Quarterly Retail Commodity Survey (QRCS)

Detailed information for third quarter 2003

This survey collects detailed information about retail commodity sales in Canada to produce estimates of the distribution of the sales of various commodities at the national level, for different types of retail outlets in Canada.

Data release - January 7, 2004

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

Description

The Quarterly Retail Commodity Survey (QRCS) collects detailed information about retail commodity sales in Canada. The objective is to produce estimates of the distribution of the sales of various commodities at the national level, for different types of retail outlets in Canada. The survey is a supplement to the Monthly Retail Trade Survey (MRTS- Survey ID 2406). MRTS gathers total monthly retail sales, while QRCS collects a breakdown of these sales by commodity.

The retailers in the Monthly Survey of Large Retailers (Survey ID 5027) are also included in the sample of the Quarterly Retail Commodity Survey. The same questionnaire is used for both surveys. The data provided to the Monthly Survey of Large Retailers is incorporated into QRCS. Excluding recreational and motor vehicle dealers, these large retailers account for about 38% of retail trade. The retailers belonging to the Monthly Survey of Large Retailers are included based on their sales size and contribution to the food, clothing, home furnishings, electronics, sporting goods and general merchandise sectors of retail trade.

The information provided by QRCS can be used to track commodity sales within and across various types of retail stores, as well as to calculate commodity market share, and to gain a better understanding of the rapidly changing retail industry. The data show the type of outlets where consumers prefer to buy certain commodities, and shifts in what the different types of commodities retailers decide to sell. Analysis of this data assists in establishing trends in commodity sales over time.

The QRCS data is used by the Statistics Canada's System of National Accounts with respect to the estimates of personal expenditure. Other users of the data include federal and provincial government departments, retail analysts, market researchers, industry experts and independent consultants.

Reference period: Quarter

Subjects

• Accommodation and food
• Business, consumer and property services
• Retail and wholesale
• Retail sales by type of product

Data sources and methodology

Target population

QRCS is a sub-sample of the MRTS sample. The MRTS universe is based on the 1980 Standard Industrial Classification. Establishments included in retail trade are those primarily engaged in retail activities, such as, buying commodities for resale to the general public for personal or household consumption and providing installation and repair services. In general, retail establishments are classified by kind of business according to the principal lines of commodities sold and services provided.

Establishments such as farm machinery dealers and dealers in farm supplies which sell to the farmer as a farm operator are considered to be in wholesale trade but those which sell to farmers as household consumers are included in retail trade. On the other hand, lumber and building materials dealers, fuel dealers and sellers of office furniture, machines and equipment, such as computers, are classified in wholesale trade, whether they sell mainly to other businesses or to household consumers.

The target population consists of statistical companies on the Business Register Central Frame Database that have at least one location identified in the retail trade sector according to the 1980 Standard Industrial Classification (SIC).

• Additional documentation

Instrument design

The questionnaires were developed at Statistics Canada and were reviewed and tested in the field and focus groups were held in both official languages. In the course of developing the survey, Statistics Canada consulted with a number of large retailers as well as with industry associations including the Canadian Federation of Independent Business and the Retail Council of Canada.

Sampling

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

The QRCS sample consists of a subset of the retailers in the Monthly Retail Trade Survey (MRTS). In order to take full advantage of the information and infrastructure provided by an already existing retail survey, a two-phase sample design methodology was used where the first phase sample is the MRTS sample.

First phase: The MRTS sample is drawn from Statistics Canada's Business Register Central Frame Data Base (CFDB). The target population consists of statistical companies on the CFDB that have at least one location identified in the retail trade sector according to the 1980 Standard Industrial Classification (SIC). The population of companies is stratified according to industry (retail trade groups consisting of groupings of 4-digit SIC codes), geographic region, and size (large, medium or small based on estimated gross business income). The largest companies within each combination of trade group and geographic region are sampled in MRTS with certainty (census), as are companies that operate in more than one trade group or geographic region. The other companies are sampled on a probability basis. The MRTS sample consists of about 14,000 companies.

Second phase: The frame from which the QRCS sample is drawn is the set of companies in the MRTS sample. As in the first phase, the sampling unit is the statistical company. In order to make the second phase design as efficient as possible, the MRTS sample is completely restratified making use of the most recent information available from MRTS. The companies in the MRTS sample are restratified by trade group, province/territory and company size (large, medium or small based on MRTS sales information). Similar to the first phase, the largest companies within each combination of trade group and province/territory are sampled in QRCS with certainty. A sample of the small and of the medium-sized MRTS companies in each combination of trade group and province/territory are selected to be in QRCS. The QRCS sample consists of about 8,300 companies. Similar to MRTS, the sample is updated each month to reflect changes in the population.

Data sources

Responding to this survey is mandatory.

Data are collected directly from survey respondents.

If a respondent finds it more convenient to report their commodity data to Statistics Canada on a monthly basis, they are allowed to do so. Respondents can report annually when the distribution of their sales does not vary throughout the year. The reference period refers to the period that the commodities were actually sold in the retail stores. The collection period is the period that the interviewers collected the sales data.

Data are principally collected by the Statistics Canada Regional Offices. The system used to process and capture the collected data is called BLAISE. The retailers on the panel for the Monthly Survey of Large Retailers as well as a selected number of units are collected through the head office in Ottawa.

Respondents are given a choice of collection methods: mail or telephone. They also have the choice to report commodity data in dollars or as a percentage of total sales and receipts. Telephone follow-up is conducted to resolve edit problems with mail-back questionnaires and to collect data from respondents who have not returned the questionnaire.

The initial contact with the respondent consists of sending the respondent a package including an introductory letter informing the respondent that a Statistics Canada representative will be calling. A sample questionnaire and a commodity index are also included. This package is followed by a telephone conversation to introduce the survey to the respondent, identify the person best able to provide the data and obtain a detailed profile of what the business sells over a one-year time frame. A profile is a list of all the commodities sold by the retailer. The questionnaire is then tailored to the commodities sold by the retailer. Commodities no longer sold are removed from the respondent's profile and will not appear on the next tailored questionnaire.

Commodity indexes were developed to assist interviewers and respondents in choosing the most appropriate commodity codes to classify the type of items being sold by retailers. There are two indexes -- one is organised by commodity code and the other one is an alphabetical listing by commodity.

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

Error detection

During data collection, on-line edits are performed to check for consistency between the current period's data and the last period's data. If the commodities reported for the current period are inconsistent with the previous period, the data are verified with the respondent. Edits to ensure that the captured information is numerically valid and that all data fields are completed are also performed, as well as edits to ensure that the reporting period dates are valid.

Once the data are received back at head office an extensive series of processing steps is undertaken to thoroughly verify each record received. Edits are performed at the micro level to ensure that: the commodities sold make sense for the type of store; the sum of the individual commodities equals the total sales reported and that there are no missing fields; the total sales reported to this survey is in line with the sales reported to Monthly Retail Trade Survey; and there are no large fluctuations in commodity sales from period to period. Records failing these edits are subject to manual inspection and possible corrective action.

Imputation

An automated imputation system is used to impute for missing or erroneous data. Non-respondents, as well as respondents with one or more fields flagged for imputation (due to incomplete or inconsistent data identified during the editing process), are subject to imputation. Since the RCS sample is monthly-based, the imputation system processes the data for one reference month at a time. The system makes use of the auxiliary information available from MRTS. Since all retailers in RCS are also in MRTS, the total sales for each record is obtained from the MRTS file after the MRTS edit and imputation process has been completed. The commodity fields are then imputed one at a time using the following methods.

For non-respondents, the system uses the most recent historical data available to determine which commodities are sold by the retailer. These commodities are flagged for imputation and the remaining commodities are set to zero. RCS uses adjusted historical imputation to impute for total non-response. Data from the retailer for the same month of the previous year is used. If that data is unavailable, the previous month data is used.

For respondents with fields requiring imputation and for non-respondents where the historical data are not available, commodity values are imputed by ratio imputation using a current auxiliary variable. Imputation groups of similar retailers are formed on the basis of type of store, geographic region and size. Values imputed to a unit will be derived from the values of respondents belonging to the same group. Respondents that are considered to be outliers (either due to extremely large fluctuations in their commodity distributions when compared to their previous data, or due to unusual commodity sales for the type of store) are excluded from the group. For each commodity requiring imputation, the ratio of the group's commodity sales to the group's total sales is applied to the unit's total sales. When there are not sufficient respondents in an imputation group, groups at successively more aggregated levels of type of store, geographic region or size are used.

Since the commodity fields are imputed one at a time, the imputation process is followed by a prorating step to ensure that all parts add up to the corresponding totals.

Estimation

The goal of QRCS is to produce quarterly estimates for the distribution of total retail sales among various commodities. The source for the level of the total retail sales is MRTS. QRCS total sales are benchmarked at the trade group level to the MRTS sales estimates. An exception to this is the Department Stores trade group (QRCS includes the sales of department store concessions while MRTS does not). The MRTS sales estimates include both the surveyed and non-surveyed portions. By benchmarking QRCS to MRTS, the sales for the non-surveyed portion of QRCS are taken into account.

Benchmarking is also done for the New Car Dealers trade group. The commodity distribution from the responses to the New Motor Vehicle Dealer Commodity Survey is applied to the MRTS retail sales for the trade group.

Since all missing commodity information is imputed (both for non-respondents and for respondents with some missing data), there is no adjustment at the estimation stage for non-response. The estimation weight that is applied to units in the QRCS sample is made up of three components that are multiplied together. The first component is a weight reflecting the first-phase of the sample (i.e. a weight to inflate the MRTS sample data to represent the entire population). The second component is a weight reflecting the second-phase (i.e. a weight to inflate the QRCS sample data to represent the entire MRTS sample). The third component is an adjustment factor to ensure that the QRCS total sales estimate equals the MRTS sales estimate at the trade group level except for the Department Stores trade.

Since the MRTS and QRCS samples are monthly-based, commodity estimates and their variances are calculated for each month of the quarter. The monthly estimates are then summed to obtain commodity estimates for the quarter. A double-expansion ratio estimator is used to produce the commodity estimates. Variances are calculated using a formula developed with the Taylor linearization method and adapted for a two-phase stratified design. Because the samples are not independent from month to month, a covariance component is included to obtain the variances for the quarter.

Quality evaluation

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

The data are examined at a macro level to ensure that the long-term trends make sense when compared to publicly available information in media reports, company press releases, etc. Large fluctuation in year-over-year sales for commodities are analysed to determine if they are in error or if sales for these commodities accurately reflect retail activity. Subject matter officers follow up with the company to confirm the data and to document reasons for large fluctuations in sales.

Disclosure control

Statistics Canada is prohibited by law from releasing any information it collects which could identify any person, business, or organization, unless consent has been given by the respondent or as permitted by the Statistics Act. Various confidentiality rules are applied to all data that are released or published to prevent the publication or disclosure of any information deemed confidential. If necessary, data are suppressed to prevent direct or residual disclosure of identifiable data.

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

Revisions and seasonal adjustment

Each release, current quarter preliminary estimates as well as previous quarter revised estimates are made available. Once a year, annual revisions are performed. The revisions mainly stem from responses received after the initial release of the quarter's data. Data are also revised due to revisions to the retail sales level provided by MRTS.

QRCS total sales estimates are benchmarked at the trade group level to the sales estimates (before seasonal adjustment) from the Monthly Retail Trade Survey (MRTS). Total sales for QRCS differ slightly from the sales published by MRTS in that the sales of department store concessions are included in QRCS and not in MRTS.

QRCS estimates are not adjusted for seasonality.

Data accuracy

The commodity estimates are derived from a sample survey and, as such, are subject to both sampling and non-sampling errors. Sampling errors are present because observations are made only on a sample and not on the entire population. The sampling error depends on factors such as the size of the sample, variability in the population, sampling design and method of estimation. The coefficient of variation (CV), which is the estimated standard error expressed as a percentage of the estimate, is used to measure the degree to which sampling error potentially exists within the sample. Estimates with smaller CVs are more reliable than estimates with larger CVs.

Non-sampling errors are not related to sampling and may occur for many reasons. Population coverage errors, differences in the interpretation of questions, incorrect information from respondents, and mistakes in recording, coding and processing data are examples of non-sampling errors. Non-response is an important source of non-sampling error. While the impact of non-sampling errors is difficult to evaluate, measures such as response rates and imputation rates can be used as indicators of the potential level of non-sampling error.

Each commodity estimate is assigned a code from A to F (where A is most reliable and F is to be used with caution) as an indicator of data quality. This quality indicator code is a joint measure of the magnitude of the CV and the imputation rate. The imputation rate is the proportion of the estimated sales which comes from imputed data. For example, if the total estimated sales for a commodity is $1 million, and $150,000 is from imputed data, then the imputation rate is 15%.

The final quality indicator code is determined by first assigning a code based only on the CV. The code is then adjusted to take into account the imputation rate for that estimate. Estimates with a CV in the range of 0% to 5% are assigned an A; 5% to 10% a B; 10% to 16.5% a C; 16.5% to 25% a D; 25% to 33% an E. If the imputation rate is below 10%, the CV code becomes the final quality indicator code. If the imputation rate is between 10% and 33%, the CV code is downgraded by one (i.e. an A would become a B, a C would become a D). If the imputation rate is between 33% and 60%, the CV code is downgraded by two (i.e. an A would become a C, a C would become an E).

At the total retail level, the major commodity estimates currently have the following quality indicator codes:

Food and beverages: B
Health and personal care products: B
Clothing, footwear and accessories: A
Home furnishings and electronics: B
Motor vehicles, parts and services: A
Automotive fuels, oils and additives: B
All other goods and services: B