Data quality, concepts and methodology: Concepts and variables measured

Price coverage

The FPPI includes all farm products sold off the farm, which are estimated in the farm cash receipts series published in Farm Cash Receipts - Agriculture Economic Statistics, catalogue no. 21-011-X.

Prices are based on either administrative data sources, or monthly surveys of agricultural producers or commodity purchasers. Where possible, administrative data sources are used to reduce response burden. The commodity price is collected at point of first transaction, where the fees deducted before a producer is paid are excluded (e.g., storage, transportation and administrative costs), but any bonuses and premiums that can be attributed to specific commodities are included. Commodity-specific program payments are not included in the price.

Administrative price data come from a wide variety of sources. Many are collected directly from marketing boards, for example, the Ontario Wheat Producers Marketing Board and the Nova Scotia Grain Marketing Board; market associations such as CANFAX. Some are collected and processed by provincial agricultural or statistical departments. Some data are generated from the regulatory activities of Agriculture and Agri-Food Canada and the Canadian Wheat Board.

Where administrative data are not available, prices are collected using the monthly Farm Product Prices Survey, which consists of a series of surveys. There are two farm surveys: the monthly farm Potato Prices Survey and the quarterly farm Ontario Tame Hay and Straw Prices Survey. The reference date for the farm surveys is the 15th of each month and prices are collected by telephone during the five working days following the 15th.

There are two monthly company surveys: the Non- Board Grains and Speciality Crops Survey in the Prairies and the Grain Survey in the Maritimes. Administrative data, the Non-Board Grains and Specialty Crops Survey and the Maritime Grain Survey cover activity for the whole reference month, producing average weighted prices that reflect grades marketed.

There are some commodities within the farm cash receipts series that do not have prices; there are only values. These are generally smaller commodities such as maple, forest, floriculture and nursery products, and the other fruit and vegetable categories. The farm cash receipts for these unpriced items are used to establish the basket shares of these items, while their price movement comes from the aggregate index under which the item falls. For instance, the total vegetable index would be used to proxy the “other vegetable” category.

Data accuracy

The methodology of the index and the price series which construct the index have been designed to control error and to reduce the potential effects of these. However, both administrative and survey data are subject to various kinds of errors. Survey data are mainly subject to response and data capture errors. In reporting prices each month, farm survey respondents are asked to report the average prices prevailing in their neighborhood, taking into account the various grades of each commodity marketed. Thus, average prices reported by these respondents may differ from month to month due to changes in price, quality or both.

Every effort is made to ensure that administrative data are conceptually correct for the use to which they are put. Users should be aware that administrative data are generally compiled for an organization's own needs, and not for survey purposes. Any anomalies or inconsistencies detected are verified with the source, and where necessary, adjustments are made to reconcile data with the conceptual framework of our series. The administrative agencies used are considered to be the best sources available, and data received from them is judged to be of very good quality, even in those circumstances where adjustments have been made. Much price data come from the accounting departments of various marketing agencies, and are originally used to issue cheques to producers, are audited, and so quality is deemed to be excellent.

Comparability of data and related sources

The current methodology (a chained index with monthly baskets) was applied to the index starting with 1992; there was no change in methodology for the indexes before 1992. Although the index levels of the 1997=100 series will be different from those of the 1986=100 series, they will continue to show the same percent changes for the period ending in December 1991. The index prior to 1992 was constructed as a chain fixed-basket index.

Since the index is chained annually, measures of twelve-month price change are not measures of pure price change, as they would be in a direct seasonal basket index, but are distorted by basket shifts.

However, since we calculate unlinked indexes for year y-1 and y based on the basket for years y-6 to y-2, we use this unlinked index to derive a measure of pure price change between any month m of year y-1 and the same month of year y.

The current index includes all farm cash receipts in the basket, whereas the previous index, prior to 1992, had been based upon products that account for about 95% of total farm cash receipts from farming operations. Commodities included in the index vary by province, reflecting the regional agricultural production. For instance there are no special crop sub-indexes in the Maritime provinces.

Other quality indicators and assessments

Statistics Canada had been using the 1980 Standard Industrial Classification system (SIC) for collecting, compiling and disseminating economic statistics on Canada’s businesses. Beginning with the 1997 reference year, the North American Industry Classification System (NAICS) was introduced. It adds hatcheries and farms with only Christmas trees to the agricultural sector. The index base weights used for the commodities affected by NAICS were adjusted to reflect the new classification in 1997.

Previous