Monthly Survey of Food Services and Drinking Places (MSFSDP)
The Monthly Survey of Food Services and Drinking Places provides estimates of the value of sales of restaurants, caterers, and drinking places.
Detailed information for December 2013
Data release - February 27, 2014
- Questionnaire(s) and reporting guide(s)
- Variables and definitions
- Data sources and methodology
- Data accuracy
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-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.
- Accommodation and food
- Business, consumer and property services
Data sources and methodology
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).
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.
This is a sample survey with a cross-sectional design.
The frame, derived from Statistics Canada's Business Register, includes about 82,500 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.
The sample is based on a stratified simple random design. The frame is stratified according to Province / Territory and by 4 digit NAICS: 7221 (Full service restaurants), 7222 (Limited service eating places), 7223 (Caterers and food service contractors), and 7224 (Drinking places). These strata are further stratified based on a revenue measure of enterprise size derived from the Business Register. The take-all stratum contains the complex enterprises and the large enterprises. There are two take-some strata that contain the medium size enterprises and in these strata a simple random sample is taken. The take-none strata contain the small enterprises where no sample is taken; instead administrative data (GST) is used for estimation. Overall, the sample size is about 2,000 enterprises covering more than 10,000 establishments.
An initial sample was drawn in 1998, when the survey was converted to NAICS. New samples were re-drawn in 2004, 2007 and 2008. The sample is refreshed each month by including a sample of births from the population. Every few years, all establishments in the sample are updated to take into account changes in their revenue, dead units are removed from the sample, and some small and medium-sized units are rotated out, while others are rotated into the sample. Starting with the reference month of March 2009, estimates are based on a restratified sample based on the list of enterprises and establishments on the Business Register as of December 2008. For additional information, consult the article "Monthly Survey of Food Services and Drinking Places: Impact of restratification" in the Documentation section below.
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 Head Office staff.
Data are collected through a mail-out/mail-back process, while providing respondents with the option of telephone or other electronic filing methods.
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 new take-none strata and for the take-some strata for the full-service restaurants and the limited-service eating places for the following provinces: Quebec, Ontario, Manitoba, Saskatchewan, Alberta and British Columbia.
View the Questionnaire(s) and reporting guide(s) .
Data are examined for inconsistencies and errors using automated edits coupled with analytical review. Where possible, data will be verified using alternate sources.
Imputation is used to estimate for non-response and missing data. Imputation methods include the use of historical monthly trends from data collected in previous years and from current GST data. Data from imputation sources that fail statistical edits are considered as outliers and are not used in the imputation process.
Sales are estimated by multiplying each data response by its sampling weight. For the take-all strata the weight is 1 since all enterprises are selected in the sample. For the take-some strata, the sample is selected by simple random sampling and the weight is therefore the inverse of the probability of selection in the stratum.
In the population represented by the take-some strata of NAICS 7221 and 7222, in the provinces of British Columbia, Alberta, Saskatchewan, Manitoba, Ontario, and Quebec (where the overall sample size is about 800 enterprises) Ratio Estimation is used. This method of estimation is a modification and extension to the use of the sampling weights. It improves the quality of the estimate by taking advantage of the high correlation between survey data and GST data. The use of administrative data also reduces response burden and survey costs. The approach of ratio estimation (for this part of the population) began with the January 2009 estimates.
There is no sample for the take-none strata. Instead, sales are estimated using the ratio estimation approach for all provinces and NAICS based on the data from the take-some strata.
The standard error and CV of the estimates are derived directly from the stratified simple random sample.
Prior to dissemination, 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, historic trends, and comparisons with other data sources.
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.
Revisions and seasonal adjustment
Monthly, preliminary estimates are provided for the reference month. Raw estimates, based on late responses, are revised for the two previous months. Seasonally adjusted estimates are computed using the X-12-ARIMA software and revised for the three previous months. At the end of each calendar year, seasonally adjusted estimates are revised to equal the sum of the raw estimates. Seasonal adjustment options are also reviewed on an annual basis and updated as required.
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 December 2008 and run in parallel with the old sample until May 2009. This new sample also reflected improvements made to the Business Register since the last sample refreshment (restratification) in March 2007. 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.
The first historical revisions were applied and released in the June 2009 preliminary estimates. Historical revisions in the raw data were required to correct for significant non-sampling errors. These revisions included replacing imputed data with reported data, corrections to previously reported data or to industry classifications, and estimates for new births that were not known at the time of the original release. Also, the revisions reflected the improvements made to imputation methods, calendarization, and modelling of administrative (Goods and Services Tax) data.
With the release of the August 2010 preliminary estimates, historical revisions were applied back to January 2008. Similar to the revisions done in the previous year, corrections were required to adjust for non-sampling errors. Furthermore, additional revisions were made starting in January 2009 to address changes to the estimation methodology. The seasonally adjusted data series were revised back to January 2005.
A new round of revisions were made and published in the release of the April 2011 preliminary estimates. Historical data were revised back to January 2010 and the seasonally adjusted data series were revised back to January 2008.
With the release of the September 2012 preliminary estimates, historical revisions have been made going back to January 2011. These revisions include replacing imputed data with reported data, introducing new births, and utilizing the most recent administrative data. The seasonally adjusted data series have been revised back to January 2008.
With the release of the March 2013 preliminary estimates, historical revisions have been made going back to January 2012. These revisions include replacing imputed data with reported data, introducing new births, and utilizing the most recent administrative data. The seasonally adjusted data series have been revised back to January 2009.
The response rate, which is about 75%, is a measure of the proportion of those sample units that have responded in time for inclusion in the estimate.
The coefficient of variation is used to measure the sampling error of the estimates. The coefficient of variation, at the national level for total sales, is 0.8%. However, this coefficient of variation does not include the component of sampling error derived from the model (as described in the section on estimation). This applies to the take-none strata and the take-some strata for NAICS 7221 and 7222, in British Columbia, Alberta, Saskatchewan, Manitoba, Ontario, and Quebec.
Codes for estimated coefficients of variation codes are provided in the tables below,
A - Excellent CV is 0.0% to 5.0%
B - Very good CV is 5.1% to 10.0%
C - Good CV is 10.1% to 15.0%
D - Acceptable CV is 15.1% to 25.0%
E - Use with caution CV is 25.1% to 35.0%
F - Unreliable CV is larger than 35.1%
- ARCHIVED - Monthly Survey of Food Services and Drinking Places: Impact of restratification (PDF Version, 90.42kb)
- Technical Details on Estimation in the Monthly Survey of Food Services and Drinking Places (MFS)
- Date modified: