from the various phases of a survey are referred to as nonsampling errors. For example, these types of errors can occur when a respondent provides incorrect information or does not answer certain questions; when a unit in the target population is omitted or covered more than once; when GST data for records being modeled for a particular month are not representative of the actual record for various reasons; when a unit that is out of scope for the survey is included by mistake or when errors occur in data processing, such as coding or capture errors.

Prior to publication, combined survey results are analyzed for comparability; in general, this includes a detailed review of individual responses (especially for large businesses), general economic conditions and historical trends.

A common measure of data quality for surveys is the coefficient of variation (CV). The coefficient of variation, defined as the standard error divided by the sample estimate, is a measure of precision in relative terms. Since the coefficient of variation is calculated from responses of individual units, it also measures some non-sampling errors.

The formula used to calculate coefficients of variation (CV) as percentages is:

CV (X) = S(X) * 100% / X
where X denotes the estimate and S(X) denotes the standard error of X.

Confidence intervals can be constructed around the estimates using the estimate and the CV. Thus, for our sample, it is possible to state with a given level of confidence that the expected value will fall within the confidence interval constructed around the estimate. For example, if an estimate of $12,000,000 has a CV of 2%, the standard error will be $240,000 (the estimate multiplied by the CV). It can be stated with 68% confidence that the expected values will fall within the interval whose length equals the standard deviation about the estimate, i.e. between $11,760,000 and $12,240,000.

Alternatively, it can be stated with 95% confidence that the expected value will fall within the interval whose length equals two standard deviations about the estimate, i.e. between $11,520,000 and $12,480,000.

Finally, due to the small contribution of the non-survey portion to the total estimates, bias in the non-survey portion has a negligible impact on the CVs. Therefore, the CV from the survey portion is used for the total estimate that is the summation of estimates from the surveyed and non-surveyed portions.

13. 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.

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.

Monthly Retail Trade Survey (MRTS) Data Quality Statement

Objectives, uses and users
Concepts, variables and classifications
Coverage and frames
Sampling
Questionnaire design
Response and non-response
Data collection and capture operations
Editing
Imputation
Estimation
Revisions and seasonal adjustment
Data quality evaluation
Disclosure control

1. Objectives, uses and users

1.1. Objective

The Monthly Retail Trade Survey (MRTS) provides information on the performance of the retail trade sector on a monthly basis, and when combined with other statistics, represents an important indicator of the state of the Canadian economy.

1.2. Uses

The estimates provide a measure of the health and performance of the retail trade sector. Information collected is used to estimate level and monthly trend for retail sales. At the end of each year, the estimates provide a preliminary look at annual retail sales and performance.

1.3. Users

A variety of organizations, sector associations, and levels of government make use of the information. Retailers rely on the survey results to compare their performance against similar types of businesses, as well as for marketing purposes. Retail associations are able to monitor industry performance and promote their retail industries. Investors can monitor industry growth, which can result in better access to investment capital by retailers. Governments are able to understand the role of retailers in the economy, which aids in the development of policies and tax incentives. As an important industry in the Canadian economy, governments are able to better determine the overall health of the economy through the use of the estimates in the calculation of the nation’s Gross Domestic Product (GDP).

2. Concepts, variables and classifications

2.1. Concepts

The retail trade sector comprises establishments primarily engaged in retailing merchandise, generally without transformation, and rendering services incidental to the sale of merchandise.

The retailing process is the final step in the distribution of merchandise; retailers are therefore organized to sell merchandise in small quantities to the general public. This sector comprises two main types of retailers, that is, store and non-store retailers. The MRTS covers only store retailers. Their main characteristics are described below. Store retailers operate fixed point-of-sale locations, located and designed to attract a high volume of walk-in customers. In general, retail stores have extensive displays of merchandise and use mass-media advertising to attract customers. They typically sell merchandise to the general public for personal or household consumption, but some also serve business and institutional clients. These include establishments such as office supplies stores, computer and software stores, gasoline stations, building material dealers, plumbing supplies stores and electrical supplies stores.

In addition to selling merchandise, some types of store retailers are also engaged in the provision of after-sales services, such as repair and installation. For example, new automobile dealers, electronic and appliance stores and musical instrument and supplies stores often provide repair services, while floor covering stores and window treatment stores often provide installation services. As a general rule, establishments engaged in retailing merchandise and providing after sales services are classified in this sector. Catalogue sales showrooms, gasoline service stations, and mobile home dealers are treated as store retailers.

2.2. Variables

Sales are defined as the sales of all goods purchased for resale, net of returns and discounts. This includes commission revenue and fees earned from selling goods and services on account of others, such as selling lottery tickets, bus tickets, and phone cards. It also includes parts and labour revenue from repair and maintenance; revenue from rental and leasing of goods and equipment; revenues from services, including food services; sales of goods manufactured as a secondary activity; and the proprietor’s withdrawals, at retail, of goods for personal use. Other revenue from rental of real estate, placement fees, operating subsidies, grants, royalties and franchise fees are excluded.

Trading Location is the physical location(s) in which business activity is conducted in each province and territory, and for which sales are credited or recognized in the financial records of the company. For retailers, this would normally be a store.

Constant Dollars: The value of retail trade is measured in two ways; including the effects of price change on sales and net of the effects of price change. The first measure is referred to as retail trade in current dollars and the latter as retail trade in constant dollars. The method of calculating the current dollar estimate is to aggregate the weighted value of sales for all retail outlets. The method of calculating the constant dollar estimate is to first adjust the sales values to a base year, using the Consumer Price Index, and then sum up the resulting values.

2.3. Classification

The Monthly Retail Trade Survey is based on the definition of retail trade under the NAICS (North American Industry Classification System). NAICS is the agreed upon common framework for the production of comparable statistics by the statistical agencies of Canada, Mexico and the United States. The agreement defines the boundaries of twenty sectors. NAICS is based on a production-oriented, or supply based conceptual framework in that establishments are groups into industries according to similarity in production processes used to produce goods and services.

Estimates appear for 21 industries based on special aggregations of the 2012 North American Industry Classification System (NAICS) industries. The 21 industries are further aggregated to 11 sub-sectors.

Geographically, sales estimates are produced for Canada and each province and territory.

3. Coverage and frames

Statistics Canada’s Business Register ( BR) provides the frame for the Monthly Retail Trade Survey. The BR is a structured list of businesses engaged in the production of goods and services in Canada. It is a centrally maintained database containing detailed descriptions of most business entities operating within Canada. The BR includes all incorporated businesses, with or without employees. For unincorporated businesses, the BR includes all employers with businesses, and businesses with no employees with annual sales that have a Goods and Services Tax (GST) or annual revenue that declares individual taxes.  annual sales greater than $30,000 that have a Goods and Services Tax (GST) account (the BR does not include unincorporated businesses with no employees and with annual sales less than $30,000).

The businesses on the BR are represented by a hierarchical structure with four levels, with the statistical enterprise at the top, followed by the statistical company, the statistical establishment and the statistical location. An enterprise can be linked to one or more statistical companies, a statistical company can be linked to one or more statistical establishments, and a statistical establishment to one or more statistical locations.

The target population for the MRTS consists of all statistical establishments on the BR that are classified to the retail sector using the North American Industry Classification System (NAICS) (approximately 200,000 establishments). The NAICS code range for the retail sector is 441100 to 453999. A statistical establishment is the production entity or the smallest grouping of production entities which: produces a homogeneous set of goods or services; does not cross provincial boundaries; and provides data on the value of output, together with the cost of principal intermediate inputs used, along with the cost and quantity of labour used to produce the output. The production entity is the physical unit where the business operations are carried out. It must have a civic address and dedicated labour.

The exclusions to the target population are ancillary establishments (producers of services in support of the activity of producing goods and services for the market of more than one establishment within the enterprise, and serves as a cost centre or a discretionary expense centre for which data on all its costs including labour and depreciation can be reported by the business), future establishments, establishments with a missing or a zero gross business income (GBI) value on the BR and establishments in the following non-covered NAICS:

  • 4541 (electronic shopping and mail-order houses)
  • 4542 (vending machine operators)
  • 45431 (fuel dealers)
  • 45439 (other direct selling establishments)

4. Sampling

The MRTS sample consists of 10,000 groups of establishments (clusters) classified to the Retail Trade sector selected from the Statistics Canada Business Register. A cluster of establishments is defined as all establishments belonging to a statistical enterprise that are in the same industrial group and geographical region. The MRTS uses a stratified design with simple random sample selection in each stratum. The stratification is done by industry groups (the mainly, but not only four digit level NAICS), and the geographical regions consisting of the provinces and territories, as well as three provincial sub-regions. We further stratify the population by size.

The size measure is created using a combination of independent survey data and three administrative variables: the annual profiled revenue, the GST sales expressed on an annual basis, and the declared tax revenue (T1 or T2). The size strata consist of one take-all (census), at most, two take-some (partially sampled) strata, and one take-none (non-sampled) stratum. Take-none strata serve to reduce respondent burden by excluding the smaller businesses from the surveyed population. These businesses should represent at most ten percent of total sales. Instead of sending questionnaires to these businesses, the estimates are produced through the use of administrative data.

The sample was allocated optimally in order to reach target coefficients of variation at the national, provincial/territorial, industrial, and industrial groups by province/territory levels. The sample was also inflated to compensate for dead, non-responding, and misclassified units.

MRTS is a repeated survey with maximisation of monthly sample overlap. The sample is kept month after month, and every month new units are added (births) to the sample.  MRTS births, i.e., new clusters of establishment(s), are identified every month via the BR’s latest universe. They are stratified according to the same criteria as the initial population. A sample of these births is selected according to the sampling fraction of the stratum to which they belong and is added to the monthly sample. Deaths occur on a monthly basis. A death can be a cluster of establishment(s) that have ceased their activities (out-of-business) or whose major activities are no longer in retail trade (out-of-scope). The status of these businesses is updated on the BR using administrative sources and survey feedback, including feedback from the MRTS. Methods to treat dead units and misclassified units are part of the sample and population update procedures.

5. Questionnaire design

The Monthly Retail Trade Survey incorporates the following sub-surveys:

Monthly Retail Trade Survey - R8

Monthly Retail Trade Survey (with inventories) – R8

Survey of Sales and Inventories of Alcoholic Beverages

The questionnaires collect monthly data on retail sales and the number of trading locations by province or territory and inventories of goods owned and intended for resale from a sample of retailers. The items on the questionnaires have remained unchanged for several years. For the 2004 redesign, the general questionnaires were subject to cosmetic changes only. The questionnaire for Sales and Inventories of Alcoholic Beverages underwent more extensive changes. The modifications were discussed withstakeholders and the respondents were given an opportunity to comment before the new questionnaire was finalized. If further changes are needed to any of the questionnaires, proposed changes would go through a review committee and a field test with respondents and data users to ensure its relevancy.

6. Response and non-response

6.1. Response and non-response

Despite the best efforts of survey managers and operations staff to maximize response in the MRTS, some non-response will occur. For statistical establishments to be classified as responding, the degree of partial response (where an accurate response is obtained for only some of the questions asked a respondent) must meet a minimum threshold level below which the response would be rejected and considered a unit non-response.  In such an instance, the business is classified as not having responded at all.

Non-response has two effects on data: first it introduces bias in estimates when non-respondents differ from respondents in the characteristics measured; and second, it contributes to an increase in the sampling variance of estimates because the effective sample size is reduced from that originally sought.

The degree to which efforts are made to get a response from a non-respondent is based on budget and time constraints, its impact on the overall quality and the risk of non-response bias.

The main method to reduce the impact of non-response at sampling is to inflate the sample size through the use of over-sampling rates that have been determined from similar surveys.

Besides the methods to reduce the impact of non-response at sampling and collection, the non-responses to the survey that do occur are treated through imputation. In order to measure the amount of non-response that occurs each month, various response rates are calculated. For a given reference month, the estimation process is run at least twice (a preliminary and a revised run). Between each run, respondent data can be identified as unusable and imputed values can be corrected through respondent data. As a consequence, response rates are computed following each run of the estimation process.

For the MRTS, two types of rates are calculated (un-weighted and weighted). In order to assess the efficiency of the collection process, un-weighted response rates are calculated. Weighted rates, using the estimation weight and the value for the variable of interest, assess the quality of estimation. Within each of these types of rates, there are distinct rates for units that are surveyed and for units that are only modeled from administrative data that has been extracted from GST files.

To get a better picture of the success of the collection process, two un-weighted rates called the ‘collection results rate’ and the ‘extraction results rate’ are computed. They are computed by dividing the number of respondents by the number of units that we tried to contact or tried to receive extracted data for them. Non-monthly reporters (respondents with special reporting arrangements where they do not report every month but for whom actual data is available in subsequent revisions) are excluded from both the numerator and denominator for the months where no contact is performed.

In summary, the various response rates are calculated as follows:

Weighted rates:

Survey Response rate (estimation) =
Sum of weighted sales of units with response status i / Sum of survey weighted sales

where i = units that have either reported data that will be used in estimation or are converted refusals, or have reported data that has not yet been resolved for estimation.

Admin Response rate (estimation) =
Sum of weighted sales of units with response status ii / Sum of administrative weighted sales

where ii = units that have data that was extracted from administrative files and are usable for estimation.

Total Response rate (estimation) =
Sum of weighted sales of units with response status i or response status ii / Sum of all weighted sales

Un-weighted rates:

Survey Response rate (collection) =
Number of questionnaires with response status iii/ Number of questionnaires with response status iv

where iii = units that have either reported data (unresolved, used or not used for estimation) or are converted refusals.

where iv = all of the above plus units that have refused to respond, units that were not contacted and other types of non-respondent units.

Admin Response rate (extraction) =
Number of questionnaires with response status vi/ Number of questionnaires with response status vii

where vi = in-scope units that have data (either usable or non-usable) that was extracted from administrative files

where vii = all of the above plus units that have refused to report to the administrative data source, units that were not contacted and other types of non-respondent units.

(% of questionnaire collected over all in-scope questionnaires)

Collection Results Rate =
Number of questionnaires with response status iii / Number of questionnaires with response status viii

where iii = same as iii defined above

where viii = same as iv except for the exclusion of units that were contacted because their response is unavailable for a particular month since they are non-monthly reporters.

Extraction Results Rate =
Number of questionnaires with response status ix / Number of questionnaires with response status vii

where ix = same as vi with the addition of extracted units that have been imputed or were out of scope

where vii = same as vii defined above

(% of questionnaires collected over all questionnaire in-scope we tried to collect)

All the above weighted and un-weighted rates are provided at the industrial group, geography and size group level or for any combination of these levels.

Use of Administrative Data

Managing response burden is an ongoing challenge for Statistics Canada. In an attempt to alleviate response burden and survey costs, especially for smaller businesses, the MRTS has reduced the number of simple establishments in the sample that are surveyed directly and instead derives sales data for these establishments from Goods and Service Tax (GST) files using a statistical model. The model accounts for differences between sales and revenue (reported for GST purposes) as well as for the time lag between the survey reference period and the reference period of the GST file.

For more information on the methodology used for modeling sales from administrative data sources, refer to ‘Monthly Retail Trade Survey: Use of Administrative Data’ under ‘Documentation’ of the IMDB.

Table 1 contains the weighted response rates for all industry groups as well as for total retail trade for each province and territory. For more detailed weighted response rates, please contact the Marketing and Dissemination Section at (613) 951-3549, toll free: 1-877-421-3067 or by e-mail at retailinfo@statcan.

6.2. Methods used to reduce non-response at collection

Significant effort is spent trying to minimize non-response during collection. Methods used, among others, are interviewer techniques such as probing and persuasion, repeated re-scheduling and call-backs to obtain the information, and procedures dealing with how to handle non-compliant (refusal) respondents.

If data are unavailable at the time of collection, a respondent's best estimates are also accepted, and are subsequently revised once the actual data become available.

To minimize total non-response for all variables, partial responses are accepted. In addition, questionnaires are customized for the collection of certain variables, such as inventory, so that collection is timed for those months when the data are available.

Finally, to build trust and rapport between the interviewers and respondents, cases are generally assigned to the same interviewer each month. This action establishes a personal relationship between interviewer and respondent, and builds respondent trust.

7. Data collection and capture operations

Collection of the data is performed by Statistics Canada’s Regional Offices.

Table 1: Weighted response rates by NAICS, for all provinces and territories: February 2015
Table summary
This table displays the results of Table 1: Weighted response rates by NAICS Weighted Response Rates (appearing as column headers).
  Weighted Response Rates
Total Survey Administrative
NAICS - Canada  
Motor Vehicle and Parts Dealers 90.6 91.3 63.8
Automobile Dealers 92.1 92.5 58.3
New Car Dealers 93.5 93.5 Note ...: not applicable
Used Car Dealers 71.4 73.9 58.3
Other Motor Vehicle Dealers 80.7 82.4 72.5
Automotive Parts, Accessories and Tire Stores 75.7 77.7 60.3
Furniture and Home Furnishings Stores 86.6 90.9 53.6
Furniture Stores 92.3 93.9 66.0
Home Furnishings Stores 77.1 84.8 47.6
Electronics and Appliance Stores 72.5 72.9 54.9
Building Material and Garden Equipment Dealers 89.9 93.4 49.6
Food and Beverage Stores 88.5 90.1 68.3
Grocery Stores 92.4 94.1 72.9
Grocery (except Convenience) Stores 94.8 96.1 79.0
Convenience Stores 54.5 59.7 25.0
Specialty Food Stores 65.0 69.0 47.3
Beer, Wine and Liquor Stores 77.3 78.1 35.0
Health and Personal Care Stores 88.3 88.2 89.9
Gasoline Stations 72.4 72.9 63.2
Clothing and Clothing Accessories Stores 87.7 88.9 30.8
Clothing Stores 88.4 89.6 33.0
Shoe Stores 84.9 85.7 16.7
Jewellery, Luggage and Leather Goods Stores 86.0 87.4 24.8
Sporting Goods, Hobby, Book and Music Stores 87.2 90.5 39.2
General Merchandise Stores 97.6 97.7 80.6
Department Stores 100.0 100.0 Note ...: not applicable
Other general merchandise stores 95.7 95.9 80.6
Miscellaneous Store Retailers 59.2 62.9 28.6
Total 87.2 88.4 63.1
Regions  
Newfoundland and Labrador 80.2 81.9 23.5
Prince Edward Island 85.1 86.5 0.0
Nova Scotia 89.2 90.6 50.4
New Brunswick 86.3 88.2 49.5
Québec 86.6 88.4 62.1
Ontario 89.3 90.3 65.6
Manitoba 86.1 86.6 54.4
Saskatchewan 89.2 90.2 62.0
Alberta 86.3 87.1 69.5
British Columbia 84.6 85.6 61.0
Yukon Territory 83.8 83.8 Note ...: not applicable
Northwest Territories 84.7 84.7 Note ...: not applicable
Nunavut 73.4 73.4 Note ...: not applicable

Weighted Response Rates

Respondents are sent a questionnaire or are contacted by telephone to obtain their sales and inventory values, as well as to confirm the opening or closing of business trading locations. Collection of the data begins approximately 7 working days after the end of the reference month and continues for the duration of that month.

New entrants to the survey are introduced to the survey via an introductory letter that informs the respondent that a representative of Statistics Canada will be calling. This call is to introduce the respondent to the survey, confirm the respondent's business activity, establish and begin data collection, as well as to answer any questions that the respondent may have.

8. Editing

Data editing is the application of checks to detect missing, invalid or inconsistent entries or to point to data records that are potentially in error. In the survey process for the MRTS, data editing is done at two different time periods.

First of all, editing is done during data collection. Once data are collected via the telephone, or via the receipt of completed mail-in questionnaires, the data are captured using customized data capture applications. All data are subjected to data editing. Edits during data collection are referred to as field edits and generally consist of validity and some simple consistency edits. They are used to detect mistakes made during the interview by the respondent or the interviewer and to identify missing information during collection in order to reduce the need for follow-up later on. Another purpose of the field edits is to clean up responses. In the MRTS, the current month’s responses are edited against the respondent’s previous month’s responses and/or the previous year’s responses for the current month. Field edits are also used to identify problems with data collection procedures and the design of the questionnaire, as well as the need for more interviewer training.

Follow-up with respondents occurs to validate potential erroneous data following any failed preliminary edit check of the data. Once validated, the collected data is regularly transmitted to the head office in Ottawa.

Secondly, editing known as statistical editing is also done after data collection and this is more empirical in nature. Statistical editing is run prior to imputation in order to identify the data that will be used as a basis to impute non-respondents. Large outliers that could disrupt a monthly trend are excluded from trend calculations by the statistical edits. It should be noted that adjustments are not made at this stage to correct the reported outliers.

The first step in the statistical editing is to identify which responses will be subjected to the statistical edit rules. Reported data for the current reference month will go through various edit checks.

The first set of edit checks is based on the Hidiriglou-Berthelot method whereby a ratio of the respondent’s current month data over historical (last month, same month last year) or auxiliary data is analyzed. When the respondent’s ratio differs significantly from ratios of respondents who are similar in terms of industry and/or geography group, the response is deemed an outlier.

The second set of edits consists of an edit known as the share of market edit. With this method, one is able to edit all respondents, even those where historical and auxiliary data is unavailable. The method relies on current month data only. Therefore, within a group of respondents, that are similar in terms of industrial group and/or geography, if the weighted contribution of a respondent to the group’s total is too large, it will be flagged as an outlier.

For edit checks based on the Hidiriglou-Berthelot method, data that are flagged as an outlier will not be included in the imputation models (those based on ratios). Also, data that are flagged as outliers in the share of market edit will not be included in the imputation models where means and medians are calculated to impute for responses that have no historical responses.

In conjunction with the statistical editing after data collection of reported data, there is also error detection done on the extracted GST data. Modeled data based on the GST are also subject to an extensive series of processing steps which thoroughly verify each record that is the basis for the model as well as the record being modeled. Edits are performed at a more aggregate level (industry by geography level) to detect records which deviate from the expected range, either by exhibiting large month-to-month change, or differing significantly from the remaining units. All data which fail these edits are subject to manual inspection and possible corrective action.

9. Imputation

Imputation in the MRTS is the process used to assign replacement values for missing data. This is done by assigning values when they are missing on the record being edited to ensure that estimates are of high quality and that a plausible, internal consistency is created. Due to concerns of response burden, cost and timeliness, it is generally impossible to do all follow-ups with the respondents in order to resolve missing responses. Since it is desirable to produce a complete and consistent microdata file, imputation is used to handle the remaining missing cases.

In the MRTS, imputation is based on historical data or administrative data (GST sales). The appropriate method is selected according to a strategy that is based on whether historical data is available, auxiliary data is available and/or which reference month is being processed.

There are three types of historical imputation methods. The first type is a general trend that uses one historical data source (previous month, data from next month or data from same month previous year). The second type is a regression model where data from previous month and same month, previous year are used simultaneously. The third type uses the historical data as a direct replacement value for a non-respondent. Depending upon the particular reference month, there is an order of preference that exists so that top quality imputation can result. The historical imputation method that was labelled as the third type above is always the last option in the order for each reference month.

The imputation method using administrative data is automatically selected when historical information is unavailable for a non-respondent. Trends are then applied to the administrative data source (monthly size) depending on whether the structure is simple, e.g. enterprises with only one establishment, or the unit has a more complex structure.

10. Estimation

Estimation is a process that approximates unknown population parameters using only part of the population that is included in a sample. Inferences about these unknown parameters are then made, using the sample data and associated survey design. This stage uses Statistics Canada's Generalized Estimation System (GES).

For retail sales, the population is divided into a survey portion (take-all and take-some strata) and a non-survey portion (take-none stratum). From the sample that is drawn from the survey portion, an estimate for the population is determined through the use of a Horvitz-Thompson estimator where responses for sales are weighted by using the inverses of the inclusion probabilities of the sampled units. Such weights (called sampling weights) can be interpreted as the number of times that each sampled unit should be replicated to represent the entire population. The calculated weighted sales values are summed by domain, to produce the total sales estimates by each industrial group / geographic area combination. A domain is defined as the most recent classification values available from the BR for the unit and the survey reference period. These domains may differ from the original sampling strata because units may have changed size, industry or location. Changes in classification are reflected immediately in the estimates and do not accumulate over time. For the non-survey portion, the sales are estimated with statistical models using monthly GST sales.

For more information on the methodology for modeling sales from administrative data sources which also contributes to the estimates of the survey portion, refer to ‘Monthly Retail Survey: Use of Administrative Data’ under ‘Documentation’ of the IMDB.

The measure of precision used for the MRTS to evaluate the quality of a population parameter estimate and to obtain valid inferences is the variance. The variance from the survey portion is derived directly from a stratified simple random sample without replacement.

Sample estimates may differ from the expected value of the estimates. However, since the estimate is based on a probability sample, the variability of the sample estimate with respect to its expected value can be measured. The variance of an estimate is a measure of the precision of the sample estimate and is defined as the average, over all possible samples, of the squared difference of the estimate from its expected value.

11. Revisions and seasonal adjustment

Revisions in the raw data are required to correct known non-sampling errors. These normally include replacing imputed data with reported data, corrections to previously reported data, and estimates for new births that were not known at the time of the original estimates.

Raw data are revised, on a monthly basis, for the month immediately prior to the current reference month being published. That is, when data for December are being published for the first time, there will also be revisions, if necessary, to the raw data for November. In addition, revisions are made once a year, with the initial release of the February data, for all months in the previous years. The purpose is to correct any significant problems that have been found that apply for an extended period. The actual period of revision depends on the nature of the problem identified, but rarely exceeds three years. The revision period can be extended when historical revisions or restratitfication are done.

Retail trade data are seasonally adjusted using the X12-ARIMA method. This consists of extrapolating a year's worth of raw data with the ARIMA model (auto-regressive integrated moving average model), and of seasonally adjusting the raw time series. Finally, the annual totals of the seasonally adjusted series are forced to the annual totals of the original series.

The seasonally adjusted data also need to be revised. In part, they need to reflect the revisions identified for the raw data. Also, the seasonally adjusted estimates are calculated using X-12-ARIMA, and are sensitive to the most recent values reported in the raw data. For this reason, with the release of each month of new data, the seasonally adjusted values for the previous three months are revised.  A seasonally adjusted time series is a time series that has been modified to eliminate the effect of seasonal and calendar influences. For this reason, the seasonally adjusted data allows for more meaningful comparisons of economic conditions from month to month.

Once a year, seasonal adjustments options are reviewed to take into account the most recent data. Revised seasonally adjusted estimates for each month in the previous years are released at the same time as the annual revision to the raw data. The actual period of revision depends on the number years the raw data was revised.

12. Data quality evaluation

The methodology of this survey has been designed to control errors and to reduce their potential effects on estimates. However, the survey results remain subject to errors, of which sampling error is only one component of the total survey error. Sampling error results when observations are made only on a sample and not on the entire population. All other errors arising from the various phases of a survey are referred to as nonsampling errors. For example, these types of errors can occur when a respondent provides incorrect information or does not answer certain questions; when a unit in the target population is omitted or covered more than once; when GST data for records being modeled for a particular month are not representative of the actual record for various reasons; when a unit that is out of scope for the survey is included by mistake or when errors occur in data processing, such as coding or capture errors.

Prior to publication, combined survey results are analyzed for comparability; in general, this includes a detailed review of individual responses (especially for large businesses), general economic conditions and historical trends.

A common measure of data quality for surveys is the coefficient of variation (CV). The coefficient of variation, defined as the standard error divided by the sample estimate, is a measure of precision in relative terms. Since the coefficient of variation is calculated from responses of individual units, it also measures some non-sampling errors.

The formula used to calculate coefficients of variation (CV) as percentages is:

CV (X) = S(X) * 100% / X
where X denotes the estimate and S(X) denotes the standard error of X.

Confidence intervals can be constructed around the estimates using the estimate and the CV. Thus, for our sample, it is possible to state with a given level of confidence that the expected value will fall within the confidence interval constructed around the estimate. For example, if an estimate of $12,000,000 has a CV of 2%, the standard error will be $240,000 (the estimate multiplied by the CV). It can be stated with 68% confidence that the expected values will fall within the interval whose length equals the standard deviation about the estimate, i.e. between $11,760,000 and $12,240,000.

Alternatively, it can be stated with 95% confidence that the expected value will fall within the interval whose length equals two standard deviations about the estimate, i.e. between $11,520,000 and $12,480,000.

Finally, due to the small contribution of the non-survey portion to the total estimates, bias in the non-survey portion has a negligible impact on the CVs. Therefore, the CV from the survey portion is used for the total estimate that is the summation of estimates from the surveyed and non-surveyed portions.

13. 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.

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.

The Daily – Analysis Component

Archived information

Archived information is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

Consultation objectives

In May 2014, Statistics Canada invited key stakeholders of its official release bulletin, The Daily, to moderated group discussions aimed at improving the analytical content of The Daily.

The Daily has been the official voice of Statistics Canada since 1932. Every new product or dataset must be announced to the public on the website in The Daily.

The agency is currently evaluating a range of possible changes to its Daily articles. These changes are designed to improve the analytical content of Daily releases by providing users with additional information and analysis that can aid in the interpretation of data.

Consultation methodology

Statistics Canada held moderated group discussions with its users during which participants were asked a series of feedback questions on the analytical content of The Daily.

How to get involved

The consultations are now closed.

Individuals who wish to obtain more information or to take part in future consultations may contact Statistics Canada by sending an email to consultations@statcan.gc.ca.

Please note that Statistics Canada selects participants for each consultation to ensure feedback is sought from a representative sample of the target population for the study. Not all applicants will be asked to participate in a given consultation.

Statistics Canada is committed to respecting the privacy of consultation participants. All personal information created, held or collected by the Agency is protected by the Privacy Act. For more information on Statistics Canada's privacy policies, please consult the Privacy notice.

Results

Most users do not want The Daily to undergo major changes, only enhancements.

What worked

Participants liked the summaries, charts and links. They found the structure to be logical, the articles well written and the seasonally adjusted data clearly identified.

Areas for improvement

Participants would like to have text for every release. Data availability announcements, which contain neither data nor information, require users to search the website or to call Statistics Canada for information.

For certain subjects, atypical data, and trend cycles, participants would like to see more detailed, contextual analysis. They would also like to see topical studies on the effects of external events, such as floods.

Overall, participants were satisfied with the layout of The Daily landing page, but indicated that the order, including the 'New products and studies' section, could be made clearer. Charts were not always easy to interpret or referred to in the text.

Participants would like to see links to more detailed information in Daily texts, including information sources, full-length articles, and in-depth analysis. Participants also found the presentation of and links to CANSIM data were not intuitive. A more comprehensive list of table numbers and titles was suggested.

Recommendations

  • Include text or a data table with every Daily release.
  • Provide more detailed, contextual analysis with hyperlinks in the text to relevant methodological information.
  • Include more topical studies.
  • Refer to charts in the text and add source information (or links) to the charts.
  • Provide CANSIM table titles with the table numbers.
  • Increase awareness of The Daily features within the new My StatCan module.

Statistics Canada thanks participants for their participation in this consultation. Their insights guide the Agency's web development and ensure that the final products meet users' expectations.

Date modified:

Statistics Canada’s The Daily Readership Survey

Archived information

Archived information is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

Consultation objectives

In January 2014, Statistics Canada invited readers of its official release bulletin, The Daily, to complete a survey. The feedback provided will help improve how the agency produces and delivers The Daily.

Consultation methodology

Participants were asked to complete a short, anonymous questionnaire. The survey was available online and email invitations were sent to My Account and Upcoming Releases subscribers, Website Evaluation Survey participants who use The Daily, as well as to a list of journalists. Links to access the survey were available on the Consulting Canadians page, on The Daily and on the Media room pages.

A total of 374 visitors participated over a 21-day period from January 6 to 27, 2014.

How to get involved

This consultation is now closed.

Individuals who wish to obtain more information or to take part in a consultation should contact Statistics Canada by sending an email to consultations@statcan.gc.ca.Please note that Statistics Canada selects participants for each consultation to ensure feedback is sought from a representative sample of the target population for the study. Not all applicants will be asked to participate in a given consultation.

Please note that Statistics Canada selects participants for each consultation to ensure feedback is sought from a representative sample of the target population for the study. Not all applicants will be asked to participate in a given consultation.

Statistics Canada is committed to respecting the privacy of consultation participants. All personal information created, held or collected by the Agency is protected by the Privacy Act. For more information on Statistics Canada's privacy policies, please consult the Privacy notice.

Results

Overall satisfaction

In total, 81% of respondents were satisfied with The Daily.

Information sought and planned use

Consultation results show that 42% of respondents’ primary purpose for reading The Daily was for research for work or school and 36% wanted to stay generally informed about the economy and society.

Among participants, 44% read The Daily primarily for statistics and indicators, 35% for articles and studies and 11% for data tables and charts. Overall, 70% of participants reported that it was either easy or very easy to find what they were looking for in The Daily.

Areas of improvement

Ease of access to data (navigation and search) and website content (information availability) remain areas for improvement.

Consultation participant profile

Participants came principally from three main sectors:  federal government (20%), media (20%) and provincial government (19%). Among participants, 75% were frequent users who read The Daily at least once per week.

Statistics Canada thanks participants for their participation in this consultation. Their insights guide the agency’s web development and ensure that the final products meet users’ expectations.

Date modified:

Highlights and analysis

Archived information

Archived information is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

Get to know Canadian farmers and their families

This report, the final data release from the 2011 Census of Agriculture, provides a socio-economic profile of Canada's farm population. It was produced by linking data from the 2011 Census of Agriculture with data from the 2011 National Household Survey and the 2011 Census of Population. Until 2006, the population information came from the Census of Population's long-form questionnaire, which was a mandatory questionnaire distributed to 20% of the Canadian households. In 2011, the data source was the voluntary National Household Survey, which was distributed to approximately 33% of Canadian households.  Data users should be aware of changes to these two databases when doing comparisons of results between the 2011 Agriculture–National Household Survey (Ag-NHS) Linkage database and previous Agriculture–Population Linkage databases. Please refer to the Quality Statement about the Agriculture–National Household Survey linkage for more details.

This 2011 Agriculture–National Household Survey linkage provides socio-economic data for Canada's farm population, farm operators, and farm families. Canada's farm population is defined as farm operators (individuals involved in the day-to-day management decisions in operating a census farmFootnote 1) as well as the individuals in their households. There could be up to three farm operators reported for each farm on the 2011 Census of Agriculture. The 2011 Ag-NHS database does not include the farm population in the Territories or those that were residents of collective dwellingsFootnote 2 on May 10, 2011. In this report, the four Atlantic provinces have been combined. In total, 205,170 farms were included in 2011 Ag-NHS linkage, representing 99.7% of all farms enumerated in Canada in the 2011 Census of Agriculture.

 

Insight into Canada's farm population

The Canadian farm population totalled 650,395 in 2011, accounting for 1 out of every 50 Canadians, or 2.0% of the total population of CanadaFootnote 3. The farm population in Ontario was the highest among the provinces at 174,905, accounting for 1.4% of Ontario's total population. At 103,885, the farm population in Saskatchewan made up 10.3% of its total population, the largest proportion out of all the provinces (Table 1).

Farm population refers to farm operators and the individuals in their households.

 

This table displays the results of table 1 the farm population as a percentage of the total population in canada farm population, total population and farm population as a percentage of the total population (appearing as column headers).

Table 1 The farm population as a percentage of the total population in Canada, 2011
  Farm population Total population Farm population as a percentage of the total population
Canada 650,395 32,746,505 2
Atlantic provinces 26,310 2,286,650 1.2
Quebec 101,675 7,732,520 1.3
Ontario 174,905 12,651,790 1.4
Manitoba 49,155 1,174,345 4.2
Saskatchewan 103,885 1,008,760 10.3
Alberta 129,810 3,567,975 3.6
British Columbia 64,650 4,324,455 1.5
Note: The total population was sourced from the NHS to be comparable with the Ag-NHS since both exclude collective dwellings.
Source: Statistics Canada, Agriculture–NHS Linkage Database, 2011 and National Household Survey, 2011.

On a national basis, Ontario was home to over a quarter (26.9%) of Canada's farm population. Another 20.0% were in Alberta, 16.0% in Saskatchewan and 15.6% in Quebec. Together, the Atlantic provinces made up 4.0% of Canada's farm population (Chart 1).

Chart 1 Distribution of the farm population as a proportion of all farms in Canada, 2011

Text description of image

Approximately 45.0% of the farm population were farm operators, while the other 55.0% were other members of their households.

Age distribution

In general, the farm population was older than the total populationFootnote 4. Although those under 35 years old made up the largest portion (38.5%) of the farm population, the proportion in the total population (43.2%) was higher (Chart 2). Nearly a third of both the farm population and the total population were between the ages of 35 years and 54 years (29.9% and 29.7% respectively). Those aged 55 years or older made up a slightly higher percentage of the farm population (31.6%) compared to the total population (27.1%).

This broader farm population consists of farm operators and the other people in their households. Nearly half (48.2%) of farm operators were aged 55 or older in 2011, while about 1 in 12 (8.2%) were under 35 years old. Comparatively, in the self-employed labour force about one-third (33.2%) were 55 or older and about 1 in 6 (15.6%) were under 35.

Chart 2 Age distribution of farm operators, the self-employed labour force, the farm population and the total population in Canada, 2011

Text description of image

The farm populations in British Columbia and the Atlantic provinces had the largest proportion of individuals aged 55 years or older (35.7% and 36.0% respectively). The farm populations in Quebec and Ontario, in turn, were largely comprised of younger individuals, with 42.3% of the farm population in Quebec and 40.8% of the farm population in Ontario under the age of 35 years.

Approximately 1 in 10 (9.6%) of those aged under 35 in the farm population was a farm operator, with the remaining 90.4% comprising other members of the household (Chart 3). Conversely, the majority of the farm population aged 35 years or older were farm operators (67.1%).

Chart 3 Proportion of the farm population that were operators or other household members by age category in Canada, 2011

Text description of image

Mother tongue

The most common mother tongue in the Canadian farm population was English (72.3%), followed by French (17.0%), German (4.7%) and Dutch (1.8%) (Table 2).

In the total population, English and French were the most frequently reported mother tongues (57.4% and 21.3% respectively), followed by Panjabi (Punjabi) (1.3%) and Chinese (1.2%).

Mother tongue is defined as the first language a person learned at home in childhood and still understood at the time of the 2011 NHS.

 

Within the farm population, English was the predominant mother tongue for every province except Quebec, where French was the dominant language, followed by English. In British Columbia, Punjabi was the second most common mother tongue, spoken by 8.8% of the farm population. In the other provinces, the second most common mother tongues were either German or French.

This table displays the results of table 2 mother tongues in the farm population in canada. The information is grouped by rank (appearing as row headers), mother tongue, farm population and percentage, calculated using all languages, 650,395 , 100.0, all others (including multiple responses), 12,205 and 1.9 units of measure (appearing as column headers).

Table 2 Mother tongues in the farm population in Canada, 2011
Rank Mother tongue Farm population Percentage
  All languages 650,395 100.0
1 English 470,410 72.3
2 French 110,260 17.0
3 German 30,800 4.7
4 Dutch 11,970 1.8
5 Panjabi (Punjabi) 5,770 0.9
6 Ukrainian 5,045 0.8
7 Italian 1,310 0.2
8 Polish 1,140 0.2
9 Hungarian 835 0.1
10 Portuguese 650 0.1
  All others (including multiple responses) 12,205 1.9
Source: Statistics Canada, Agriculture–NHS Linkage Database, 2011.

Residence

Of the 650,395 people in the farm population, the majority (90.0%) reported living in rural areas, while the remaining 10.0% lived in population centres.

In sharp contrast, the majority (81.2%) of the total population of Canada lived in population centres, while the remaining 18.8% lived in rural areas. Close to 1 in 10 (9.5%) of this rural population was part of the farm population.

Rural areas are classified as all areas outside population centres. Population centres refer to areas with a population of at least 1,000 and a population density of 400 persons or more per square kilometre, based on the current census population count. Population centres are further classified as small population centres with a population between 1,000 and 29,999, medium population centres with a population between 30,000 and 99,999, and large urban population centres with a population of 100,000 or more.

 

Compared to the other provinces, Ontario (93.4%), Manitoba (92.9%) and Alberta (92.1%) had the largest proportions of the farm population living in rural areas. In contrast, British Columbia had the smallest proportion of the farm population living in rural areas (77.1%).

Among the total population, the largest proportions living in rural areas were in the Atlantic provinces (44.9%) and Saskatchewan (32.8%). Similar to the farm population, British Columbia had the smallest proportion of its total population living in rural areas (13.9%).

In total, the rural farm population represented 9.5% of the total rural population in Canada (Table 3). This proportion varied across the country, with the largest proportion in Saskatchewan (27.8%) and the smallest in the Atlantic provinces (2.4%).

This table displays the results of table 3 rural farm population as a percentage of the total rural population in canada rural farm population, total rural population and rural farm population as a percentage of total rural population (appearing as column headers).

Table 3 Rural farm population as a percentage of the total rural population in Canada, 2011
  Rural farm population Total rural population Rural farm population as a percentage of total rural population
Canada 585,180 6,151,880 9.5
Atlantic provinces 24,155 1,027,600 2.4
Quebec 90,735 1,511,525 6.0
Ontario 163,435 1,775,670 9.2
Manitoba 45,660 317,310 14.4
Saskatchewan 91,785 330,540 27.8
Alberta 119,570 588,915 20.3
British Columbia 49,840 600,305 8.3
Source: Statistics Canada, Agriculture–NHS Linkage Database, 2011 and the National Household Survey, 2011.

Among the farm population living in population centres, most lived in small population centres (43.4%), followed closely by large population centres (42.3%). In comparison, among the total population living in population centres, most lived in large population centres (74.3%), followed by small population centres (15.0%).

In total, 22.9% of the farm population in British Columbia lived in population centres, the largest proportion among the provinces, followed by 11.6% in Saskatchewan and 10.8% in Quebec.  Among the farm population in British Columbia living in population centres, the majority lived in large population centres (67.4%).  Conversely, in Saskatchewan the largest proportion lived in small population centres (62.4%).

Average household size

In 2011, the average household size in the farm population was 2.9 persons. This compares to an average of 2.5 persons in private households in the total population.

The average farm household size ranges across the country from a high of 3.1 in Ontario to a low of 2.7 in Saskatchewan. In the total population, the average household size ranges from a high of 2.6 in Ontario and Alberta to a low of 2.3 in Quebec.

Larger households were more common in the rural farm population than in the total rural population. The proportion of four-person households, five-person households and six-or-more-person households was larger in the rural farm population than in the total rural population (Chart 4). The proportion of one-person households in the rural farm population was smaller than in the total rural population.

Chart 4 Household size of the rural farm population and the total rural population in Canada, 2011

Text description of image

Religious affiliation

In 2011, 82.6% of the farm population reported a religious affiliation. This was higher than the total population, in which 76.1% reported a religion affiliation.

Of those reporting a religious affiliation in the farm population, the most frequently reported religions were Roman Catholic, the United Church and Lutheran. For the total population reporting a religion affiliation, the most frequently reported were Roman Catholic, the United Church and Anglican. Some groups, like the Mennonites, were more present in the farm population than in the total population (Table 4).

This table displays the results of farm population 2011 farm population 2011, total population 2011 and number of persons, calculated using number of persons and percentage units of measure (appearing as column headers).

Table 4 Reported religious affiliations of the farm population and the total population in Canada, 2011
  Farm population 2011 Total population 2011
Number of persons Percentage Number of persons Percentage
All persons that reported religious affiliations 537,320 100 24,929,085 100
Roman Catholic 206,905 38.5 12,699,430 50.9
United Church 100,830 18.8 2,004,845 8
Lutheran 30,085 5.6 477,405 1.9
Christian Footnote 5 29,470 5.5 1,470,310 5.9
Anglican 29,065 5.4 1,608,990 6.5
Mennonite 28,270 5.3 175,780 0.7
Baptist 17,220 3.2 634,450 2.5
Presbyterian 16,080 3 472,065 1.9
ProtestantFootnote 6 13,015 2.4 549,190 2.2
Christian Reformed Church 8,255 1.5 65,395 0.3
All other religions 58,100 10.8 4,771,270 19.1

Footnotes

Footnote 5

Christian, not included elsewhere, includes those who report only 'Christian' and Christian religions not included elsewhere such as Born Again Christian, Apostolic not included elsewhere, Messianic Jew, Hutterite, etc.

Return to footnote 5 referrer

Footnote 6

Not otherwise specified.

Return to first footnote 6 referrer

Source: Statistics Canada, Agriculture–NHS Linkage Database, 2011 and the National Household Survey, 2011.

For the farm population reporting a religious affiliation, Roman Catholic was the most frequently reported religion for each province except for Manitoba where it was the United Church. For the total population reporting a religious affiliation, Roman Catholic was the most frequently reported in every province.

In 2011, 17.4% of the farm population in Canada, accounting for 113,075 persons, reported no religious affiliation, compared to 23.9% of the total population, or 7,817,420 persons.

Among the provinces, the farm populations of British Columbia and Alberta most frequently reported no religious affiliation (37.4% and 23.4% respectively). The same trend was apparent in the total population, where these same provinces most frequently reported no religious affiliation (44.1% and 31.6%).

Income of Canadian farm families

The median total income for economic families in the farm population was $74,604 in 2010. In comparison, the median total income for economic families in the total population was $76,458. The 197,045 farm families represent 2.1% of the economic families in Canada. Economic families do not include one-person households, which represented 26,205 households or 11.6% of all farm households in 2011.

Farm economic family or farm family refer to an economic family where at least one person is a farm operator. The two terms are used interchangeably in this report. Economic family refers to a group of two or more persons who live in the same dwelling and are related to each other by blood, marriage, common-law, adoption or a foster relationship.

 

Chart 5 Median income for farm economic families and all economic families in Canada, 2010

Text description of image

The median total income for farm economic families varied across the provinces, ranging from $62,641 in Quebec to $80,928 in Alberta (Chart 5).  In the total population, the median income for all economic families ranged from $66,502 in the Atlantic provinces to $93,393 in Alberta.  In most provinces, the median income for farm families was lower than for economic families in the total population.

The median income of farm economic families in rural areas was $72,700, compared to a median income of $93,590 for those in population centres. The median income for farm families in rural areas was lower than for families in population centres in each province.

In rural areas, the median income of farm families was $72,700 compared to $71,422 for all economic families in Canada. In all provinces except for Quebec, Ontario and Alberta, the median income for rural farm families was higher than for all rural economic families (Chart 6).

Chart 6 Median income of farm economic families and all economic families in rural areas in Canada, 2010

Text description of image

Median income by farm type

Poultry and egg production represented 2.1% of farm families in 2011 and had the highest median income in 2010 compared to other types of farming, at $90,250 (Table 5). The second highest median income, at $82,473, was for the 3.9% of farm families in "greenhouse, nursery and floriculture production". Nearly one-third (29.7%) of farm families were involved in the most common type of farming, "oilseed and grain farming", which had the third highest median income ($80,865). Families involved in "dairy cattle and milk production" represented 7.9% of farm families and had the lowest median income of $65,010. This may be due to a number of factors, including that the "dairy cattle and milk production" farm type also had the lowest percentage of operators working off the farm in 2011.

Farm types are based on the North American Industrial Classification System (NAICS) provides a framework for classifying farms based on the commodities produced and the commodity value.

 

This table displays the results of table 5 median income of farm economic families by naics farm type in canada. The information is grouped by type of farming (by naics) (appearing as row headers), median total income for farm families (2010 dollars) and percent of farm families (appearing as column headers).

Table 5 Median income of farm economic families by NAICS farm type in Canada, 2010
Type of farming (by NAICS) Median total Income for farm families (2010 dollars) Percent of farm families
All types of farming 74,604 100.0
Poultry and egg production 90,250 2.1
Greenhouse, nursery and floriculture production 82,473 3.9
Oilseed and grain farming 80,865 29.7
Fruit and tree nut farming 80,505 4.1
Other animal production 77,587 11.3
Vegetable and melon farming 76,608 2.6
Other crop farming 71,544 17.3
Hog and pig farming 68,594 1.7
Sheep and goat farming 67,612 1.9
Beef cattle ranching & farming including feedlots 66,873 17.6
Dairy cattle and milk production 65,010 7.9
Source: Statistics Canada, Agriculture–NHS Linkage Database, 2011.

A closer look at operators of Canadian farms

Of the 292,795 farm operators in the Ag-NHS database, there were 212,185 (72.5%) male operators and 80,605 (27.5%) female operators in Canada's provinces (Table 6). Similarly, in the total population, the majority of the self-employed labour force was male in 2011 (64.0% male and 36.0% female) according to the 2011 NHS.

Compared with the other provinces, British Columbia had the largest proportion of female farm operators (36.5%), followed by Alberta (29.2%) and Ontario (28.4%).

Farm operators refer to persons involved in the day to day management decisions of a census farm that were aged 15 years or older. Up to three farm operators could be reported per farm.

The self-employed labour force refers to persons that were in the labour force and self-employed with or without a business as well as unpaid family workers that were 15 years or older at the time of the 2011 National Household Survey.

 

This table displays the results of table 6 proportion of farm operators and the total self-employed labour force by gender in canada farm operators (%) and self-employed labour force (%), calculated using male, female and total units of measure (appearing as column headers).

Table 6 Proportion of farm operators and the total self-employed labour force by gender in Canada, 2011
  Farm operators (%) Self-employed labour force (%)
Male Female Total Male Female Total
Canada 72.5 27.5 100 64 36 100
Atlantic provinces 77.4 22.6 100 64.2 35.8 100
Quebec 74.1 25.9 100 63.4 36.6 100
Ontario 71.6 28.4 100 64.3 35.7 100
Manitoba 76.1 23.9 100 67.3 32.7 100
Saskatchewan 77 23 100 66.7 33.3 100
Alberta 70.8 29.2 100 64.2 35.8 100
British Columbia 63.5 36.5 100 62.7 37.3 100
Source: Statistics Canada, Agriculture–NHS Linkage Database, 2011 and the National Household Survey, 2011.

Age distribution of farm operators

In general, farm operators were older than the total self-employed labour force. Nearly half of farm operators were 55 years or older (48.2%) in 2011. In comparison, 1 in 3 people in the total self-employed labour force was aged 55 years or older (33.2%), according to the 2011 NHS.

British Columbia also had the largest proportion of farm operators aged 55 years or older among the provinces. In comparison, Saskatchewan had the largest proportion of the total self-employed labour force aged 55 years or older. Among the provinces, Quebec had the smallest proportion of farm operators aged 55 years or older. This was also the case for the total self-employed labour force, where Quebec also had the smallest proportion (Table 7).

This table displays the results of table 7 proportion of farm operators and the total self-employed labour force by age category in canada farm operators (%) and self-employed labour force (%), calculated using under 35 years, 35 to 54 years, 55 years or older and all ages units of measure (appearing as column headers).

Table 7 Proportion of farm operators and the total self-employed labour force by age category in Canada, 2011
  Farm operators (%) Self-employed labour force (%)
Under 35 years 35 to 54 years 55 years or older All ages Under 35 years 35 to 54 years 55 years or older All ages
Canada 8.2 43.5 48.2 100 15.6 51.2 33.2 100
Atlantic provinces 6.1 41 52.9 100 13.1 51 35.9 100
Quebec 10.9 49.4 39.7 100 16.9 53.4 29.7 100
Ontario 8.2 42.5 49.3 100 15.2 51.9 33 100
Manitoba 8.8 45.9 45.3 100 14.8 46.9 38.3 100
Saskatchewan 8.9 41.8 49.3 100 15.9 41.5 42.6 100
Alberta 7.4 43.1 49.6 100 17 49.4 33.6 100
British Columbia 5.4 40.5 54.1 100 14.7 51.6 33.7 100
Source: Statistics Canada, Agriculture–NHS Linkage Database, 2011 and the National Household Survey, 2011.

Female farm operators were proportionally older than females in the total self-employed labour force. Nationally, there were 45.3% of female farm operators 55 years or older, compared to 29.1% of females in the total self-employed labour force. Male farm operators were also older than males in the total self-employed labour force, with 49.4% of male operators aged 55 years or older, compared to 35.5% of males in the total self-employed labour force.

Median age of farm operators

According to the 2011 Ag-NHS, the median age of farm operators was 54. The median age of the total self-employed labour force was 49.

The median age of male farm operators was 54, while the median age of female operators was 53. In the total self-employed labour force, the median age of males (50 years) was older than the median age for females (48 years).

The median age is the point where exactly one half of the population is older and the other half is younger.

 

Among the provinces, the lowest median age for farm operators was 52 years in Quebec. In British Columbia, the median age of operators was 56 years—the highest of the provinces.

In each of the provinces, the median age of female operators was lower than for male operators except in Saskatchewan where the median age for female and male operators was the same.

Operators on multi-generational farms

One in seven operators was on a multi-generational farm (14.4%). Among the provinces, Quebec had the largest proportion of operators on multi-generational farms (20.3%), while Alberta had the lowest proportion (12.3%).

A multi-generational farm includes at least two operators with an age difference of 20 years or more.

 

Highest level of education

In 2011, the majority of farm operators had completed at least a secondary school education (78.3%). The remaining 21.7% of farm operators and 12.5% of the total self-employed labour force reported no certificate, diploma or degree in the 2011 NHS.

A larger proportion of farm operators than of the total self-employed labour force reported their highest level of education attainment as a secondary school certificate, trades certificate, or college diploma (Chart 7).  In contrast, approximately 1 in 6 farm operators reported university credentials as their highest level of education attainment, compared to one-third of the total self-employed labour force.

Among the farm operators, the proportion that attained a secondary school education or higher was larger for female operators (86.3%) than for male operators (75.3%).  A trades certificate (including apprenticeship) was the highest education attainment for 17.4% of male operators and 9.3% of female operators. Over a quarter, 26.4%, of all the female operators reported a college diploma as their highest level of education compared to 16.6% of male operators. At 23.2%, the proportion of female operators with university credentials was larger than the 15.0% of male operators with university credentials.

Chart 7 Highest level of schooling for farm operators and the total self-employed labour force Canada, 2011

Text description of image

Major field of study

In 2011, over half of farm operators reported a post-secondary education (51.6%), while 65.8% of the total self-employed labour force attained a post-secondary education. Over half of female farm operators, 58.8%, and nearly half of male operators, 48.9%, attained a post-secondary education.

For male operators with a post-secondary education, "agriculture" was the most frequently reported field of study (28.3%), followed by "mechanic and repair technologies/technicians" (15.5%), and "construction trades" (8.6%). For males with a post-secondary education in the total self-employed labour force, "business, management, marketing and related support services" (16.9%) and "construction trades" (9.3%) were the most frequently reported major fields of study.

The most frequently reported fields of study for female operators with a post-secondary education were "health professions and related programs" (23.6%), "business, management, marketing and related support services" (22.7%), and "education" (13.5%), followed by "agriculture" (9.9%). In comparison, the  two most frequently reported fields of study for females in the total self-employed labour force were "business, management, marketing and related support services" (20.7%) and "health professions and related programs" (18.5%).

Main occupation

The most frequently reported main occupations by farm operators were "management occupations", "trades, transport and equipment operators and related occupations" and "natural resources, agriculture and related production occupations". In comparison, in the total self-employed labour force, the most frequently reported main occupations were "management occupations", "trades, transport and equipment operators and related occupations" and  "sales and service occupations" (Table 8).

Occupation refers to the kind of work persons were doing during the week of Sunday, May 1 to Saturday, May 7, 2011, as determined by their kind of work and the description of the main activities in their job.
The occupational data of the 2011 National Household Survey are produced using the National Occupational Classification (NOC) 2011.
Occupations related to agricultural were classified as "management occupations" or "natural resources, agriculture and related production occupations".

 

This table displays the results of table 8 the five most frequently reported main occupations by farm operators and the total self-employed labour force in canada farm operators and total self-employed labour force (appearing as column headers).

Table 8 The five most frequently reported main occupations by farm operators and the total self-employed labour force in Canada, 2011
  Farm operators Total self-employed labour force
All 100 100
Management occupations 50.7 24.3
Trades, transport and equipment operators and related occupations 12.2 17.4
Natural resources, agriculture and related production occupations 10.2 3.8
Business, finance and administration occupations 7.6 11.6
Sales and service occupations 6.3 13.8
Other occupations 12.9 29.1
Source: Statistics Canada, Agriculture–NHS Linkage Database, 2011 and the National Household Survey, 2011.

The occupations most frequently reported by male and female farm operators differed. Female operators most frequently reported "management occupations" (35.5%), "business, finance and administration occupations" (20.6%), and "sales and service occupations" (11.5%). Male operators, most frequently reported "management occupations" (56.5%), "trades, transport and equipment operators and related occupations" (16.0%), and "natural resources, agriculture and related production occupations" (10.5%).

In the total self-employed labour force, "management occupations" were the most frequently reported main occupation for males (27.8%), while for females, "sales and service occupations" (20.8%) were the most frequently reported.

Feature: Immigrant farming population: A closer look at immigrant farmers who made Canada their home

How many immigrants were among the farm population?

There were 44,790 immigrants in the farm population in Canada, making up 6.9% of the total farm population. Immigrants made up a smaller proportion of the farm population than of the total population, with immigrants representing 20.7% of the total population in Canada.

Although the majority of the farm population in each province was non-immigrant, the proportion of immigrants in the farm population ranged from a high of 19.8% in British Columbia to a low of 2.1% in Saskatchewan.

An immigrant refers to a person who is or has ever been a landed immigrant or a permanent resident (who landed in Canada prior to May 10, 2011).

Non-immigrant refers to a person who was not an immigrant. For this article, this includes those that were Canadian born as well as non-permanent residents.

 

Ontario had the highest number of immigrants among the provinces, followed by British Columbia (Table 9). Proportionally, however, there were more immigrants in the farm population in British Columbia (19.8%) than in Ontario (8.6%). In the total population, Ontario had the highest number and largest proportion of immigrants, while British Columbia was second in both number and proportion.

This table displays the results of table 9 proportion of immigrants in the farm population and in the total population in canada immigrants in the farm population and immigrants in the total population, calculated using number, percentage of farm population and percentage of total population units of measure (appearing as column headers).

Table 9 Proportion of immigrants in the farm population and in the total population in Canada, 2011
  Immigrants in the farm population Immigrants in the total population
Number Percentage of farm population Number Percentage of total population
Canada 44,790 6.9 6,768,515 20.7
Atlantic provinces 1,345 5.1 92,990 4.1
Quebec 2,700 2.7 974,895 12.6
Ontario 14,995 8.6 3,611,365 28.5
Manitoba 3,145 6.4 184,500 15.7
Saskatchewan 2,170 2.1 68,775 6.8
Alberta 7,645 5.9 644,115 18.1
British Columbia 12,775 19.8 1,191,875 27.6
Source: Statistics Canada, Agriculture–NHS Linkage Database, 2011 and the National Household Survey, 2011.

Where were Canada's immigrant farm population born?

The five most frequently reported places of birth for Canada's immigrant farm population were the Netherlands, the United Kingdom, the United States, Germany and India (Table 10). In the total population, the most frequently reported places of birth for immigrants born outside of Canada were India, China, the United Kingdom, and the Philippines.

The most frequently reported place of birth for the immigrant farm population differed among the provinces. The Netherlands was the leading place of birth in Ontario and Alberta, while the United Kingdom was the leading place of birth in Manitoba and Saskatchewan. In the Atlantic provinces, the United States was the leading place of birth for the immigrant farm population. In Quebec, Switzerland and France were the most frequently reported places of birth for the immigrant farm population, while India was the most common place of birth for the immigrant farm population in British Columbia.

This table displays the results of table 10 places of birth for immigrants in the farm population and total population in canada. The information is grouped by rank (appearing as row headers), place of birth, farm population, total population and number of immigrants, calculated using number of immigrants, percentage, all, 44,790 , 6,766,320, all others, 10,215 , 22.8, 4,259,710 and 63.0 units of measure (appearing as column headers).

Table 10 Places of birth for immigrants in the farm population and total population in Canada, 2011
Rank Place of birth Farm population Total population
Number of immigrants Footnote 7 Percentage Number of immigrants Footnote 8 Percentage
  All 44,790   6,766,320  
1 Netherlands 8,695 19.4 98,355 1.5
2 United Kingdom 6,645 14.8 536,060 7.9
3 United States 5,750 12.8 262,695 3.9
4 Germany 3,820 8.5 151,810 2.2
5 India 3,785 8.5 547,695 8.1
6 Switzerland 2,400 5.4 18,645 0.3
7 Italy 945 2.1 256,775 3.8
8 China Footnote 9 890 2 545,245 8.1
9 Mexico 880 2 69,645 1
10 Belgium 750 1.7 19,685 0.3
  All others 10,215 22.8 4,259,710 63

Footnotes

Footnote 7

Born outside Canada Note

Return to footnote 7 referrer

Footnote 8

Born outside Canada Note

Return to first footnote 8 referrer

Footnote 9

China excludes Hong Kong Special Administrative Region and Macao Special Administrative Region.

Return to footnote 9 referrer

Source: Statistics Canada, Agriculture–NHS Linkage Database, 2011 and the National Household Survey, 2011.

When did the immigrant farm population immigrate to Canada?

Over one-third (34.4%) of the immigrant farm population immigrated prior to 1971. This compares to 18.6% of the immigrants in the total population (Chart 8).

Recent immigrants, those that immigrated between 2001 and 2011, made up 14.6% of the immigrant farm population. In comparison, recent immigrants made up 31.8% of the immigrants in the total population.

Chart 8 Period of immigration of the immigrant farm population and the total immigrant population to Canada

Text description of image

The place of birth of immigrants differs by immigration period. The most frequently reported places of birth for immigrants in the farm population that immigrated prior to 1971 were the Netherlands (22.0%), the United Kingdom (19.5%) and Germany (12.0%). For immigrants in the total population that immigrated prior to 1971, the most frequently reported places of birth were the United Kingdom (22.2%), Italy (17.0%) and Germany (7.7%).

For recent immigrants in the farm population, the Netherland (20.7%), the United Kingdom (17.0%) and the United States (14.2%) were the most frequently reported places of birth, while in the total population, China (12.4%), India (11.3%) and the Philippines (10.5%) were the most common places of birth for recent immigrants.

What was the age distribution of the immigrant farm population?

The immigrant farm population was older than the non-immigrant farm population. Over half of the immigrant farm population was 55 years or older (51.3%) compared to one-third (30.2%) of the non-immigrant farm population. The proportion of the farm population aged 65 years or older was also higher for immigrants than non-immigrants (28.0% compared to 13.2%).

This is similar to the story in the total population where the immigrant population was also older than those that were not immigrants. The proportion of the total population aged 55 years or older was 36.1% for immigrants and 24.8% for those that were not immigrants.

What were the mother tongues of the immigrant farm population?

The most common mother tongues for Canada's immigrant farm population were English (34.5%), Dutch (17.8%), German (16.7%), Punjabi (8.0%) and French (3.8%). In the total population, the most frequently reported mother tongues for immigrants were English (23.0%), Chinese (4.8%), Tagalog (4.7%) and Spanish (4.5%).

In all of the provinces, the most common mother tongue for the immigrant farm population was English except for Quebec where French was the most common mother tongue. The second most common mother tongue for the immigrant farm population in Quebec, Manitoba and Saskatchewan was German. Dutch was the second most common mother tongue for the immigrant farm population in the Atlantic provinces, Ontario and Alberta, while Punjabi was the second most common mother tongue in British Columbia.

Where does the immigrant farm population live?

Of the 44,790 people in the immigrant farm population, 83.5% lived in rural areas; this can be compared to the non-immigrant farm population of 605,605 people, of which 90.5% lived in rural areas.  In the total population, 5.5% of immigrants and 22.2% of those who were not immigrants lived in rural areas.

In most provinces, the immigrant farm population was less likely to live in rural areas than the non-immigrant farm population. This was most apparent in British Columbia and Ontario. In the British Columbia farm population, 67.9% of immigrants lived in rural areas compared to 79.4% of non-immigrants. In Ontario, 88.7% of the immigrant farm population lived in rural areas, compared to 93.9% of the non-immigrant farm population. Manitoba was the exception as the proportion of the immigrant farm population in the province that lived in rural areas was actually higher than the proportion of the non-immigrant farm population (94.0% compared to 92.8%).

Chart 9 Distribution of the farm population across small, medium and large population centres by immigration category in Canada, 2011

Text description of image

For the farm population living in population centres, immigrants were more likely than non-immigrants to live in large population centres (67.8% compared to 39.0%) (Chart 9).  Similarly, among the total population living in population centres, a higher proportion of immigrants (91.6%) than those who were not immigrants (68.8%) resided in large population centres.

What were the religious affiliations of the immigrant farm population?

In the farm population, immigrants were less likely to report a religious affiliation than non-immigrants (80.9% and 82.7%, respectively). In contrast, immigrants in the total population were more likely to report a religious affiliation than those that were not immigrants (79.9% and 75.1% respectively).

Of the immigrant farm population reporting a religious affiliation, the most frequently reported religions were Roman Catholic (30.4%), Sikh (9.9%), and Christian (7.2%). In the total population, the most frequently reported religions by immigrants were Roman Catholic (35.2%), Muslim (13.3%) and Christian (7.7%). 

For the non-immigrant farm population that reported a religious affiliation, Roman Catholic (39.1%) was still the most frequently reported, followed by the United Church (19.7%) and Lutheran (5.6%).

Roman Catholic was the most frequently reported religious affiliation for the immigrant farm population in each province except for British Columbia where Sikh was the most frequently reported religion.

Focus on immigrant farm operators

In 2011, 9.0% (26,300) of Canada's farm operators were immigrants. This compares to 25.7% of the total self-employed labour force in Canada that were immigrants.

The proportion of farm operators that were immigrants ranged from a high of 24.4% in British Columbia to a low of 2.2% in Saskatchewan. In the total self-employed labour force, the proportion of immigrants ranged from a high of 35.7% in Ontario to a low of 5.8% in Saskatchewan.

What was the median age of immigrant farm operators?

Immigrant operators were on average older than non-immigrant operators. The median age of immigrant operators was 58 years compared to 54 years for non-immigrant operators. Similar to the farm operator population, the median age of immigrants in the total self-employed labour force was higher than non-immigrants (50 years and 49 years, respectively).

What was the highest level of education for immigrant farm operators?

The proportion of immigrant farm operators that reported having a certificate, diploma or degree was higher than that for non-immigrants (81.4% and 78.0%, respectively). The proportion of operators with a postsecondary education was also higher for immigrant operators (61.3%) than for non-immigrants (50.7%). Similarly, the proportion of immigrant farm operators that reported having a college diploma or university credentials was also higher than the proportion for non-immigrants (46.5% and 35.5% respectively).

What were the major fields of study for immigrant operators with a post-secondary education?

Immigrant farm operators with a postsecondary education were more likely to have reported "agriculture" as their field of study than non-immigrant farm operators.  Of the immigrant farm operators with a postsecondary education, 25.8% reported "agriculture" as their field of study compared to 22.2% of non-immigrant operators with a postsecondary education.

What were the most common types of farming for immigrant operators?

"Oilseed and grain farming" was the most common type of farming among immigrant operators (15.8%). This compares to nearly one-third of non-immigrant operators (29.4%). The other most frequent types of farming among immigrant operators were "other animal production" (15.4%), "other crop farming" (13.7%), "fruit and tree nut farming" (12.3%) and "beef cattle ranching and farming" (11.6%). Among non-immigrant operators, after "oilseed and grain farming", the most frequent farm types were "beef cattle ranching and farming" (18.5%), "other crop farming" (17.7%), "other animal production" (11.8%) and "dairy cattle and milk production" (7.5%).

The proportion of operators that were immigrants varies greatly by farm type. In 2011, 1 in 4 operators (26.5%) of "fruit and tree nut" farms was an immigrant. Immigrant operators also made up 17.9% of all operators in "greenhouse, nursery and floriculture production". 1 in 6 operators (16.2%) in "sheep and goat farming" was an immigrant, and among operators in "vegetable and melon farming", 15.9% were immigrants. The proportion of immigrant operators was lowest in "oilseed and grain farming" (5.0%) and in "beef cattle ranching and farming" (5.8%).

 

Feature: Profile of operators under 35 years: A closer look at the younger generation of farm operators

How many operators were under 35 years of age?

There were 24,055 farm operators under the age of 35 years, accounting for 1 in 12 (8.2%) of all operators in the 2011 Ag-NHS database. In comparison, 1 in 6 in the total self-employed labour force was under the age of 35 years (15.6%).

Among the provinces, Quebec had the largest proportion of younger operators, while British Columbia had the smallest proportion of younger operators (Chart 10).

Among the total self-employed labour force, Alberta and Quebec had the highest proportions of those aged 15 years to 34 years, while the lowest proportion was in the Atlantic provinces.

Chart 10 Proportions of farm operators and of the self-employed labour force that were under 35 years of age in Canada, 2011

Text description of image

The majority of operators under 35 years were male (73.5%), consistent with the proportion of males in the total operator population (72.5%). In the total self-employed labour force, males also made up the majority of those under 35 years at 59.4%.

What was the highest level of education for younger operators?

Operators under the age of 35 years were more likely to have attained at least a secondary school education than operators aged 35 and older. For operators under the age of 35 years, 85.9% reported having a secondary school education or higher compared with 77.6% of operators aged 35 years or older.

Quebec and British Columbia had the largest proportions of operators under the age of 35 that attained at least a secondary school education (91.4% of operators in Quebec and 90.9% of operators in British Columbia).

Operators under 35 years were more likely to have completed a post-secondary education (59.5%) than operators aged 35 years or older (50.9%). Quebec had the largest proportion of operators under the age of 35 years that reported an education attainment beyond secondary school (78.8%). In Quebec, operators under the age of 35 years most frequently reported a college diploma (37.7%) or a trades certificate (29.9%) as their highest level of education attainment.

What were the major fields of study for operators under 35 years with a post-secondary education?

Of all operators that completed a post-secondary education, proportionally more operators under 35 years reported their major field of study as "agriculture" than operators 35 years or over (41.8% and 20.5% respectively). This was also true for immigrant operators that had a post-secondary education. In total, 47.2% of immigrant operators under 35 years reported "agriculture" as their major field of study, compared to 25.0% of immigrant operators aged 35 years and older.

What were the main occupations of operators under 35 years?

Nearly half of farm operators under 35 reported their main occupation as "management occupations", while in the total self-employed labour force, "trades, transport and equipment operators and related occupations" was the most frequently reported by those under 35 years (Table 11).

This table displays the results of table 11 main occupation reported by farm operators under 35 years and those under 35 years in the self-employed labour force in canada percent of operators under 35 years and percent of the self-employed labour force under 35 years (appearing as column headers).

Table 11 Main occupation reported by farm operators under 35 years and those under 35 years in the self-employed labour force in Canada, 2011
  Percent of operators under 35 years Percent of the self-employed labour force under 35 years
All 100 100
Management occupations 44.7 17
Trades, transport and equipment operators and related occupations 14.2 19.4
Natural resources, agriculture and related production occupations 13.3 5.7
Sales and service occupations 7 14.3
Business, finance and administration occupations 6.9 7.9
Other occupations 13.8 35.7
Source: Statistics Canada, Agriculture–NHS Linkage Database, 2011 and the National Household Survey, 2011.

How many operators under 35 years were on multi-generation farms?

In 2011, 34.5% of operators under 35 years were on multi-generational farms, compared to 12.6% of operators 35 years or older.

For operators under 35 years, Quebec and British Columbia had the largest proportions on multi-generational farms at 48.4% and 45.7% respectively. Saskatchewan had the smallest proportion of operators under 35 years on multi-generational farms at 25.9%.

 

The 2011 Census of Agriculture and the National Household Survey Linkage data are available without charge in CANSIM: tables 004-0100 to 004-0129.

For more information about farms in Canada, consult the CEAG 2011 release at:  /eng/ca2011/index 

For custom data requests from the Ag-NHS linkage please contact STATCAN.infostats-infostats.STATCAN@canada.ca

Notes

Footnotes

Footnote 1

Refers to a farm, ranch or other agricultural operation that produces at least one of the following products intended for sale: crops, livestock, poultry, animal products, greenhouse or nursery products, Christmas trees, mushrooms, sod, honey or bees, and maple syrup products.

Return to footnote 1 referrer

Footnote 2

Dwelling used for commercial, institutional or communal purposes, such as a hotel, a hospital or a work camp.

Return to first footnote 2 referrer

Footnote 3

This refers to the population of Canada's provinces, which was published in the 2011 National Household Survey.

Return to first footnote 3 referrer

Footnote 4

The total population includes the farm population.

Return to first footnote 4 referrer

Date modified:

Statistics Canada - Producer Prices Division

2012/2013

This document is confidential when completed.

Si vous préférez recevoir ce questionnaire en français veuillez composer le 1-877-604-7828.

Please provide your email address.

This information is collected under the authority of the Statistics Act, Revised Statutes of Canada, 1985, Chapter S-19.

Completion of this questionnaire is a legal requirement under this act.

Purpose of this survey

This survey collects financial, wage and contractor fee information that is used to produce price indexes. These indexes measure change in prices for informatics professional services. You as the respondent will benefit from completing this questionnaire by now having the ability to benchmark your company against other companies in the same industry (in aggregate form only).

Statistics Canada uses this information to better measure the volume of activity in the computer services industry. For the purpose of this survey, "informatics professional services" covers the following types of businesses: software publishers; data processing; hosting and related services; computer systems and related services; Internet publishing and broadcasting, and web search. Your information may also be used by Statistics Canada for other statistical and research purposes.

Confidentiality

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. Statistics Canada will use the information from this survey for statistical purposes.

Record linkages

In order to enhance the information you provide in this survey, Statistics Canada plans to combine the responses relating to your organization with the information you previously provided on this survey Statistics Canada may also combine the information you provide with other survey or administrative data sources.

Your participation is important

Your participation is vital to ensure that the information collected in this survey is accurate and comprehensive.

Fax or email transmission disclosure

Statistics Canada advises you that there could be a risk of disclosure during facsimile or email. However, upon receipt, Statistics Canada will provide the guaranteed level of protection afforded to all information collected under the authority of the Statistics Act.

Return Procedures

Please return the completed questionnaire to Statistics Canada within 15 days of receipt by mail using the return envelope.  You can also fax it to us at 1-888-883-7999 or email to bsso@statcan.gc.ca.

Lost the return envelope or need help? Call us at 1-877-604-7828 or mail to: Statistics Canada, 150 Tunney's Pasture Driveway, Ottawa, Ontario, K1A 0T6

Visit our website at www.statcan.gc.ca

If necessary, please make address label corrections (please print)

  • Legal Name
  • Business Name
  • Title of Contact
  • First Name of Contact
  • Last Name of Contact
  • Address (number and street)
  • City
  • Province/ territory
  • Postal Code/Zip Code
  • Country
  • Language Preference
    • English
    • French

A. Introduction

Instructions:

Please use this page as a quick reference for definitions of the Business Activities listed on the next page in Section B.

Software Publishing

This Canadian industry includes establishments primarily engaged in publishing computer software, usually for multiple clients and generally is referred to as packaged software. Establishments in this industry carry out operations necessary for producing and distributing computer software, such as designing, providing documentation, assisting in installation and providing support services to software purchasers. These establishments may design and publish, or publish only.

Examples: Packaged computer software publishing (including designing and developing), Packaged computer software (all formats), all formats, Packaged publishers games.

Data Processing, Hosting and Related Services

This Canadian industry includes establishments primarily engaged in providing hosting or data processing services. Hosting establishments may provide specialized hosting activities, such as web hosting, video and audio streaming services, application hosting, application services provisioning, or may provide general time-share mainframe facilities to clients. Data processing establishments may provide complete processing and preparation of reports from data supplied by the customer; specialized services, such as automated data entry; or they may make data processing resources available to clients on an hourly or time-sharing basis.

Examples: Application hosting, Automatic data processing, Computer input preparation services, Computer processing services, Computer time-sharing services, Data entry services, Data processing services, Disk and diskette conversion services, Input preparation services, Leasing of computer time, Microfilm recording and imaging services, Optical scanning data services, Rental of computer time, Computer service bureaus, Video and audio streaming services, Web hosting.

Internet Publishing, Broadcasting and Web Search Portals

This Canadian industry includes establishments primarily engaged in publishing and/or broadcasting content on the Internet or operating web sites, known as web search portals, that use a search engine to generate and maintain extensive databases of Internet addresses and content in an easily searchable format. The Internet publishing and broadcasting establishments in this industry provide textual, audio, and/or video content of general or specific interest. These establishments do not provide traditional (non-Internet) versions of the content that they publish or broadcast. Establishments known as web search portals often provide additional Internet services, such as e-mail, connections to other web sites, auctions, news, and other limited content, and serve as a home base for Internet users.

Examples: Internet directory publishing; Internet book publishing; Internet broadcasting; Internet entertainment sites; Internet game sites; Internet newspaper publishing; Internet periodical publishing; Internet software publishing; Publishing, maps, street guides and atlases (exclusively on Internet); Technical books, publishing (exclusively on Internet); Web search portals.

Computer Systems Design and Related Services

This Canadian industry includes establishments primarily engaged in providing expertise in the field of information technologies through one or more activities, such as writing, modifying, testing and supporting software to meet the needs of a particular customer, including the creation of Internet home pages; planning and designing computer systems that integrate hardware, software and communication technologies; on-site management and operation of clients' computer and data processing facilities; providing advice in the field of information technologies; and other professional and technical computer-related services.

Examples: Computer consulting services, Disaster recovery services, Facilities management services, Hardware consulting services, Custom computer programs or systems software development; Custom computer software consulting services, programming services, systems analysis and design; Computer-aided design (CAD) systems services; Computer-aided engineering (CAE) systems services; Data processing facilities management services; Design and system analysis, computer services (software); Facilities management services, computer support services; Information management system design services; Internet page design services, custom; Local area network (LAN) systems integrators; Management information systems design consulting services; Office automation, computer systems integration;  Computer hardware requirements analysis; Software installation services; Custom software programming; Custom software systems analysis and design; Systems analysis and design, computer services (software); Systems engineering (system integration); Web page developing.

B. Business Activities

We have selected one business activity for your company:

Internet Publishing, Broadcasting and Web Search Portals

This Canadian industry includes establishments primarily engaged in publishing and/or broadcasting content on the Internet or operating web sites, known as web search portals, that use a search engine to generate and maintain extensive databases of Internet addresses and content in an easily searchable format. The Internet publishing and broadcasting establishments in this industry provide textual, audio, and/or video content of general or specific interest. These establishments do not provide traditional (non-Internet) versions of the content that they publish or broadcast. Establishments known as web search portals often provide additional Internet services, such as e-mail, connections to other web sites, auctions, news, and other limited content, and serve as a home base for Internet users.

Was your company engaged in the business activity identified above in 2013?

  • Yes Please go to Section C.
  • No Please select one of the business activities below that best represents your business and complete the questionnaire.

Note: If you did not perform the pre-selected activity at all in 2013, then select the activity that represents your main business activity from the choices below.

Descriptions and examples of the business activities are given in section A.

  • Software Publishing
  • Data Processing, Hosting and Related Services
  • Internet Publishing, Broadcasting and Web Search Portals
  • Computer Systems Design and Related Services
  • Other - Please Specify:

C. Reporting Period

Please report information for your fiscal years in 2012 and 2013

2012:

  • Fiscal year end date (year/month/day):
  • Number of months:

2013:

  • Fiscal year end date (year/month/day):
  • Number of months:

D. Revenue Share

Please provide the dollar value for revenue received from each of the following business activities in 2012 and 2013 from all provinces. Please report in Canadian dollars for your company's Canadian operations. Please see Section A "Introduction", for the details of each activity.

If your company operates in more than one location then please provide the total from all locations (provinces and territories) in Canada.

Reporting Instructions:

  • Include: Fees charged to clients for employees and contract workers and expenses (cost + mark-up) recovered from clients (e.g. hardware, software, travel and accommodation, subcontracted services)
  • Exclude: Revenue earned by foreign operations. Software sales unrelated to informatics professional services and all taxes collected for remittance to a government agency.

Business Activity

  • Software Publishing:
    • Revenue in 2012 (CAN$):
    • Revenue in 2013 (CAN$):
  • Data Processing, Hosting and Related Services:
    • Revenue in 2012 (CAN$):
    • Revenue in 2013 (CAN$):
  • Internet Publishing, Broadcasting and Web Search Portals:
    • Revenue in 2012 (CAN$):
    • Revenue in 2013 (CAN$):
  • Computer Systems Design and Related Services:
    • Revenue in 2012 (CAN$):
    • Revenue in 2013 (CAN$):
  • Other :
    • Revenue in 2012 (CAN$):
    • Revenue in 2013 (CAN$):
  • Total:
    • Revenue in 2012 (CAN$):
    • Revenue in 2013 (CAN$):

E. Operating Revenue & Expenses

Reporting Instructions:

Please provide the dollar value for the revenue and expenses for the fiscal years indicated, only for the business activity selected in Section B.

If your company has locations in other provinces and territories across Canada, then please provide the total from all locations in Canadian dollars only.

Please do not report revenue and expenses unrelated to the business activity selected in Section B.             

Revenue:

Operating Revenue:

  • Include: Fees charged to clients for employees and contract workers and expenses (cost + mark-up) recovered from clients (e.g. hardware, software, travel and accommodation, and sub-contracted services).
  • Exclude: Revenue earned by foreign operations. Software sales unrelated to informatics professional services and all taxes collected for remittance to a government agency.
    • Operating Revenue in 2012 (CAN$):
    • Operating Revenue in 2013 (CAN$):

Expenses:

Expenses for Employees:

  • Include: Wages, salaries, benefits and bonuses paid to full-time, part-time and temporary employees whose time was charged to the business activity selected in Section B
  • Exclude: Overhead expenses (e.g. wages, salaries and benefits  and bonuses of administrative staff, building occupancy costs, purchased services such as legal and accounting services).
    • Expenses for Employees in 2012 (CAN$):
    • Expenses for Employees in 2013 (CAN$):

Expenses for Contract Workers:

  • Include: Fees paid to contract workers for their work on the business activity selected in Section B.
    • Expenses for Contract Worker in 2012 (CAN$):
    • Expenses for Contract Worker in 2013 (CAN$):

Other Expenses:

  • Include: All other expenses incurred for work on the business activity selected in Section B (e.g. software, hardware upgrades, office expenses, travel and accommodation). 
  • Exclude: Overhead such as taxes refunded by government, rent, utilities and insurance.
    • Other Expenses in 2012 (CAN$):
    • Other Expenses in 2013 (CAN$)

F. Personnel

Average number of paid employees during the reporting period for the business activity selected in Section B.

  • To calculate the average number employed, add the number of employees in the last pay period of each month of the reporting period and divide this sum by the number of months (usually 12).
  • Exclude: Partners, proprietors and non-salaried personnel.
    • Average number of paid employees in 2012:
    • Average number of paid employees in 2013:

Full-time employees during the reporting period for the business activity selected in Section B.

  • Full-time employment consists of persons who usually work 30 hours or more per week.
  • To calculate the average number of full-time employees: add the number of full-time employees in the last pay period of each month of the reporting period and divide this sum by the number of months (usually 12).
    • Average number of paid employees who worked full-time in 2012:
    • Average number of paid employees who worked full-time in 2013:

G. Average Annual Percentage Change in Labour Rates

Average annual percentage change for salaries and wages paid to employees and fees paid to contract workers.

For the fiscal year  indicated and the business activity selected in Section B, please complete the average annual percentage change for Salaries and wages paid to employees and fees paid to contract workers. Please follow the example below:

Example:  Your company has 3 employees who can charge their time to the activity selected in Section B.  Two of these employees received annual increases of 1% and 5%.  The third employee did not receive an increase (0%). The sum of the three wage rate changes (1%+5%+0%) is 6.0%. When you divide by the number of employees (+6% / 3 employees), the result is an average annual percentage changes in wage rates of 2.0%. 

If there is no variation in the average annual percentage change of salaries and wages or of fees paid to contract workers, then write "0".

Salaries and wages rates:

  • Please report the average annual percentage change (+,-) in the salaries and wages paid to employees whose time is charged to the business activity  selected in Section B for all provinces and territories.
  • Exclude: The salary or wage changes for general and administrative staff.
    • Salaries and Wages Rates in 2012 (%):
    • Salaries and Wages Rates in 2013 (%):

Fees paid to contract workers:

  • Please report the average annual percentage change (+,-) in the fees paid to contract workers whose time is charged to the business activity  selected in Section B for all provinces and territories.
    • Fees Paid to Contract Workers in 2012 (%):
    • Fees Paid to Contract Workers in 2013 (%):

H. Contact Information

Name of authorized person to contact about this questionnaire (please print)

  • First Name of Authorized Person:
  • Last Name of Authorized Person:
  • Title of Authorized Person:
  • Telephone Number:
  • Extension:
  • Fax Number:
  • E-mail Address:
  • Website Address:

I certify that the information contained herein is complete and correct to the best of my knowledge.

Date:

Signature:

I. Administration

Time to complete questionnaire

How long did you spend collecting and reporting the information needed to complete this questionnaire?

Pre-filled questionnaire

In order to facilitate the completion of next year's questionnaire, we can provide you with a copy of the information you provided this year.  Do you authorize us to send a pre-filled questionnaire containing the information you provided this year?

Please check

  • Yes (Please send a pre-filled questionnaire next year)
  • No (Please send a blank questionnaire)

J. Comments






Please make a copy of this completed questionnaire for your records.

Thank you for completing this questionnaire.