Data quality, concepts and methodology: Methodology
Archived Content
Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.
Introduction
Statistics Canada has conducted an annual survey of the greenhouse industry since 1955, of the nursery industry since 1956 and of the sod industry since 1982. The survey is designed to provide an indication of the economic situation of the primary greenhouse, sod and nursery stock production industries. The survey is designed to provide insight into the general economic situation of the industry. To this end, data are provided on market structure and production inputs.
The survey is designed to collect information on area planted, labour, payroll, cost for growing-on and for resale, gross revenues, as well as the distribution of gross sales by channel of distribution.
The greenhouse target population includes all of the commercial greenhouse growers as identified by the 2006 Census of Agriculture and reconciled with the previous known population. For the sod and nursery operations, the survey focuses on operations that grow some or all the commodities they sell. As a result, firms that only distribute sod or nursery stock, or just landscape and lay sod are excluded from the survey. The survey is also designed to estimate the total value added of the agricultural side of nursery/sod but also picks up value added from the non-growing (i.e. marketing) portion of each operation. Note too that value added from landscaping and laying sod is excluded from the revenue estimates even when the responding firm grows nursery stock or sod. The survey estimates for greenhouses, sod farms and nurseries are calculated using probability sampling methods.
Concepts and definitions
Greenhouse industry
Prior to 1986, the target population included all commercial greenhouse growers. In 1986, the target population was redefined to include only commercial greenhouse growers with a growing area of 7,500 square feet or more.
In 1992, it was decided to include all of the commercial greenhouse growers as identified by the 1991 Census of Agriculture and reconciled with the previous known population. Due to the changed methodology, the estimates since 1992 are not comparable with the corresponding 1991 estimates.
A greenhouse can be defined as an operation where plants are grown under glass, plastic or similar type of protection. Since 2008, we also ask respondents to report area under poly-film or polyethylene.
A breakdown of sales, indicating channels of distribution and flower varieties sold, assists in the determination of the market structure. All sales figures published represent prices paid to the grower. Data collected on important inputs of production include the number of employees, gross yearly payroll, total investment and area operated under glass and plastic.
Greenhouse operating expenses include payroll, plant material purchases, heating, lighting, ventilation costs, other crop expenses (fertilizer, pesticides, and growing mediums) and other operating expenses (interest, land taxes, insurance, packaging, repairs to farm buildings, machinery, equipment and vehicles, and contract work). Since 2008, electricity expenses for lighting, airflow fans and heating are reported inside a specific expense item.
The concept for full-time employment was revised for Greenhouse, Sod and Nursery (GSN) 2003 data in accordance with the definition used at Statistics Canada and endorsed by the International Labour Organization. Full-time employment consists of persons who usually work 30 hours or more per week at their main or only job. Prior to 2003, full-time employment was defined as persons working 8 months or longer and part-time employment consisted of persons working less than 8 months of the year. Since 2008, respondents to the survey must report the number of seasonal workers (less than 8 months contract) and permanent (more than 8 months contract).
The tables for cut flowers, potted plants, cuttings and propagating material have been updated to reflect changes in the industry.
Nursery and sod industry
Before 1992, estimates essentially pertained to commercial operations that allocated a portion of their surface area to sod or seedlings and had annual gross sales of $75,000 or more. Since 1992, estimates have pertained to nurseries with 5 acres or more, and sod operations of 40 acres or more in regions other than the Atlantic (10 acres or more in the Atlantic region).
The survey estimates in previous years were simply a tabulation of all responding operations fitting the above two criteria. Due to the changed methodology, the 1992 and subsequent estimates are not directly comparable with the corresponding 1991 estimates.
The survey collects dollar data on two commodities: sod and nursery stock. Sod is a standard commodity whereas nursery stock requires precise definition. The term "nursery stock" covers a diverse range of non-edible, living plant material grown in field or in containers and sold with its root system intact. Nursery stock includes annual and perennial plants ranging from woody plants (e.g. trees, shrubs and rose bushes) to bedding plants and potted outdoor flowers.
The end purpose of the plants may be ornamental or functional (e.g. fruit trees, grape vines and forestry seedlings). Plant sizes range from seedlings to fully grown trees. "Nursery stock" excludes Christmas trees and various inputs such as growing medium (e.g. peat moss and top soil), fertilizer and pesticides. Since 2008, nursery and sod operating expenses are separated.
Methodology and data quality
Survey frame
The survey frame has been designed to include all operations cultivating greenhouse, sod and nursery products based on the latest Census of Agriculture. Institutional farms, farms on Indian reservations, Hutterite colonies as well as operations from Yukon, Northwest Territories and Nunavut were excluded from the survey. After these exclusions, the remaining farms represent the initial survey frame.
Observations on the 2011 survey frame were extracted from the 2010 Annual Greenhouse, Sod and Nursery Survey frame and the 2006 Census of Agriculture, to which were added some operations coming from external sources. The new operations are either new farms since the 2006 Census or farms missed by the 2006 Census.
Excluded from the survey frame were areas of glass or plastic greenhouses maintained for non-commercial plants. Also excluded were greenhouse-only operations in Quebec, Ontario and British Columbia with less than 5,000 square feet and sales of less than $10,000. Greenhouse-only operations in the remaining provinces with less than 1,000 square feet were excluded from the survey frame. Furthermore, sod-only operations with less than 40 acres (10 acres in the Atlantic) and nursery-only operations with less than 5 acres were excluded from the survey frame.
Sample selection
For each province, four strata were defined. The first stratum includes operations with a greenhouse area of less than 1 acre, sales of less than $110,000, no nursery or sod area, and that reported flower production in the 2006 Census of Agriculture. The second stratum includes all operations with a greenhouse area of less than 1 acre, sales of less than $110,000, no nursery or sod area, and that did not report flower production in the 2006 Census of Agriculture. The third stratum includes all operations with a greenhouse area of less than 1 acre, sales of less than $110,000, nursery area greater than zero and less than 5 acres, and no sod area. All remaining operations were placed in the fourth stratum. Within each province, operations were sampled in the first three strata and a census was taken of all units in the fourth stratum. For each stratum where sampling was done, the minimum sample size was 15 and the maximum design weight was 5.
Overall, the sample size was determined in order to achieve a target CV of 0.01 (1%) for sales, greenhouse area and nursery area at the provincial level. The final sample size for this survey was 5,342.
For this survey, we used two questionnaire versions: one full version requesting all the required information and one abridged version that essentially only requests totals. Following the sample selection, we allocated one of the two versions of the questionnaire to each operation sampled. The allocation was done based on province of operation and sales volume. We allocated the full version to 3,913 operations and the abridged version to 1,429 operations.
Data collection
The reference period for this survey is calendar year 2011. Respondents received a copy of the questionnaire through the mail in early December 2011 and most were interviewed by telephone between January 9 and January 29, 2012.
Data collection for the 2011 Annual Greenhouse, Sod and Nursery Survey was undertaken using a "Computer Assisted Telephone Interview" (CATI) system. Some respondents, who have previously mentioned that they did not want to be interviewed over the phone, mailed back the questionnaire to the Ottawa Head Office.
During the data collection period, some respondents asked to not be interviewed over the phone. They were asked to complete the questionnaire and send it back to the Ottawa Head Office.
Response rates
Information was received from 83.0% of the sample. Non-responding units (i.e. no-contacts and refusals) are dealt with by adjusting the initial sample weights (raising factor adjustment).
Edit and imputation
With the introduction of the CATI methodology, it is now possible to implement edit procedures at the time of the interview. Computer programmed edit checks in the CATI system inform interviewers during the interview of possible data errors, which can then be corrected immediately by the interviewer and respondent. The CATI system significantly reduces the need for subsequent telephone follow-up, thereby reducing respondent burden and survey processing time. Erroneous data and partially completed questionnaires are imputed using historical data.
Sampling and non-sampling errors
The statistics contained in this publication are based on a random sample of agricultural operations and, as such, are subject to sampling and non-sampling errors. The overall quality of the estimates depends on the combined effect of these two types of errors. Sampling errors arise because estimates are derived from sample data and not from the entire population. Sampling error is quantified using the coefficient of variation (CV). The lower the CV, the smaller the sampling error.
Non-sampling errors are errors which are not related to sampling and may occur throughout the survey operation for many reasons. For example, non-response is an important source of non-sampling error. Coverage, differences in the interpretation of questions, incorrect information from respondents, errors in recording, coding and processing of data are other examples of non-sampling errors.
Estimation
We weighted the information collected from the respondents to produce estimates representative of all of the operations in the greenhouse, sod and nursery industry in our survey frame. We produced estimates at the national and provincial level for different totals and sub-totals of industry variables. During the weighting, we took into consideration both versions of the questionnaires and ensured that all of the estimates produced for all variable types remained representative of all of the operations in the survey frame.
Revisions
For each publication, we release a set of preliminary estimates for the current reference year and a set of revised estimates for the previous reference year.
Data quality
The estimates in this publication are based on a probability sample of farming operations. The potential error introduced by sampling can be estimated from the sample itself by using a statistical measure called the coefficient of variation (CV). Over repeated surveys, 95 times out of 100, the relative difference between a sample estimate and what would have been obtained from an enumeration of all farming operations would be less than twice the coefficient of variation. The sample estimate plus or minus twice the CV is referred to as the confidence interval.
For the 2011 Annual Greenhouse, Sod and Nursery Survey, CVs at the Canada level range generally from 2% to 10% for the variables that are more frequently reported (area, investment, expenditures, sales, etc.), which makes them very reliable. Some variables on plant, flower and tree varieties have CVs greater than 25%. The coefficients of variation are available upon request.
To improve data qualification, the symbols E and F have been introduced into the publication. An E means that the data are widely dispersed and that the estimate should be used with caution. An F replaces the estimate where the data are too unreliable for publication.
- Date modified: