Homeowner Repair and Renovation Survey (HRRS)

Detailed information for 2001

This survey gathered information about spending by homeowners in Canada's ten provinces on housing repairs and renovations.

Data release - December 2, 2002

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

Description

This survey gathers information about spending by homeowners in Canada's ten provinces on housing repairs and renovations.

In the early years, funding was provided by Canada Mortgage and Housing Corporation (CMHC) or jointly by CHMC and Statistics Canada.

Starting in 1997, funding was provided by Statistics Canada's Project to Improve Provincial Economic Statistics (PIPES).

Many different organizations and individuals use the information:

- Statistics Canada uses the results to estimate residential repair and renovation expenditures in the System of National Accounts. This estimate is an input into Gross Domestic Product.

- Municipal, provincial and federal governments use the data to make policy and program decisions that affect homeowner households. The Canada Mortgage and Housing Corporation is one government agency that uses these data.

- Individuals and families use the information from the Homeowner Repair and Renovation Survey to compare their spending habits with those of other households. Are they spending more or less than others with similar characteristics to themselves? Are Canadian households spending more on repairing or on renovating their homes?

- Businesses offering residential repair and renovation goods and services can better understand the needs of homeowners based on data from this survey. Businesses must understand changes in how homeowners spend in order to provide convenient and suitable goods and services.

- Trade associations can represent their members better by knowing how much Canadian homeowners are spending in the repair and renovation market.

Reference period: Calendar year

Collection period: March following the reference year

Subjects

• Income, pensions, spending and wealth

Data sources and methodology

Target population

This survey was conducted as a supplement to the Labour Force Survey (LFS, record number 3701).

The LFS covers 98 percent of the Canadian population. The survey excludes from its coverage the Northwest Territories and Nunavut. Yukon is included in the LFS but excluded from the HRRS. Residents of Indian reserves and Crown lands, inmates of institutions and full-time members of the Canadian Armed Forces are also excluded.

Only homeowner households living in private dwellings are represented in the HRRS. This excludes military, logging and construction camps and collective households such as hotels, large lodging houses and clubs. Also excluded are private households living on boats or in motor homes, in tents, or other dwellings (other than mobile homes) that can be moved on short notice.

Instrument design

The current HRRS questionnaire was introduced in 1997. At that time, minor changes were made to the questionnaire in order to make more use of the power of Computer Assisted Interviewing (CAI).

Since the original questionnaire had been designed as a paper questionnaire, the questionnaire redesign represented an opportunity to make use of the power of CAI. This included the incorporation of question wording that depended upon answers to earlier questions, more complex question flows and some on-line edit checking for logical inconsistencies.

Sampling

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

Since the HRRS is a supplement to the LFS, its sample design is closely tied to it. The current design of the LFS can be described in broad terms as a stratified, multi-stage design employing probability sampling at all stages of the design.

Sample size = approximately 24,000 dwellings.

Data sources

Responding to this survey is voluntary.

Data are collected directly from survey respondents.

At the conclusion of the LFS interviews, interviewers administered the HRRS questionnaire to pre-selected homeowner households. The HRRS questions are designed for Computer Assisted Interviewing (CAI) either over the telephone or in person. CAI not only eliminates the need for paper questionnaires, it also offers the possibility of integrating many valuable controls into the survey. Such controls include programming the logical flow of the questions, specifying the types of answers required, performing on-line edits and giving immediate feedback to the respondent and/or interviewer with on-screen prompts. The CAI application also captures notes and comments provided by the respondent and the interviewer.

Interviews were conducted by telephone from centralized locations. Between the 1999 and the 2001 surveys, Statistics Canada adopted collection procedures located in Regional Offices to replace telephone interviews conducted from the interviewer's home. This change allows for extra quality control over data collection.

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

Error detection

Using CAI, HRRS data were automatically captured during the interview. Data were then transmitted electronically to head office after each working day during the ten-day collection period. Notes written by interviewers and extreme expenditure values were all examined. Further edits were performed when needed.

Imputation

Despite all attempts made by the interviewers, some non-response is inevitable. Non-response can be divided into two types: total and partial non-response.

Total non-response occurs when there is no information available about household spending on repairs and renovations. To correct for total non-response, adjustment weights are used to adjust the basic survey weight (see next section).

Partial non-response occurs when only some information about a household is missing and is compensated for by imputation. For the HRRS, imputation for missing expenditure fields was done using a donor imputation method called hot-deck. Essentially, this method consists of forming imputation classes based on the expenditure item and economic region and dividing these classes into two groups: donors and receivers. A donor is randomly assigned to each receiver within an imputation class while making sure that each donor is used only once.

Estimation

The estimation of population characteristics from a survey is based on the premise that each sampled unit represents, in addition to itself, a certain number of unsampled units in the population. To reflect this representation and in order to derive estimates of the population of homeowners from the HRRS data, a survey weight is attached to each responding household in the sample. This survey weight comes from the LFS and is essentially equal to the inverse of the probability of selection of the household.

Some adjustments must be performed to this survey weight before deriving a final weight for each HRRS responding household. This final weight will later be used for analytical purposes. First, there is an adjustment to account for the fact that the HRRS sample is a subsample of the LFS and is formed of only four of the six LFS rotation groups. Then, an adjustment is applied to the weight of the HRRS responding households to compensate for the loss of those selected households that did not participate in the survey.

The final adjustment to the weight is made to correct for coverage errors. The sub-weights are compared to independently derived estimates of the population and adjusted so that the survey estimates of the population conform to these control totals. These final weights are used in the HRRS tabulations.

Quality evaluation

Data are compared to the results from previous years and to other data sources: the Census of Population (record number 3901), administrative and other Statistics Canada surveys.

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.

For tabular data, the minimum number of records allowed per cell (usually 30) ensures that confidentiality requirements are not violated.

Revisions and seasonal adjustment

This methodology does not apply to this survey.

Data accuracy

Response rate for 2001 = 68%

Coefficients of variation for summary level spending categories for Canada and the provinces are supplied in the publication 62-201-XIB.