Canadian Commercial Aerospace, Defence, Industrial Marine and Industrial Security Sector Survey
Detailed information for 2007
Status:
Active
Frequency:
Irregular
Record number:
2933
The objective of this survey is to produce statistical information on the revenues, costs, investments, employment and markets of the firms engaged in aerospace, defence or marine manufacturing or in the security industry in Canada.
Data release - November 9, 2009
Description
The objective of this survey is to produce statistical information on the revenues, costs, investments, employment and markets of the firms engaged in aerospace, defence or marine manufacturing or in the security industry in Canada. The data from this project will be used by Industry Canada to determine the competitive position of the sector and its needs in terms of federal government programs.
The survey is sponsored by the Aerospace, Defence and Marine Branch of Industry Canada. The data from this project will be used by Industry Canada to gain a better understanding of the current status of these sectors, as well as to develop policies and programs that will support the advancement of businesses in these sectors.
Reference period: Fiscal year
Subjects
- Manufacturing
- Transportation equipment
Data sources and methodology
Target population
The target population comprises enterprises engaged in the manufacture and delivery of products and services related to the defence, aerospace and commercial and civil marine sectors in Canada. This population includes enterprises belonging to a diverse group of industries classified under the North American Industry Classification System (NAICS).
Instrument design
The paper questionnaire was jointly developed by Industry Canada and Statistics Canada. The questionnaire from the 2002 survey was updated and revised in conjunction with Industry Canada. This new questionnaire was then tested with potential respondents in the spring of 2008 - 6 English respondents in Toronto, and 6 French respondents in Montreal. The questionnaire was then revised once again based on the comments and suggestions of the companies in the questionnaire testing.
Sampling
This survey is a census with a cross-sectional design.
The survey frame consists of 1,547 units in the Canadian Commercial Aerospace, Defence, Industrial Marine and Industrial Security Sector. Statistics Canada conducts a census of all units on the survey frame.
Data sources
Data collection for this reference period: 2008-09-18 to 2008-12-15
Responding to this survey is mandatory.
Data are collected directly from survey respondents.
A mailout/mailback questionnaire is used. A minimum of 5 follow-up attempts by phone are made to non-respondents to encourage them to complete the questionnaire.
View the Questionnaire(s) and reporting guide(s) .
Error detection
Consistency edits were applied during post-collection data processing. These edits ensure, for example, that the sum of total exports plus domestic sales is not greater than the value of total sales, or that the sum of sales to the various subsectors is equal to the value reported for total sales. If records failed one or more edits, they were reviewed to correct the problems.
Imputation
In order to correct for item non-response, imputation was done via nearest-neighbour donor imputation. The nearest-neighbour imputation strategy uses a matching process to find a similar responding record (donor) for each non-respondent (recipient) and missing values are copied from donor to recipient. Similar records are found using a distance that takes into account the collection priority level and the activity sectors assigned to each unit. In some cases the imputed values are adjusted to ensure that the resulting microdata record is consistent (i.e. sum of parts equal to totals). Statistics Canada has developed a generalized software application called BANFF which was used to process the data for this survey.
In general, the amount of imputation required was very low in this survey. The attached Table 3 gives a summary of the imputation rates for each section of the questionnaire. Since each section is comprised of many cells, the values presented in the table are the median and maximum cell level imputation rates within each section.
Estimation
Since the survey was actually conducted as a census of all units on the frame, the only weighting that was applied to the micro-data was to correct for unit non-response. Weighting was applied such that responding units were used to represent similar non-responding units, where a similar unit was one in the same sector class and priority class, as determined by the sector flags and the collection priority flags provided by Industry Canada. For example, if there were 50 units on the frame in a particular sector flag by collection priority class, and only 30 of them responded to the survey, the weight used for each respondent record in this class is 50 / 30 or 1.60. The weight is applied to each respondent record in estimating population totals, means and ratios to ensure that every unit on the frame is represented properly whether they respond or not. As well, if the response rates differ between classes, calculating the weights within classes ensures that this does not introduce a bias. Effectively, the set of respondents is treated as a random sample from the population and these units are assigned weights so that they can be used to draw conclusions on the entire population. This also allows for estimation of quality indicators based on variance estimates (expressed as standard errors and co-efficients of variation) that can be used to assess the reliability of the survey results. The Generalized Estimation System developed by Statistics Canada was used to produce the weighted population estimates and their associated variance estimates.
Quality evaluation
The reported data are compared to previous results of the survey to evaluate the quality of the data. Industry Canada has detailed knowledge of the sector and the companies therein, and is able to provide feedback on their activities to help identify concerns relating to the data quality.
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.
Confidentiality analysis includes the detection of possible "direct disclosure", which occurs when the value in a tabulation cell is composed of a few respondents or when the cell is dominated by a few companies. 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
There are two types of errors which can impact the data: sampling errors and non sampling errors. Since this was a census survey, no sampling was done and therefore there were no sampling errors.
Non-sampling errors are all types of errors that alter the estimated values. For example, when respondents only answered some questions, imputation had to be used to compensate for the remaining values. Another problem is that respondents can interpret and answer the questions differently than expected. It can be difficult to identify these types of errors in many cases. Subject matter expertise may be the only way to detect those errors.
The survey frame was built by Industry Canada, who selected the companies to appear on the frame. The frame does not include all of the businesses active in the sectors of interest for this survey, but Industry Canada believes that those businesses that are not included do not make a significant contribution to the revenues and employment of the sectors of interest. If it turns out that many larger businesses were, in fact, not included in the survey, it would not be possible to say that the estimates do in fact essentially represent the sectors they were designed to represent. This would contribute to the overall non-sampling error. Since the frame was not created by Statistics Canada, Statistics Canada cannot evaluate its properties in terms of coverage (overcoverage or undercoverage) or the quality of the classification variables used in stratification and domain estimation. All inferences are made based on the assumption that the frame matches exactly to the target population, and does not contain classification errors (i.e. province, firm type, etc.).
The response rate of the Canadian Commercial Aerospace, Defence, Industrial Marine and Industrial Security Sector Survey, 2007 was 58.5%.
- Date modified: