Survey on Small R&D Performers
Detailed information for 2007
The objective of this survey is to better understand the issues and difficulties encountered by enterprises performing research and development (R&D) work, at low levels of expenditures.
Data release - April 3, 2009
This survey collects information to better understand the behaviour of small R&D performers and recipients of refundable tax credit. Small R&D performers conduct R&D activities more often on an occasional basis rather than continuous. An interest for analysts is to explore factors influencing occasional R&D performers should they wish to perform R&D on a more permanent basis. This survey also addresses the role and impact of support programs in firms' decisions to conduct R&D activities and how theses activities are conducted.
This survey collected information on the benefits of R&D work and on the success rate of the enterprise to reach some strategic objectives such as the reduced costs or the enhanced quality of products etc.
The information compiled by this survey will also be used by the Canadian government to analyze policies supporting economic development of small R&D performers.
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 based on a framework that ties them together in a coherent picture.
- Research and development
- Science and technology
Data sources and methodology
This survey targets Canadian enterprises performing between $10,000 and $250,000 on R&D performance activities. This survey covers the complete Canadian industrial sector; (NAICS 11 to 91: North American Industry Classification System, Statistics Canada, 2007). It is possible to find in this population large enterprises, in terms of revenue or employment, which declare small amounts of R&D expenditures.
The questionnaire collects information on research and development activities of the last five years on performers that spend between 10 000 and 250 000 dollars on R&D. The questionnaire also collects information on the reasons and motivations of R&D activities. The questionnaire asks respondents if they received tax credits and the causes that motivate their request for tax credits. Finally, the questionnaire includes a section on benefits of R&D work.
The questions have been developed to fulfill the needs of Statistics Canada's partners. The questionnaires target specific information needed to complete the current knowledge on behaviors and characteristics of small R&D performers in Canada.
The questions have been tested with twenty respondents located in Montréal and Ottawa. The questionnaire was elaborated after consultation with partners (ISQ consortium, NRC-IRAP, Finance Canada and NSERC).
This is a sample survey with a cross-sectional design.
The survey on Small Research and Development Performers sample was drawn from the administrative portion of the frame for the Survey on Research and Development in Canadian Industry (RDCI, record number 4201). The business register (August 2008) was used to update the frame.
Sample Unit: Enterprise
Only establishments satisfying the following criteria were sampled:
- Research and development expenditures between $10,000 and $250,000.
- Have performed research and development in any of the following years; 2004, 2005 and 2006.
The sample size is 2150 units. The survey was drawn from an initial population of more than 16 thousands units. The average probability of selection was approximately one over eight. In addition, the target population is composed mainly of small R&D performers. As a consequence, there is virtually no industry dominance. For a few strata, probability of being selected was higher.
Stratification has been done for 21 industrial sectors (NAICS 1111-9191) (manufacturing and services sectors) and by provinces. For each industrial sector, two strata were created (stratum 1 for small enterprises that spend between $10,000$ and $100,000 and stratum 2 for enterprises of medium size that spend between $100,000 and $250,000).
Data collection for this reference period: 2008-10-01 to 2008-12-31
Responding to this survey is mandatory.
Data are collected directly from survey respondents.
Data were collected using a paper mail-out, mail-back questionnaire.
View the Questionnaire(s) and reporting guide(s) .
A computer program was crafted to perform a systematic detection of missing values and to assess the coherence between respondents' answers to various questions. Some questions were manually scrutinized to resolve some inconsistencies that could not be solved through the computer program. In some cases, respondents were called back to clarify their answers.
Based on the error/editing scan performed, an automatic imputation program was applied. Missing values were imputed by using a donor with characteristics similar to the recipient.
Following edit and imputation, all valid records were weighted to reflect to target survey population. Estimations were computed using the Generalized Estimation System software, which takes into account the stratification of the sample and computes standard errors for all estimates.
Quality assurance measures were implemented at each step of the data collection and processing cycle to monitor the quality of the data.
Comments made by respondents were evaluated by subject matter experts for potential incorporation into existing choices of responses. No systematic patterns were found in the comments.
All tables were scrutinized for inconsistencies and coherence.
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 avoid potential residual disclosure, the original industrial distribution of 21 industry groups was reduced to five for publication purposes. Each cell was examined to ensure that the number of records behind estimates is sufficient to avoid any potential disclosure of confidential information.
Revisions and seasonal adjustment
This methodology does not apply to this survey.
The original survey population was extracted form the Research and Development in Canadian Industry database. In this database, the main source of information on small R&D performers comes from the recipients of R&D tax credits under the Scientific Research and Experimental Development Tax Credit programme. As a consequence, any firm conducting R&D but never applying for a tax credit was not part of the survey population and therefore not covered. To alleviate to some extent this issue, three years were used to avoid selecting only the recent recipients of R&D tax credit. There are potentially a number of firms that may have performed R&D that were excluded from this survey because these firms never applied for a tax credit. The survey could not estimate the extent of this potential under coverage.
The overall response rate was 65%, which is the number of valid questionnaires received over the total number of respondent and non-respondents minus out-of-scope and out of business.
Imputation rates vary from question to question. Defined as the number of units imputed over the number of respondents to a specific question, imputation rates range from 0,3% (for the lowest) to 22.5% for question #6. The next highest imputation rates were 20,9% for question #17 and 19,6% for question #14 (in particular #14b).
The Generalized Estimation System software provides weighted estimates and reliability estimates called standard errors. The standard error takes into account both the variability of the estimate and the sample design. Estimates are classified on a scale from highly reliable to too unreliable to be published. In this latter case, data are suppressed from tables. The majority of data estimates published were highly reliable (letter "A", for standard errors between 0 and 2,5%) and reliable (letter "B", for standard errors between 2,5% and 7,5%). The majority of data estimates published were highly reliable (letter "A", for standard errors between 0 and 2,5%) and reliable (letter "B", for standard errors between 2,5% and 7,5%).