Research and Development in Canadian Industry (RDCI)

Detailed information for 2012





Record number:


This survey collects data to monitor science and technology related activities in Canada and to support the development of science and technology policy.

Data release - 2012 Intentions (2010 questionnaire), October 3, 2012. Preliminary figures (2011 questionnaire), October 17, 2013. Actual data (2012 questionnaire), August 19, 2014


This survey highlights expenditures and personnel devoted annually to scientific research and development (R&D) by Canadian industry and non-profit industrial research institutes and associations. It collects data essential to assure the availability of pertinent statistical information to monitor science and technology related activities in Canada and to support the development of science and technology policy.

The results, published in the publication "Industrial research and development: intentions" (catalogue 88-202) are also used as a key component in the Gross Expenditures of Research & Development (GERD) in Canada series.

Industrial R&D data are combined with data from other R&D performing sectors: Federal Science Expenditures and Personnel, Activities in the Social Sciences and Natural Sciences (record no. 4212), higher education and Research and Development of Canadian Private Non-profit Organizations (record no. 4204).

These data serve many users from: federal & provincial government science analysts who develop and monitor programs aimed at stimulating science and technology in Canadian industry to international organizations such as the OECD (Organisation for Economic Co-operation and Development) and UNESCO (United Nations Educational, Scientific and Cultural Organization). University researchers, research councils, business enterprises, research institutes and associations, science journal writers, the general public and the media are all users of R&D data.

Statistical activity

Science and technology (S&T) and the information society are changing the way we live, learn and work. The concepts are closely intertwined: science generates new understanding of the way the world works, technology applies it to develop innovative products and services, and the information society is one of the results of the innovations.

People are looking to Statistics Canada to measure and explain the social and economic impacts of these changes.

The purpose of this program is to develop useful indicators of S&T activity in Canada and to present them coherently.

Reference period: Fiscal year


  • Research and development
  • Science and technology

Data sources and methodology

Target population

The universe for the RDCI consists of all firms known or believed to be involved in the performance or funding of R&D. The frame for this survey has a long history spanning back to the conception of the survey more than 50 years ago. Firms are identified through many sources but most frequently firms are added from the CRA (Canada Revenue Agency) T661 files.

The survey population comprised:
- all firms that had reported R&D expenditures in the three previous reference year's surveys;
- firms with an approved claim for a federal R&D income tax incentive for the three previous reference years;
- firms that were identified by respondents in surveys of government science and technology activities as R&D contractors or grantees for the reference year and forward;
- firms that were reported by other firms as funding or performing R&D in the prior collection cycle; and
- firms identified as funding or performing R&D in the reference year and forward through newspaper, journal articles or publicly available directories.

Instrument design

The questionnaire was designed and developed using the OECD guidelines as outlined in the Frascati Manual (2002). The questionnaire's evaluation and testing is an ongoing process, insofar as to how it covers inputs to R & D activity, including sources of funds and type of expenditures, R&D personnel and payments made or received for patents, licenses and technological know-how. For reference year 2009 and onward data on R&D spending by field of science or technology are available.


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

A sample of approximately 2,000 responding units, was selected from the frame consisting of the following groups:

1. A "must take" stratum consisting of special entities such as industrial non-profit organizations, technology purchasers or vendors, and known R&D performers that do not file scientific research and experimental development (SR&ED) tax incentive applications. These special entities were all selected to be included in the sample because there are no other sources of data available for them. Industrial non-profit organizations are not eligible for tax incentives while some commercial firms opt not to make a claim.

2. The "take all" stratum comprises the largest R&D performers in each of the industrial groups. These large R&D performers represented about two-thirds of R&D expenditures in each of the specified industry groups for the previous reference year. All units in this stratum were included in the sample.

3. The "take some" stratum is composed of mid-size R&D performers in each of the specified industry group. A sample of units from this stratum was included in the sample.

4. A coverage study stratum containing 50 enterprises are included in the sample based on criteria designed to test the existing survey coverage. Enterprises which meet the testing criteria, but are not on the current survey frame, are added to the sample. Units that are in scope contribute to the estimate and will be included in the survey frame for future collection cycles. Testing criteria include industrial classification NAICS 541710 (Research and development services in physical, engineering and life sciences), reporting R&D expenses in tax filings (GIFI L9282) and indicating R&D activities or expenditures in other Statistics Canada surveys.

5. A "take none" stratum comprised of the smallest R&D performers, those firms whose total R&D expenditures comprised the bottom 5% of all R&D expenditures in each industrial group, was created to reduce response burden. Firms in the "take none" stratum were excluded from the sample

Data sources

Responding to this survey is mandatory.

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

This annual postal survey is aimed at all Canadian industries known to be performing or funding research and development. Respondents are given the option of completing the questionnaire electronically. The survey collects data for four years. For example the 2009 survey conducted in 2010, collects data on actual R&D spending in 2008 and 2009, on the preliminary figures for 2010, and on the spending intentions for 2011. Two follow-up postcards are mailed to all outstanding respondents. Telephone contact is made to all non-reporting establishments to discuss reporting options and to make special arrangements, including partial response in some instances.

The data for most of the small performers and funders is taken directly from the CRA T661 file. Firms have up to eighteen months, after their fiscal year end, to submit a claim to CRA. The processing of these records is stopped in March (15 months after the end of the calendar year), therefore figures will need to be revised in the following survey cycle. These revisions affect the two years prior to the survey year.

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

Error detection

In the pre-grooming stage of processing, edit checks are performed to identify missing and invalid entries that would point to data records that are in error. Extreme errors resulting from processing were identified in the tax and questionnaire data.

Data editing

Editing is a process to ensure that survey data are acceptable, complete, consistent and correct. There are three main categories of edits: validity, consistency and distribution edits. Validity and consistency edits are done one record or questionnaire at a time. However, distribution edits are performed by looking at data across questionnaires.

Validity edits

Validity edits identify incoherence in the data. Examples of validity edits include:
- Respondents reporting intramural R&D performance with no R&D personnel;
- Wages and salaries for R&D which are greater than the firms total wages and salaries;
- Units of measure issues (U.S. vs. Canadian dollars, dollars vs. thousands of dollars)

Consistency edits

Consistency edits verify the relationships between questions. Consistency edits may also be applied to the logical flow of the questionnaire, or may involve the use of administrative data or historical data. These types of edits typically verify relationship between questions.

For the RDCI, some examples of consistency edits are:
- Wages and salaries and other current costs on R&D performed should equal total current costs;
- Total current costs at Canada level should equal the total current expenditures reported for provinces and territories;
- The total R&D expenditures reported for Canada should equal the total sources of funds for R&D performed;
- Total R&D expenditures should match the total for all fields of science or technology;
- Total R&D personnel should likewise be the same across all questions.

Distribution edits

A question on the RDCI allows for the distribution of values for expenditures and personnel across provinces, while another new question allocates expenditures and personnel across science types. Expenditures are also allocated across sources of funding. These distribution questions are edited to identify outliers which are then validated.


It is not usually possible to resolve all records in error during the pre-grooming stage. Imputation replaces items that fail the edit rules to fix partial non-response or total non-response.

Imputation for RDCI uses the following data sequence:
- Actual respondent estimates from the prior year for planned expenditures;
- SR&ED tax data;
- Random ratio donors anchored to historical data.

Deterministic imputation

Deterministic imputation is done as part of the editing process. It is generally specified as action items to be performed using logic decision tables. In deterministic imputation only one value is deemed possible. Deterministic imputation is generally of the form A+B=C. An example would be 'total professionals + total technical and administrative staff = total R&D personnel.

Imputation by substitution

Imputation by substitution involves the use of an external data source. An auxiliary data source such as historical data or administrative data is used for missing data. For the RDCI, COA4 (explained below) and PD7 (explained below) files were used to impute revenues and employment data. The T661 (Scientific Research and Experimental Development) tax credit applications were used as an alternate data source that was treated as respondent data.

For SR&ED tax filers, revenue figures were adjusted to reflect corporate income tax data for the corresponding filer. The tax data are from T2 corporate income tax files which are mapped to the Statistics Canada's Chart of Accounts (COA) classification by firm. The variable COA4 relating to (Total) Revenues of a firm was used to improve data quality for missing or inconsistent total revenues.

The Payroll Deductions total employment data (PD7) file was also used to improve the quality of missing or inconsistent total employment data. Payroll Deduction data are monthly data, and therefore, an annual average was calculated from the Canada Revenue Agency (CRA) monthly Payroll Deduction file for all firms that reported having one or more employees in at least one of the twelve months of the tax year.


Imputation method based on estimators generally refers to the use of ratios based on historical data or other variables on the questionnaire. To estimate R&D expenditures two years past the base year, editing rules were applied using donor ratios and a response was imputed based on the response of a similar firm in the same industry group. Data are modeled using mathematical formulae.

Donor records for imputation were determined by imputation class, which were defined by population subgroups, NAICS group and size. Size was determined by total R&D expenditure (total intramural and extramural expenditures) which was used to group enterprises. For the suite of RDCI surveys, the following imputation methods were employed: deterministic imputation, substitution, and use of estimators.

Quality evaluation

The quality of any statistical information is measured in a large part by the degree to which the final product meets the original objective of the survey. The RDCI survey intention is to monitor science and technology related activities in Canada and to support the development of science and technology policy.

In order to assure the highest quality of the data from this survey we continually monitor the coverage of our survey population, to ensure that all known enterprises in the universe are accounted for. If data are not obtained directly from the respondent many other sources are researched, including funders' reports where available, previous questionnaires or administrative data, and published data such as financial statements.

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.

To prevent disclosing data on individual respondents, many items or cells must be grouped together to provide sufficient observations for dissemination (e.g., socioeconomic objectives).

Revisions and seasonal adjustment

Administrative data are used for this survey. Updates of the administrative data and filing timelines necessitate historical revisions to the data. Normally the two years prior to the current survey year are revised.

Data accuracy

One of the problems in a survey of this type is to ensure that the quality of the data is satisfactory. It cannot be expected that all firms funding R&D will be surveyed, will respond and will report correctly.

There are sources of information such as federal government grant contract lists to aid in identifying firms and editing returns.

The coverage, however, is probably not complete. This is especially true for the smaller companies in the service industries. In addition, R&D is a term subject to inconsistencies. Thus, the data, although reasonably accurate, cannot be regarded as precise.


Date modified: