Appendix 1
Survey descriptionFootnote 13
Target population
The target population was defined to ensure relevant coverage of business in Canada to meet informational needs on strategies and tactical decisions, innovation, and GVCs. To lower the response burden for small enterprises in Canada, the target population was limited to enterprises with 20 or more employees that collected revenues of at least $250,000 in 2009. In addition, several industries were not included due to either low levels of business R&D or a high presence of public sector activities.Footnote 14 The resulting target population consisted of 67,686 enterprises,Footnote 15 which represents approximately 10 percent of all enterprises in Canada with at least one employee within the in-scope sectors.
Although the size threshold limits the focus of the survey to a relatively small portion of the total number of enterprises in Canada, it comprises a substantial proportion of Canadian employment. As shown in Box A1, enterprises with 20 or more employees cover approximately 77 percent of employment in Canada and 87 percent of employment within the manufacturing sector.
| Industry | Percentage of enterprisesFootnote k | Percentage of employeesFootnote l |
|---|---|---|
| All industries | 10% | 77% |
| Manufacturing | 25% | 87% |
Sources: Statistics Canada, Business Registry, December 2009. Statistics Canada, Survey of Employment, Payrolls and Hours, April 2010. | ||
Sample
From a population of 67,686 enterprises, 6,233 were sampled using a random sample methodology stratified by industry and enterprise size.
Enterprise size classes
The SIBS survey is stratified by three enterprise size classes, which are based on the number of employees. The enterprise size classes include small enterprises, which have 20–99 employees; medium enterprises, which have 100–249 employees; and large enterprises, which have at least 250 employees.
The selected size classes are somewhat different from the small and medium-sized taxonomy traditionally used for enterprises in Canada.Footnote 16 This reflects, at least in part, a desire to be comparable to other national innovation surveys. However, two of the most relevant national innovation surveys, the European Community Innovation Survey (CIS) and the U.S. Business R&D and Innovation Survey (BRDIS), use different employee thresholds in their sample selectionFootnote 17 (See Box A2). While direct comparison with these two surveys cannot be performed without some adjustments, international comparison of Canadian results with European or U.S. surveys is still possible.Footnote 18
| SIBS | CIS | BRDIS |
|---|---|---|
| 10–49 | 5–24 | |
| 20–99 | 25–49 | |
| 50–249 | 50–99 | |
| 100–249 | 100–249 | |
| 250+ | 250+ | 250–499 |
| 500–999 | ||
| 1,000–4,999 | ||
| 5,000–24,999 | ||
| 25,000+ |
Industry detail
Overall, the sample spans 133 industries and industry groupings from the two- to six-digit level of aggregation, according to the North American Industry Classification System (NAICS 2007). These industry categories were selected based on the level of industry R&D expenditure as identified in Statistics Canada's Industrial Research and Development: Intentions 2008, as well as consultations with Industry Canada and Foreign Affairs and International Trade Canada.
The sampling strategy was designed to achieve the greatest potential for empirical research at the enterprise level to support policy analysis. The resulting sample varies considerably across industries. For example, the sampling for the manufacturing sector (34 percent of the manufacturing enterprises' target population) was high relative to most sectors, and was necessary to achieve the desired industry detail given the number of R&D intensive industries within manufacturing, see Box A3.
| NAICS | Industry description | Total sample size after stratification | Target population |
|---|---|---|---|
| 11 | Agriculture, forestry, fishing and hunting | 15 | 2,276 |
| 21 | Mining, quarrying, and oil and gas extraction | 142 | 988 |
| 22 | Utilities | 102 | 172 |
| 23 | Construction | 22 | 9,559 |
| 31–33 | Manufacturing | 4,394 | 12,846 |
| 41 | Wholesale trade | 54 | 7,219 |
| 44–45 | Retail trade | 26 | 13,280 |
| 48–49 | Transportation and warehousing | 426 | 3,664 |
| 51 | Information and cultural industries | 340 | 1,344 |
| 52 | Finance and insurance | 152 | 2,541 |
| 53 | Real estate and rental and leasing | 15 | 2,033 |
| 54 | Professional, scientific and technical services | 514 | 5,397 |
| 55 | Management of companies and enterprises | 15 | 1,019 |
| 56 | Administrative and support, waste management and remediation services | 16 | 5,348 |
| All industries | 6,233 | 67,686 |
Response rate and data quality
The survey was considered complete if six mandatory questions were answered. These six questions covered various issues such as: enterprise strategic decisions, whether the enterprise introduced a product or process innovation, and whether the enterprise had business activities outside of Canada. Of the 6,233 enterprises sampled, 4,249 returned completed questionnaires resulting in a 71 percent response rate.
Overall, the response rates by question among completed questionnaires vary between 68 percent and 100 percent and the response rates for the last questions are not systematically lower than for the first questions.
The estimates for a given industry aggregate are weighted by the appropriate enterprise population count, making the results representative of the target population.