Monthly Survey of Food Services and Drinking Places (MSFSDP)

Detailed information for January 2017

Status:

Active

Frequency:

Monthly

Record number:

2419

This survey provides information to measure the economic performance and health of the Food Services and Drinking Places Industry in the Canadian economy.

Data release - March 31, 2017

Description

The Monthly Survey of Food Services and Drinking Places provides estimates of the value of sales of restaurants, caterers, and drinking places by province and territory and by industry at the North American Industry Classification System (NAICS) four or six digit level. These data are used by federal and provincial governments, private associations and food service businesses for consulting, marketing and planning purposes. The provincial and federal governments use the information to estimate provincial taxation shares.

Reference period: Month

Collection period: Approximately 14 working days after the end of the reference month and continues for approximately 4 weeks.

Subjects

  • Accommodation and food
  • Business, consumer and property services

Data sources and methodology

Target population

The target population includes all statistical establishments that are classified as either food services or drinking places (NAICS 722) in the North American Industry Classification System (NAICS 2012).

Instrument design

This questionnaire collects data on monthly sales of food service establishments. The items on the questionnaire have remained unchanged for several years. Some minor modifications were made with the survey redesign of 2007 to facilitate its use and clarify a few elements. Associations representing the industry were consulted.

Sampling

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

FRAME
The Business Register is a repository of information reflecting the Canadian business population and exists primarily for the purpose of supplying frames for all economic surveys in Statistics Canada. It is designed to provide a means of coordinating the coverage of business surveys and of achieving consistent classification of statistical reporting units. It also serves as a data source for the compilation of business demographic information.

The major sources of information for the Business Register are updates from the Statistics Canada survey program and from Canada Revenue Agency's (CRA) Business Number account files. This CRA administrative data source allows for the creation of a universe of all business entities.

Included in the Business Register are all Canadian businesses* which meet at least one of the three following criteria:
(1) Have an employee workforce for which they submit payroll remittances to CRA; or

(2) Have a minimum of $30,000 in annual sales revenue; or

(3) Are incorporated under a federal or provincial act and have filed a federal corporate income tax form within the past three years.

The data provided in our products reflects counts of statistical locations by industrial activity (North American Industry Classification System), geography codes, and employment size ranges.

SAMPLING UNIT
The sampling unit is the establishment, as defined on the Business Register. Specifically, the sampling unit is the cluster of establishments in the target population belonging to the same enterprise. In most cases, a cluster will consist of a single establishment.

The frame, derived from Statistics Canada's Business Register, includes about 95,000 enterprises with one or more establishment classified to NAICS 722. Each enterprise must be classified on the Business Register as alive and active. Most are simple enterprises (with a single establishment) but about 1,300 are complex enterprises (with multiple establishments) and they have about 7,600 establishments. The sampling unit is the enterprise which is a cluster of establishments classified to NAICS 722 belonging to the enterprise.

STRATIFICATION METHOD
The sample is based on a stratified simple random design. The population is divided according to Province/Territory and by 4- or 6-digit NAICS: 722511 (Full service restaurants), 722512 (Limited service eating places), 7223 (Caterers and food service contractors), and 7224 (Drinking places). These strata are then further divided based on enterprise size (a revenue measure derived from the Business Register). There is a take-all stratum (census), up to two take-some strata (partially sampled), and a take-none stratum (not sampled). The take-none strata contain the small enterprises, and administrative data (goods and services tax - GST) is used for estimation instead of sampling.

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

The sample was selected using simple random sampling within each stratum. The sample is refreshed each month by including a stratified simple random sample of births from the population, i.e. new clusters of establishment(s). Births are identified using the Business Register.

Every few years, establishments in the sample and population are updated to take into account changes in their revenue, deaths (clusters that are no longer active or are no longer in NAICS 722), etc.

Overall, the sample size is about 2,400 questionnaires, covering more than 17,000 active establishments.

New samples were drawn in 2004, 2007, 2008, and 2015. Starting with reference month January 2016, estimates are based on a restratified sample drawn using the list of establishments on the Business Register as of November 2015.

Data sources

Responding to this survey is mandatory.

Data are collected directly from survey respondents and extracted from administrative files.

Data collection, data capture, preliminary edit and follow-up are performed by regional office staff.

Data are collected through various methods, such as telephone, mail-out/mail-back and electronic questionnaire.

Follow-up procedures are applied when a questionnaire has not been received after a pre-specified period of time.

Administrative data (from the Goods and Services Tax - GST) are the main source for the estimates for the take-none strata and are used as an auxiliary variable to improve the estimates in the take-some strata.

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

Error detection

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 Monthly Survey of Food Services and Drinking Places, data editing is done at two different time periods.

The first editing process is conducted during data collection. Once data are collected via the telephone, or electronic questionnaire or the receipt of completed mail-in questionnaires, the data are captured using customized data capture applications. All data are subjected to data editing.

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.

Statistical editing is also done after data collection and 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. Reported data for the current reference month will go through various edit checks.

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.

Imputation

Imputation is used to estimate for non-response and missing data. Imputation methods include the use of monthly trends from historical data and from administrative data (GST sales revenue). Historical and administrative data that fail statistical edits are considered outliers and are not used in the imputation process. 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.

Estimation

Estimation is a process by which Statistics Canada obtains values for the population of interest so that it can draw conclusions about that population based on information gathered from only a sample of the population. Sales are estimated using ratio estimation. Ratio estimation consists of replacing the initial sampling weights (defined as the inverse of the probability of selection in the sample) by new weights in a manner that satisfies the constraints of calibration. Calibration ensures that the total of an auxiliary variable estimated using the sample must equal the sum of the auxiliary variable over the entire population, and that the new sampling weights are as close as possible (using a specific distance measure) to the initial sampling weights.
For example, suppose that the known population total of the auxiliary variable is equal to 100 and based on a sample the estimated total is equal to 90, so that we are underestimating by approximately 10%. Since we know the population total of the auxiliary variable, it would be reasonable to increase the weights of the sampled units so that the estimate would be exactly equal to it. Now since the variable of interest is related to the auxiliary variable, it is not unreasonable to believe that the estimate of the sales based on the same sample and weights as the estimate of the auxiliary variable may also be an underestimation by approximately 10%. If this is in fact the case, then the adjusted weights could be used to produce an alternative estimator of the total sales. This alternate estimator is called the ratio estimator. In essence, the ratio estimator tries to compensate for 'unlucky' samples and brings the estimate closer to the true total. The gain in variance will depend on the strength of the relationship between the variable of interest and the auxiliary data.

The take-none portion is taken into account by including its auxiliary variable data in the known population total. This causes the weights for the sampled portion to be increased in such a way that the estimates will include the take-none portion.

Like other Statistics Canada monthly surveys, the calculated weighted sales 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 Business Register 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.

The standard error and CV of the estimates are derived directly from the stratified simple random sample.

Quality evaluation

Prior to dissemination, combined survey results are analyzed for comparability. In general, this includes a detailed review of individual responses, general economic conditions, historic trends, and comparisons with other data sources.

Disclosure control

Statistics Canada is prohibited by law from releasing any information it collects which could identify any person, company, 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.

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 (unadjusted) data are revised, on a monthly basis, for the two months 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 and October.

In addition, revisions are made once a year, with the initial release of the March 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 restratification are done.

The Monthly Survey of Food Services and Drinking Places 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. 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.

It is standard practice that every few years the sample is refreshed to ensure that the survey frame is up to date with births, deaths and other changes in the survey population. To that effect, a new sample was drawn in November 2015 and run in parallel with the old sample until May 2016. This new sample also reflected improvements made to the Business Register since the last sample refreshment (restratification) in December 2008. In addition the survey methodology has been refined to improve imputation of non-respondents, calendarization of reported data by respondents that do not report for a complete month, use of administrative (Goods and Services Tax) data, and modifications to seasonal adjustment options.

Data accuracy

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 non-sampling 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 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. While the impact of non-sampling errors is difficult to evaluate, certain measures such as response and imputation rates can be used as indicators of the potential level of non-sampling error.

Coefficients of variation (CV) and response rates are major data quality measures used to validate results from the MRTS.

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.

RESPONSE RATES
The average weighted response rate for 2016 is 89%.

NON-SAMPLING ERROR
Non-sampling 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 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. While the impact of non-sampling errors is difficult to evaluate, certain measures such as response and imputation rates can be used as indicators of the potential level of non-sampling error.

NON-RESPONSE BIAS
Since response rates in the Monthly Survey of Food Services and Drinking Places are generally high, there is a low risk of non-response bias affecting the estimates.

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.

COVERAGE ERROR
The Business Register is used to determine the target population for the survey. When coverage errors are identified (for example, if a unit in the target population is included more than once, or is included by mistake due to incorrect classification in NAICS 722), corrections are made on the Business Register and the updates will be reflected in the population after the following month's population update.

OTHER NON-SAMPLING ERRORS
Other sources of non-sampling error such as data processing errors are minimized through the data editing process.

Processing errors may occur at various stages of processing such as coding, data entry, verification, editing, weighting, and tabulation, etc. Non-sampling errors are difficult to measure. More important, non-sampling errors require control at the level at which their presence does not impair the use and interpretation of the results.

Measures have been undertaken to minimize the non-sampling errors. For example, units have been defined in a most precise manner and the most up-to-date listings have been used. Questionnaires have been carefully designed to minimize different interpretations.

Date modified: