Atlantic Agriculture Survey
The purpose of this survey is to collect information on livestock and crops. The statistics produced from the survey enable those active in the agricultural sector to observe and assess changes in the industry, measure performance and keep the agricultural community and general public informed of developments.
Detailed information for June 2013
Data release - July 19, 2013 This is the final release of the Atlantic Agriculture Survey Survey. The survey has been cancelled.
The Atlantic Agricultural Survey is a multipurpose agricultural survey with two survey occasions each year in June and November. The June sample collects data from farm operators related livestock inventories and crop areas. The November occurrence collects information on livestock inventories and crop production.
The principal data released include livestock inventories, crop area and production along with summarized supply-disposition tables. These data are used by agricultural industry analysts and producers as they make production and marketing decisions and by government analysts to monitor the livestock and crop sectors or develop agricultural policies in Canada. The data are used in the calculation of farm income estimates and flow to the Canadian System of National Accounts. Further, the data are used in the calculation of net farm income projections, produced by Agriculture and Agri-Food Canada in co-operation with Statistics Canada and the provinces.
- Crops and horticulture
- Livestock and aquaculture
Data sources and methodology
The universe is comprised of all agricultural operations in Newfoundland and Labrador, Prince Edward Island, Nova Scotia and New Brunswick with over $10,000 in sales.
The original development of the Computer Assisted Telephone Interview questions was based on the well established Livestock Survey application and the Crop Reporting application that, in part, are based on the livestock and crop modules on the Census of Agriculture. The Census of Agriculture questionnaire was tested using focus groups and pilot surveys.
The questions used in the ongoing survey have been tested using focus groups. In addition, staff observe training and collection, providing observation reports. All survey occasions include debriefing sessions where the results of the testing and observation are incorporated into the development cycle of the next survey.
This is a sample survey with a cross-sectional design.
The population consists of all farms in Atlantic Canada. The frame includes all agricultural producers, excluding community pastures and farms on First Nations reserves, who reported more than $10,000 on the Census of Agriculture, supplemented by known new hog operations. The stratification and allocation is multi-variate, by type and size of agricultural operation. The sample is selected using a stratified simple random sampling method.
June sample size equals 1,300.
November sample size equals 2,200.
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. Farm operators are contacted directly by phone and, in cases where the operator is difficult to contact, there are multiple follow-up telephone calls that ultimately result in a high contact rate.
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 and reviewing the top contributors to the unweighted and weighted estimates (for each variable in each province).
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). Some missing information is imputed manually during the edit process, and others are imputed using a "hot-deck" donor imputation method. The automated imputation system looks for donors within the stratum and then verifies that the donor record and the record to be imputed are acceptable, and a final review of the imputed data is performed.
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. Raising factor (weight) adjustment is used in the estimation process to account for respondents who could not be contacted or who refused to participate in the survey.
Once the weights have been calculated for each record, any level of required estimates may be obtained using domain estimation (i.e. provincial, agricultural region, etc.).
The survey results are evaluated through comparisons to previous estimates and other sources when available. Biological factors are used as a guide when evaluating the data or comparing to other data sets. A primary tool in the evaluation and final determination of the data involves supply-demand analysis and survey-based models that track the supply and demand of the particular commodity by province over time.
The survey results are analysed and corrected before the data are used to analyse the industry and fine-tune the estimates. The survey data are reviewed in a board environment before the commodity analyst works with the data primarily using supply-disposition analysis. The results of the industry analysis are reviewed by the board before being sent to the individual provinces. Once the data are finalised they are released to the public and published.
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.
Revisions and seasonal adjustment
Once every five years the published data are aligned with the results of the Census of Agriculture. Due to conceptual differences between the datasets, the match is not normally 1 to 1. For instance, the 2001 Census was conducted on May 15 and the livestock statistics refer to June. Any adjustments made to the data during the Census year are then smoothed in over the historical five-year period between the Censuses. The impact of the revisions is normally less than 5%, however, for specific items in certain provinces, the impact can be higher.
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 high level of accuracy for provincial level totals.
The following ratings are based on the value of the coefficient of variation (CV). Using this rating system, the vast majority of the total estimates at the provincial level are excellent or very good. This is particularly true in provinces where the livestock type is prevalent. If they are not excellent, the results are normally very good.
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 ranging normally from 97 to 99%, and item non-response is low. Overall, the impact of the edits and imputations is small. The data are generally of excellent quality, with some under-coverage due to intercensal frame degradation (i.e. coverage of new operations).
- Date modified: