Canadian Agricultural Partnership Survey (CAPS)

Detailed information for 2018-2023

Statistics Canada is conducting this survey in collaboration with Agriculture and Agri-Food Canada (AAFC), to produce statistical information on agricultural operations receiving funding through the cost-shared program between the federal and provincial or territorial governments of the Canadian Agricultural Partnership (CAP) from 2018 to 2023.

Data release - May 31, 2024

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

Description

The data will be used to better understand the characteristics of businesses engaged in agricultural operations, as well as to demonstrate outcomes and impacts of the cost-shared envelope of CAP on the participants, their businesses, and the overall agricultural sector. The results of this survey will provide a better understanding of how the CAP program serves agricultural operators which will allow the program to serve Canadians more effectively.

The CAP program and activities focus on helping the agriculture and agri-food sector deliver the greatest benefits for farmers, food processors and Canadian families to:
- Grow trade and expand markets to seize key opportunities and address emerging needs.
- Advance science and innovation, with an emphasis on innovation and sustainable growth.
- Better reflect the diversity of our communities, enhance collaboration across different jurisdictions and secure and support public trust.

Collection period: Early September until early November

Subjects

• Agriculture and food (formerly Agriculture)
• Business performance and ownership

Data sources and methodology

Target population

The target population of this survey is agricultural operations who received cost-shared funding through the Canadian Agricultural Partnership program between 2018 and 2023.

Instrument design

The Canadian Agricultural Partnership Survey was designed in collaboration with Agriculture and Agri-Food Canada. It is used to collect information on the impact of the of Canadian Agricultural Partnership.

The questionnaire was field-tested with potential respondents in 2023 and their comments on the design and content have been incorporated.

Sampling

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

Sampling unit:
Enterprise

Stratification method:
The survey uses a stratified random sampling for the sample. Units on the frame are stratified by province, provincial program and type of operation.

Data sources

Data collection for this reference period: 2023-09-05 to 2023-11-03

Responding to this survey is voluntary.

Data are collected directly from survey respondents.

The respondents are mailed or e-mailed a secure access code to respond to the electronic questionnaire.

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

Error detection

Error detection is an integral part of both collection and data processing activities. Automated edits are applied to data records during collection to identify reporting and capture errors. During data processing, other edits are used to automatically detect errors or inconsistencies that remain in the data following collection. These edits include value edits (e.g., value > 0, value > -500, value = 0), linear equality edits (e.g., value 1 + value 2 = total value), linear inequality edits (e.g., value 1 >= value 2) and equivalency edits (e.g., value 1 = value 2). When errors are found, they can be corrected using imputation. Extreme values are also flagged as outliers using automated methods based on the distribution of the collected information. Following their detection, these values are reviewed to assess their reliability.

Manual review of other units may lead to additional outliers identified. These outliers are excluded from use in the calculation of ratios and trends used for imputation and during donor imputation. In general, every effort is made to minimize the non-sampling errors of omission, duplication, misclassification, reporting and processing.

Imputation

When respondents do not completely answer the questionnaire, or when reported data are considered incorrect during the error detection steps, imputation is used to fill in the missing information and modify the incorrect information. Many methods of imputation may be used to complete a questionnaire, including manual changes made by an analyst. The automated statistical techniques used to impute the missing data include deterministic imputation, replacement using auxiliary information available from other sources, replacement based on known data relationships for the sample unit and replacement using data from a similar unit in the sample (known as donor imputation). Usually, key variables are imputed first and are used as anchors in subsequent steps to impute other related variables.

Donor imputation is performed in classes, and those classes are created to obtain as homogeneous a pool of donors as possible for the variable being imputed.

Imputation generates a complete and coherent microdata file that covers all survey variables.

Estimation

The response values for sampled units were multiplied by a final weight in order to provide an estimate for the entire population. The final weight was calculated using a certain number of factors, such as the probability for a unit to be selected in the sample, and adjustment of the units that could not be contacted or that refused to respond. Using a statistical technique called calibration, the final set of weights is adjusted in such a way that the sample represents as closely as possible the entire population.

Sampling error was measured by the standard error (SE) for proportions and by the coefficient of variation (CV) which represents the proportion of the estimate that comes from the variability associated to it. The SEs and CVs were calculated and are indicated in the data tables by quality flags.

Quality evaluation

Prior to the data release, survey results were analyzed for quality. In general, this analysis included a detailed review of individual responses.

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.

Direct disclosure occurs when the value in a tabulation cell is composed of or dominated by few establishments 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 type does not apply to this statistical program.

Data accuracy

Because the data are based on a sample, they are subject to sampling error. Estimates based on a sample vary from sample to sample and are typically different from the results that would have been obtained from a complete census with a 100% response rate.

Quality indicators were calculated using Statistics Canada's general estimation system, G-EST. The method used to calculate these indicators differed depending on the component.

Response rate

The response rate is 64.8%.

Non-sampling errors

Common sources of non-sampling errors include imperfect coverage, classification errors and non-response. Coverage errors, or imperfect coverage, arise when there are differences between the target population and the surveyed population. These differences can be caused by exclusion of units on the frame or inclusion of units outside of the target population in the sample. If the excluded population differs from the survey population, the results may be biased. In general, since these exclusions are small, one would expect the biases introduced to be small.

Non-response could occur during main collection. Survey estimates were adjusted (i.e., weighted) to account for non-response cases. Other types of non-sampling errors can include response and processing errors.

Non-response bias

The main method used to reduce non-response bias involved a series of adjustments to the survey weights to account for non-response as much as possible. A significant effort was made to minimize bias by using a well-tested questionnaire, a proven methodology, specialized interviewers and strict quality control.

Coverage error

The survey frame was created from external lists that were linked to the Business Register. If there is a difference in response between linked and unlinked units, there could be coverage error.