Alberta Retail Price Survey (ARPS)

Detailed information for October 2018




Every 2 years

Record number:


Statistics Canada is undertaking this survey to collect retail prices in 35 Alberta communities for the production of price indexes. These price indexes will estimate the difference in the retail cost of an identical (or very similar) group of products in each community and the cost of this same group of products in Edmonton.

Data release - May 29, 2019


To gather price information used by the Government of Alberta. This information will help determine the Living Cost Differential of 34 communities in Alberta compared to Edmonton.

The Government allowance and spatial indexes section of Statistics Canada is responsible for the computation of indexes of comparative retail prices and living costs, to support the statistical data needs of the Government of Alberta, Office of Statistical Information.

Reference period: October of the reference year


  • Prices and price indexes

Data sources and methodology

Target population

Select retail sector establishments in 35 specific communities in Alberta.

Instrument design

This paper questionnaire is a retail price survey used by interviewers/price collectors to record prices of a specified basket of products and services.


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

A judgment-based sampling approach was used to determine which retail outlets would be selected for price collection of products (goods and services) for the ASPI survey.

Sampling unit:

Stratification method:
Geographic stratification based on 35 communities in Alberta

Sampling and sub-sampling:
At least two establishments by type of retail outlet(e.g. retail food store, pharmacy/drug store, video rental outlet, service station, automobile repair shop, barber/beauty salon, dry cleaning services, coin laundry, taxi service, liquor/beer outlet, home appliance outlet and restaurant)are selected by interviewers to price products and services in each community.

Data sources

Data collection for this reference period: 2018-10-01 to 2018-10-31

Responding to this survey is voluntary.

Price collectors record prices in each selected retail outlet and capture prices and product detail on a paper questionnaire. A letter signed by the regional director of statistical survey operations accompanies price collectors when they enter each retail establishment to provide store managers with additional information on the objective, purpose and method of price collection. Prices collected in grocery stores can take up to four hours to identify and record in the paper questionnaire. Other types of outlets would require less time to complete as the number of prices collected are fewer than the number of prices collected in grocery stores.

Administrative data including retail scanner data is used to supplement field price collection. The data is used for statistical purposes with Statistics Canada making changes to adjust for differences in comparability (e.g. unit cost conversion calculations using administrative data).

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

Error detection

Error detection is conducted during post collection processing, using a set of error detection procedures to identify outliers and possible reporting anomalies. Records that fail these edits are reviewed for editing and correction when necessary.


Missing data are imputed using alternative sources of data or nearest neighbour imputation. In such cases, missing data are imputed using the average prices of homogeneous products within the same category or a nearby geographic area.


There are a total of twenty aggregate price indexes estimated for each community . Of these, three are major aggregate spatial price indexes:

1. all-items
2. food
3. non-food

The food and non-food major aggregate price indexes are further decomposed into seventeen sub-aggregates by commodity group, using a commodity classification unique to the program.

Price indexes are constructed through successive phases of aggregation from the lower level towards the higher level price aggregation. Lower level aggregation is at the level of uniquely defined products (goods and services) such as varieties of milk (homogenized and 2%, UHT, canned evaporated milk, and instant powdered skim milk), whose prices are sampled from retail outlets. Individual price relatives of product varieties in the community and Edmonton were aggregated using a geometric mean formula to arrive at an unweighted price index for the elementary aggregate of the product, in this case, milk.

Higher level price indexes were produced by aggregating lower level price indexes and weighting them with the relevant consumption expenditures of the average consumer (CPI basket weights). The rule of aggregation is the weighted sum of the lower level price indexes of products.

Once product varieties have been matched between locations, price relatives are calculated for each product variety. Given a comparison location (A), reference location (B), product variety (i) and price (p), the price relative of the product variety (p_i) between the two locations is:


The ratio describes how much the price of product variety (i) at location A differs from location B.

The computed spatial price relatives for a set of similar products, are used to calculate an unweighted price index for the given set of similar products, using the geometric mean formulae.

Individual price indexes are aggregated to the next level higher in the classification structure, the commodity group (e.g., from fresh milk to "Dairy products"). The different commodity group indexes are in turn aggregated to produce the all-items price index.

Since not all products have the same level of importance in the consumption basket of a consumer, higher level price aggregates for a commodity group price index are constructed by weighting each price index by the relevant expenditure weight of the specific product.

Expenditure weights used for constructing the spatial price indexes were derived from the spending patterns of consumers in Edmonton, as reported by the 2016 edition of the Survey of Household Spending (SHS). The data were normalized to account for the size and composition of the selected products used in the analysis.

Quality evaluation

An in-depth assessment of quality is conducted prior to the dissemination of estimates. This assessment is based on two key elements of quality (accuracy and coherence); as defined in Statistics Canada's guidelines for the validation of statistical outputs.

Analysis of price index differences compared to the previous cycle at the community and product level is done to ensure coherence. In addition, validation of key contributors to price change at the product level is undertaken to identify any divergence from expectations. Contextual analysis of survey results is also performed based on prevailing economic conditions to facilitate informed analysis.

Disclosure control

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 transformations are applied to collected price data in the calculation of price indexes, such that it is not possible to identify the raw price data obtained from any survey participant. Confidentiality rules are also applied to price indexes 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

The indexes are not revised nor seasonally adjusted.

Data accuracy

The statistical accuracy of this index depends on price and weight data obtained from sample surveys. Each type of input data is subject to its own errors. Processing procedures for editing and imputation help ensure the quality of data. Consequently, the aggregate indexes at all levels are considered to be statistically reliable.

Response rates:
The survey is expected to receive at least a a 70% response rate based on a combination of the number of communities and products within each community that price collection activities will take place in and uses a survey methodology designed to control errors and reduce their effect on estimates. However, the survey results remain subject to sampling and non-sampling error.

Sampling errors occur when observations are made only on a sample and not on the entire population. All other errors that arise from the various survey phases are referred to as non-sampling errors. For example, non-sampling errors can occur when a price collector provides incomplete information or does not complete certain sections of the price survey; when a unit in the target population is omitted; when an out of scope product(non-matching item with base city) is captured or when errors occur in data processing, such as coding or capture errors.

A imputation process is used to impute for the missing prices portion of the sample, achieving an effective 100% coverage. Non-response bias is also minimized during the same process.

Non-sampling error:
See response rates.

Non-response bias:
See response rates.

Coverage error:
See response rates.

Other non-sampling errors:
See response rates.

Report a problem on this page

Is something not working? Is there information outdated? Can't find what you're looking for?

Please contact us and let us know how we can help you.

Privacy notice

Date modified: