Survey of Household Spending (SHS)
Summary of changes
Activity on this program started: 1997
To continue to deliver relevant, high-quality data, Statistics Canada is modernizing its household expenditure survey. As part of this modernization effort, we are:
- streamlining and updating the survey content;
- implementing collection via electronic questionnaire;
- exploring ways to use alternative collection tools and data sources; and
- producing more relevant and accessible data products.
To reduce respondent burden and invest in this modernization effort, the SHS is moving to a two-year production cycle from SHS 2017 onwards. Data for the reference year 2019 will be released in January 2021.
The 2015 SHS is limited to the three territorial capitals (Whitehorse, Yellowknife and Iqaluit) in the North and uses the new design of the survey, which was implemented for the provinces starting in 2010. As such, data for the territories from SHS 2015 should not be compared to those from SHS 2012 and earlier.
Starting in 2015, the survey is conducted every two years in the three territorial capitals.
In 2012, the sample size for the expenditure diary was reduced to 50% of the total sample, given that the baseline for diary expenditures was established during the first years of its collection. Corresponding changes have been made in the weighting method for data collected from the diary.
Data have also been collected in the territories in 2012 according to the previous model of the SHS (questionnaire with 12-month recall period only, no diary of expenditures). As such, estimates from the territories are not combined with the provincial estimates but presented separately (see section 6 of the "User Guide for the Survey of Household Spending, 2012").
The method of adjusting for incomplete diaries has been refined with the 2011 SHS. As well, the age of household members is now defined to be at the time of the interview rather than as of December 31st of the survey year.
To ensure comparability of the data, the 2010 data have also been revised by incorporating these changes. The revised 2010 estimate of average household spending on all types of goods and services has increased by 1.3% when compared with the previously published 2010 estimate (April 2012).
A new methodology which combines a questionnaire and a diary to collect the household expenditures was introduced for the 2010 survey. The reference periods have been reduced for many expenditure items and collection is continuous throughout the year. Although the expenditure data collected since 2010 are similar to those of previous years, the changes to data collection, processing and estimation methods have created a break in the data series. As a result, caution should be used in comparing SHS data since 2010 with previous years, unless otherwise noted.
The content of the SHS has also been reviewed in 2010 to reduce the time required for the interview. A number of components regarding household equipment and dwelling characteristics and most of the questions regarding changes in household assets and liabilities have been dropped. Some definitions have also been changed. As well, starting with the 2010 survey, the data related to household income and income tax come mainly from an administrative data source.
In reference year 2009, a computer assisted telephone interview (CATI) questionnaire entitled the "Renovation and Home Repair Survey" collected information on spending for home repairs and renovations using a subset of questions that are asked during the CAPI interview. The results of this collection were combined with the same data from the CAPI interview to provide better estimates to the Canadian System of National Accounts.
Due to introduction of the new electronic CAPI questionnaire, changes were made to the data processing and quality control steps. In addition to the automatic edits built into the electronic questionnaire, balance edit comparing total revenues, expenses and changes in assets and liabilities performed by the interviewer was re-instated in 2007.
Target population - Beginning with 2006, the distinction between full year and part year members and households has been removed. Spending data are collected for the reference year for all members of the household at the time of interview.
Data were collected by personal interview using electronic questionnaires on laptop computers (computer assisted personal interviews, or CAPI) instead of the paper questionnaires used before the 2006 reference year.
Error detection - Due to introduction of the new electronic CAPI questionnaire, changes were made to the data processing and quality control steps. Automatic edits built into the electronic questionnaire replaced the balance edit and regional office edits performed in previous years.
Since the data for the 1997-2005 were based on full year members only, to maintain the comparability with prior years, the data for 2002-2005 have been revised to include both full year and part year households. Data for 1997-2001 will be revised at a later date.
As of the 2003 reference year, the Homeowner Repair and Renovation Survey (HRRS, record number 3886) has been integrated into the Survey of Household Spending.
Estimation - From 1997 to 2003, SHS used benchmarks derived from the 1996 Census. As of the 2004 reference year, the SHS uses survey weights which take into account new population projections from the 2001 Census.
Revisions and seasonal adjustment - All estimates back to 1997 are being adjusted to use new annual population estimates based on the 2001 Census. The method of calculating survey weights has been modified and standardised for all years from 1997 onward.
Data sources - In 1997, 1998, 1999 and every second year thereafter (starting with 2001), data are collected for the territories; for the other years, data are collected for the 10 provinces only.
Conducted since 1997, the Survey of Household Spending integrates most of the content found in the Family Expenditure Survey (FAMEX, record number 3504) and the Household Facilities and Equipment Survey (HFE, record number 3505). The first reference year for which metadata are available through these web pages is 1999.