Data quality, concepts and methodology: Explanatory notes

View the most recent version.

Data quality and methodology

This publication results from the compilation of data reported monthly in the "Refined Petroleum Products" Survey, covering all refining companies in Canada and major wholesalers/distributors. We would like to thank these respondents for their co-operation in the survey.

In order to ensure the accuracy and consistency of these data, the results of this survey are reconciled with other Energy section surveys such as the monthly Crude Oil and Natural Gas Production (survey ID # 2198), the monthly Oil Pipeline Statement (survey ID # 2191) and the monthly Oil Pipeline Transport (survey ID # 2148) surveys in terms of inter-provincial movements, imports, exports and deliveries to refineries by type of crude and means of delivery.

For survey data, non-sampling errors can be attributed to one or more of the following sources:

Data response error. This error may be due to questionnaire design and the characteristics of the questions, inability or unwillingness of the respondent to provide correct information or misinterpretation of the questions because of definitional difficulties.

Non-response error. Some respondents are unable to respond, while others may be too late in responding. Data for the non-responding units can be imputed using the data from responding units. The extent of error due to imputation is usually unknown and is very much dependent on any characteristics differences between the respondent group and the non-respondent group in the survey.

Processing error. These are the errors that may occur at various stages of processing the data such as coding, data entry, verification, editing, weighting, tabulation, etc.

These reported data allow for a measurement of refining activity in Canada. The supply of crude oil is indicated by source and the volume of oil and other materials charged is also shown. Production of 17 refined products or groups of products is detailed and is measured at a "net yield" level, that is, after exchanges between various processes in refinery operations. As a result, inter-products transfers are primarily those transfers needed for marketing purposes. Propane and butanes, shown at the gross level, are the exceptions to the net yield measurement and therefore interproduct transfers for these products are not restricted to marketing transfers, but include movement to gasoline blending, to petrochemicals, etc. Similarly, interproduct transfers for "other products", a classification covering unfinished products whose final product form can vary, is not restricted to marketing transfers.

Data on the exports and imports of refined products are those reported by respondents to the survey, adjusted for exports and imports made by non-responding companies. When such adjustments are made, corresponding changes are made to "domestic sales". As a consequence, figures shown for "domestic sales" should accurately reflect total Canadian demand for refined products.

A quarterly questionnaire, requesting a detailed breakout by consumer categories of the domestic sales figures shown in this publication, is completed by the same companies that report to the monthly "Refined Petroleum Products" survey. The results of this quarterly survey, together with data for all other energy forms, are published in Catalogue no. 57-003-X "Report on Energy Supply – Demand in Canada".

The questionnaire used in the monthly refined petroleum products survey is subject to regular revision to reflect changing technologies, etc. Similarly, the list of respondents is constantly updated to ensure a total coverage of refining operations and a sufficiently wide coverage of the wholesalers/distributors of refined products.

The Canadian Socio-economic Information Management System (CANSIM), which is Statistics Canada's computerized data bank, contains most of the data presented in this publication from 1973, less detailed information is available for the years 1956 to 1972. Please refer to tables 134-0001, 134-0002 and 134-0004.

Data Confidentiality

Statistics Canada is prohibited by law from publishing any statistics which would divulge information relating to any identifiable business without the previous consent in writing of that business. The disclosure of data likely to reveal specific information on the activity of a reveal specific information on the activity of a particular company is therefore systematically and rigorously controlled and regulated to prevent the publication or disclosure of any information deemed confidential. The rules regarding the confidentiality of parts of the results of merchandising surveys are therefore applied in such a way that no data are shown for cells (or statistical units) composed of fewer than three companies or of three or more if these are dominated by one or two companies. Then, in the publication itself, the information is withheld on this designated cell by entering an "X" in the appropriate cell.

Another way of preserving data confidentiality without having to enter an "X" is to combine (consolidate) the information from selected regions or selected cells) for the same item. The data in question can then simply be aggregated and published in a less detailed, but still confidential form.

Next