Farm Financial Survey (FFS)

Detailed information for 1999

This survey collects data on farm operations including land use, capital investments, capital sales, assets, liabilities, borrowings, income and expenses.

Data release - December 20, 2000

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

Description

This survey collects data on farm operations including land use, capital investments, capital sales, assets, liabilities, borrowings, income and expenses. Agriculture and Agri-Food Canada uses the data to examine the effects of agriculture programs and policies on different types of farm operations by province. Farm Financial Survey (FFS) data contribute some measures of assets and liabilities as well as capital investment and capital sales to programs within Statistics Canada. Prior to 1994 the Farm Credit Corporation conducted the survey on an ad hoc basis (Farm Credit Corporation Survey - FCC) starting in 1981.

Reference period: Calendar year

Collection period: March to April

Subjects

• Agriculture and food (formerly Agriculture)
• Farm financial statistics

Data sources and methodology

Target population

The target population for the survey consists of all Canadian agriculture operations that are active at the end of the reference year. Specific farms are excluded from the target population, such as farms with less than $10,000 in sales from agricultural activities; institutional farms; community pastures; farms on First Nations reserves; and farms that are part of multi-holding companies.

Instrument design

The original development of the survey questionnaire was based on agricultural financial balance sheet and income statement concepts, and it was pilot tested. There have been few changes to these core items. The questions on farm management practices were selected from a scale that had been extensively tested using focus groups and pilot tests. The environmental questions were informally tested within Statistics Canada and Agriculture and Agri-Food Canada. The questions on farm management practices and environmental practices are voluntary.

Team members observe training and collection and provide observation reports. Interviewers and collection managers participate in a telephone debriefing, and also provide answers to a set of written debriefing questions. All comments are considered in the following development cycle.

Sampling

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

Using a list of all farms from the 1996 Census of Agriculture and tax files, a list is created and stratified by province, by farm type and by farm size. The sample size is about 20,000 farms allocated to the strata to ensure reliable provincial estimates for key financial variables, for every farm type. The sample is then selected using Poisson sampling. This method involves computing a sampling fraction for each stratum, then each listed farm in the stratum is given a permanent random number. If the random number falls in the selection interval defined by the sampling fraction, then the corresponding farm is selected for the sample.

There is limited overlap of sample for each collection.

Data sources

Responding to this survey is mandatory.

Data are collected directly from survey respondents.

The survey information is collected by telephone interview in Statistics Canada regional offices, using a Computer Assisted Telephone Interview (CATI) application. Questionnaires are mailed to the farm operation a few weeks prior to collection for reference during the interview. Interviews are scheduled with the farm operator to reduce follow-up procedures.

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

Error detection

The CATI application used for collection contains range and consistency edits and help text. A set of reports is run to identify problem items early in collection for remedial action (e.g. variables with a significant number of edits or missing information). Processing includes checking interviewer notes, manually reviewing significant inconsistencies, reviewing the top contributors to the unweighted and weighted estimates (for each variable in each province).

Imputation

Total non-response (e.g. refusals and no contacts) is accounted for by weighting adjustments to each stratum. Some item non-response is estimated deterministically (using other information in the respondent's questionnaire). Missing information is similarly imputed manually during the edit process, and others are imputed using historical information, as well as a donor imputation method. The automated imputation computer system looks for donors based on farm size, farm type and region, and a final review of the imputed data is performed.

Estimation

Raising factor adjustment is used in the estimation process. The response values for sampled units are multiplied by a sampling weight in order to estimate for the entire surveyed population. The sampling weight is calculated using a number of factors, including the probability of the unit being selected in the sample.

Once the weights have been calculated for each record, any level of required estimates may be obtained using domain estimation (i.e.: Canada, provincial, crop districts, farm type, revenue class, etc.).

Quality evaluation

The CATI application contains edits used during collection that enables interviewers to correct errors immediately. Each survey step produces reports that are reviewed and that indicate edit failure counts, missing data counts, and percent impact of imputations on the estimates.

The survey results are evaluated through comparisons to other sources of farm financial information, such as the Agricultural Economic Statistics series, and the Census of Agriculture to assess the quality of the data.

Disclosure control

Statistics Canada is prohibited by law from releasing any data which would divulge information obtained under the Statistics Act that relates to any identifiable person, business or organization without the prior knowledge or the consent in writing of that person, business or organization. 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.

The program Estimex is used to produce all estimates. It automates confidentiality suppression by rounding the number of farms for each estimate to the nearest 5, and also performs rounding on the associated estimates of the characteristics. Cells are suppressed for estimates with a weighted farm count of less than 20 farms.

Revisions and seasonal adjustment

This methodology does not apply to this survey.

Data accuracy

Sample surveys are designed to provide the highest sampling efficiency (the smallest sample that will produce a sampling error of a given size). This optimization is usually performed for only a few variables, limited by the data items that are available at the time of sample design and selection, the resources available, and the complexity introduced by trying to optimize for many variables at one time. The sample used for these statistics was designed to produce a reasonable level of accuracy for assets and revenue by province and farm type. Consequently, other items may be less accurately estimated.

All estimates are accompanied by an alphabetic code from "a" to "f" that indicates the degree of reliability of each estimate. The codes are based on the value of the coefficient of variation (CV). The CV ratings for this survey are:

Letter Rating CV Rating
A 0.00% to 4.99% Excellent
B 5.00% to 9.99% Very good
C 10.00% to 14.99% Good
D 15.00% to 24.99% Acceptable
E 25.00% to 34.99% Use with caution
F =35.00% Too unreliable to publish

The variability in the estimates can be obtained by constructing confidence intervals around the estimate using the estimate and the coefficient of variation.

The overall response rate of the survey is very good at 80%, and item non-response is low. Overall, the impact of the edits and imputations is moderate. The data are generally of good quality, with some under-coverage due to intercensal frame degredation (i.e. coverage of new operations).