Survey of Innovation

Detailed information for 1997-1999

The information collected by this survey provides information on innovation and innovation activities of Canadian businesses and their characteristics.

Data release - January 31, 2000

• Questionnaire(s) and reporting guide(s)
• Description
• Data sources and methodology
• Data accuracy

Description

The survey is part of an on-going program to measure innovation in Canada. To meet this objective the survey collects information on new and significantly improved products and processes introduced during a three year time period. The information collected by this survey provides information innovation and innovation activities of Canadian businesses and their characteristics. Some topics can include innovation activities, sources of information, problems and obstacles, impact of innovation, cooperative and collaborative arrangements for innovation, business success factors, intellectual property protection, and use of government support programs. The survey is conducted every 3-4 years, depending on need, and covers a 3-year reference period. Industries surveyed may vary from survey to survey. Coverage is largely determined by client sponsorship.

Estimates produced from the survey are used by:
. firms for market analysis;
. trade associations to study performance and other characteristics of their industries;
. government to develop national and regional economic policies.

Surveys of innovation at Statistics Canada have included the 1993 Survey of Innovation and Advanced Technology which surveyed manufacturing firms; the Survey of Innovation, 1996 which surveyed the communications, financial services and technical business services industries; the 1999 Survey of Innovation, Advanced Technologies and practices in the Construction and Related Industries Survey; and the Survey of Innovation 1999 which surveyed manufacturing and selected natural resource industries for the reference period 1997-1999. The Survey of Innovation 2003 surveys information and communication technology industries; selected professional, scientific and technical services industries; selected natural resource support service industries; and selected transportation industries for the reference period 2001-2003.

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 based on a framework that ties them together in a coherent picture.

Collection period: end of fall of reference period

Subjects

• Innovation
• Science and technology

Data sources and methodology

Target population

The target population was all firms in the manufacturing sector (NAICS 31-33) (North American Industry Classification System, Statistics Canada, 1998) or in selected natural resource industries (NAICS 1133, 212, 2211). The requirement to produce provincial statistics was a major criterion in defining the sampling unit. Following the recommendations of Eurostat, the enterprise was chosen. Whenever possible, provincial data are to be an output of surveys at Statistics Canada.

Accordingly, within each province, for each enterprise, all establishments coded to the same 4-digit NAICS industry groups were grouped to form one sample unit or "provincial enterprise". To reduce response burden on small businesses, only provincial enterprises with at least 20 employees and provincial enterprises with a gross business income of at least $250,000 were considered in sample selection.

• Additional documentation

Instrument design

The questionnaire was designed by the Science, Innovation and Electronic Information Division of Statistics Canada in collaboration with Industry Canada, the Institute for Research in Construction of the National Research Council of Canada and Natural Resources Canada. The questionnaire was tested by an interview of a small sample of individual firms to ensure that the questions were well understood. Feedback from these firms was incorporated into the questionnaire.

Sampling

This is a sample survey.

Frame Description: Statistics Canada has many annual surveys that collect production data. To reduce response burden for the Survey of Innovation, it was decided to limit questions to those exploring the nature of innovation. In order to provide production data to accompany innovation data, the sample was chosen as a sub-sample of existing responses to surveys conducted by Statistics Canada's Manufacturing, Construction and Energy Division (MCED) and Natural Resources Canada. These included Statistics Canada's Manufacturing, Construction and Energy Division's (MCED) Annual Survey of Manufactures, Coal Mines Annual Survey, Annual Survey of Electric Power Generation, and Natural Resources Canada's Annual Census of Mines, Quarries and Sand Pits. By using a sub-sample of respondents to these existing surveys, new businesses (births since 1997) were not part of the target population. At the time of sample selection, the most current data for a common year was chosen. Only businesses that responded to the production surveys were considered for the population for the Survey of Innovation.

Sample units: The requirement to produce provincial statistics was a major criterion in defining the sampling unit. Following the recommendations of Eurostat, the enterprise was chosen. Whenever possible, provincial data are to be an output of surveys at Statistics Canada. Accordingly, within each province, for each enterprise, all establishments coded to the same 4-digit NAICS industry groups were grouped to form one sample unit or "provincial enterprise". To reduce response burden on small businesses, only provincial enterprises with at least 20 employees and provincial enterprises with a gross business income of at least $250,000 (According to Statistics Canada's Business Register) were considered in sample selection.

Stratification: Thirty-six industry strata were identified based on groupings of 4-digit NAICS codes. The sample was randomly drawn from the population of provincial enterprises stratified by province and the thirty-six industrial strata (thirty-one industry strata for manufacturing and five industry strata for the selected natural resource industries).

Sample size: As part of a Federal-Provincial Agreement, the sample in Quebec was augmented. A sample of 5,944 provincial enterprises in manufacturing and 674 provincial enterprises in selected natural resource industries was drawn for a total sample size of 6,618 provincial enterprises.

• Additional documentation

Data sources

Data collection for this reference period: September to December 1999

Responding to this survey is mandatory.

Data are collected directly from survey respondents.

Computer Assisted Telephone Interviewing (CATI) was used to collect the survey data. All sample units were "pre-contacted" to determine the name and correct mailing address for the respondent, the Chief Executive Officer (CEO) of the business or the person designated by the CEO. Questionnaires were mailed out with mail, telephone and fax follow ups carried out for to elicit a response from non-respondents.

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

Error detection

Validity and flow edits were built into the data capture system and were applied during data collection and data entry. Validity edits ensured that responses to particular questions fell within a limited range of possible values. Post collection consistency edits were applied to complete questionnaires.

Imputation

Imputation was employed for missing responses to non-mandatory questions. The Generalized Edit and Imputation System (GEIS) software was used to select donors.

There are several cases where the relevance of a subsequent set of questions relies on a response to a preceding question. The ability to proceed along a path of questions was reliant on the nature of the response and the subsequent responses are influenced by the firm behaviour indicated by the response to the preceding question. Block imputation (one donor) was used for these correlated questions as a means to avoid edit failures. For instance, only those who indicated that they had a new or significantly improved product in the first part of Question 12 complete the remaining two sections in the question. Similarly for Question 24, only firms indicating that they undertook R&D would provide information about where and on what frequency the R&D was carried out. For Question 26, respondents would provide the number of patents they had applied for only if they had applied for at least one patent.

Questions 14, 15 and 16 are all related to cooperative and collaborative arrangements. The sub-population of collaborators self-identify in the first part of question 14. Only this sub-population are asked to complete the subsequent series of questions on their reasons for collaborating (the remainder of Question 14), who their collaborators were and where they were located (Questions 15 or 16). When imputation was required, the donor responses were applied to all three questions. This eliminated the possibility of having one donor without collaborations for Question 14 and a second donor with collaborations for Question 15 or 16 which would result in an edit failure.

The group of Questions 17 to 20 were all in reference to the firm's most important innovation. Only firms who described their most important innovation in Question 17 were considered for imputation for Questions 18 to 20. Block imputation was used as these questions are correlated.

Some questions (Questions 8, 10, 13, 25, and 29) contained correlated choices. The components of the response choices were in effect, one question. When there was no response in any part of these questions block imputation was used.

The questions that did not contain correlated parts (Q1, Q2, Q9) were imputed line by line.

The change in total number of employees is a firm-specific phenomenon. During the edit process, any non-response to this question was coded as unknown.

There was no imputation for any of the qualitative questions (Q7, Q11, Q17).

Estimation

The response rate for the survey was calculated as the total number of completed questionnaires as a percentage of the total active, in-scope survey sample. The overall response rate for manufacturing industries was 95%, for a total of 5,455 completed questionnaires. The overall response rate for selected natural resource industries was 94% for a total of 582 completed questionnaires.

Given the low rate of non-response (5%) to the survey, it was decided that it would be reasonable to assume that the characteristics of the non-response population were the same as the respondent population. Accordingly, it was decided that the contribution of non-response to the estimates was to be accounted for by adjusting the sample weights of the respondent population.

Estimates based upon the responses to the survey questions are population estimates; that is, they represent the percentage of businesses in the population that exhibit a particular characteristic. The population estimates are generated through the accumulation of the product of the response variable and the sample weight for the defined tabulation cells.

A sampling error description is provided in the attached document.

• Additional documentation

Quality evaluation

The quality of the data has been checked against quality standards at Statistics Canada, namely, data relevance, accuracy, timeliness, accessibility, interpretability and coherence.

Data relevance was insured by the active collaboration in the questionnaire design between the Science, Innovation and Electronic Information Division (SIEID) of Statistics Canada, Industry Canada, the Institute for Research in Construction of the National Research Council of Canada and Natural Resources Canada.

Data accuracy was insured by conducting cognitive interviews in both official languages with potential respondents. Their comments were integrated into the final design and wording of the questionnaire.

From the close of data collection to the first data release two months elapsed, thus insuring data timeliness.

Accessibility to data users is made through a series of venues, including, the sharing of data with provincial statistical agencies, the Facilitated Access program, working papers available on Statistics Canada's web site, and descriptive tables provided to stakeholders.

To help users interpret the data, the definitions of the underlying concepts, classification, data collection methodology, as well as indicators of data accuracy are made available.

Standard Statistics Canada symbols have been used in all data tables thus insuring data coherence.

Disclosure control

Statistics Canada is prohibited by law from releasing any information it collects that 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.

In order to prevent any data disclosure, confidentiality analysis is done using the Statistics Canada Generalized Disclosure Control System (G-Confid). G-Confid is used for primary suppression (direct disclosure) as well as for secondary suppression (residual disclosure). Direct disclosure occurs when the value in a tabulation cell is composed of or dominated by few enterprises while residual disclosure occurs when confidential information can be derived indirectly by piecing together information from different sources or data series.

Revisions and seasonal adjustment

This methodology does not apply to this survey.

Data accuracy

The response rate for the survey was calculated as the total number of completed questionnaires as a percentage of the total active, in-scope survey sample. The overall response rate for manufacturing industries was 95%, for a total of 5,455 completed questionnaires. The overall response rate for selected natural resource industries was 94% for a total of 582 completed questionnaires.