Official name of the survey:
Survey of Consumer Finances.

Administrative unit responsible for the survey:
Household Surveys Division

Address:

The following people with specific expertise for various parts of the survey can be contacted at the above address:

Sampling and sample selection:

Frank Mayda
Labour Force Survey, Methods Section
Jean Talon Building 4-C2
Tel: (613)951-9807

Data collection:

Brian Williams
Survey Operations Division
Jean Talon Building 6-C6
Tel: (613)951-9467

Demographic and income variables, and editing and survey processing:

Kevin Bishop
Jean Talon Building 5-D2
Tel: (613)951-2211
Fax: (613)951-0562

Funding for the Canadian Survey of Consumer Finances is provided for by Statistics Canada. Copies of the original codebooks and other documentation can be obtained from Statistics Canada at the above address.

The principle users of the data collected in this survey are government, universities, social agencies, advocacy groups, and private sector corporations and consultants.

The Survey of Consumer Finances was first conducted in 1951, and has been conducted annually since 1972. Since 1985 the survey has been carried out as a Labour Force Survey supplement. "Census Families" and "Households" microdata were first made available in 1971; "Economic Families" microdata in 1973; and "Individual" level microdata in 1981.

The main purpose of the Survey of Consumer Finances is to measure the composition, earnings levels, and distribution of income in Canada.

The process of data collection for the 1987 Survey of Consumer Finances was begun and completed in April 1988 for the preceding year.

There are restrictions on the use of the microdata by the public. In order to satisfy bureau policy concerning the disclosure of identifying information, the individual microdata file records have been screened and masked.

A bibliography of the most important publications based on the Canadian Survey of Consumer Finances is provided in Section M. Also available in Section M is a list of available user documentation for the Survey of Consumer Finances.

B. POPULATION AND SAMPLE SIZE, SAMPLING METHODS back

A stratified cluster probability sample design is employed in the Survey of Consumer Finances. The final stage sampling frame for the survey is comprised of a list of all private dwellings in the ten Canadian provinces. This sampling frame includes the total population of private household heads, with the exception of households on Native American reserves, and households in the Yukon and Northwest Territories.

The population living in institutions such as nursing homes for the aged and ill, prisons, etc. were not covered by the sampling frame; however, military personnel living in military housing or with their families were covered. Foreign residents (defined as neither Canadian born nor landed immigrants), persons with a usual residence elsewhere, and those individuals living in seasonal dwellings, were excluded from the survey at the time of the interview. All population groups had an equal probability of selection.

As defined by the sample frame, a total of 9,244,000 households existed in the population from which the sample was selected. This represents 20,267,000 individuals of 15 years of age or more. A total of 41,670 households (87,837 individuals) were surveyed and for which the minimum amount of data was obtained. When weighted, this amounts to 9,244,000 households and 20,267,000 individuals. This was the same sample that was made available to LIS. LIS has been able to make 11,518 cases available to its users. The geographic areas of Yukon and the Northwest Territories were excluded from the survey.

Estimates of sampling variability (standard errors) have been computed for this survey.

C. MEASURES OF DATA QUALITY back

1. Item and Group Nonresponse Rates

A total of 1,399 households (3.2%) were not interviewed, or for which data were not obtained. The only information which is available regarding these households is their geographic location. 20.4% of individuals did not respond to the income questions.

For those individuals who were not interviewed, but who were members of an interviewed household, demographic and labor force information is available.

No information is available regarding the non-response rates of population sub-groups, nor for item non-response rates.

Although Yukon and the Northwest Territories were excluded from the survey, it is unlikely this would bias the sample, as the population in these areas is too small to have a significant impact on the representativeness of the sample.

2. Reporting and Under-Reporting and Income Data Quality

The quality of the income data has been evaluated by comparing the Survey of Consumer Finances income data with Canadian national accounts. These estimates are reported in Table 1. A similar evaluation of the taxation data has also been performed.

TABLE 1.
Comparison of scf estimates to adjusted personal income 1987 economic family (Millions of dollars)

(1)Taken from "SCF/National Accounts Reconciliation." No date. Statistics Canada, Households Surveys Division.

(2) SCF amounts include receipts for family units whose major source of income is military pay. These family units are not included in published tables.

(3) SCF = Social Assistance and Provincial Income Supplements plus Other Income from Government Sources plus Provincial Tax Credits.

NA = Item 6 - (Item 7+8+9+10).

* The Child Tax Credit figure and Sales Tax Credit were obtained from Revenue Canada, not National Accounts.Statistics Canada.

3. Data Corrections and Adjustments

Adjustments have been made to the income data in the Survey of Consumer Finances. In addition, "hot deck" imputation was performed by categorizing responding and non-responding individuals by demographic and labor force characteristics, and using "donor" values to complete "recipient" records.

No other adjustments for nonresponse were performed; nor were any values replaced by consulting alternative data sources.

4. General Assessment of Income Data Quality

The four main problems regarding the quality of the income data in this survey which users should be aware of are:

1. There was a low overall response rate to the income questionnaires. Approximately 20% refused to answer any of the income questions.

2. Some specific income items were undercovered. These included investment income sources and government income sources that are not universal programs.

3. The top end of the income distribution curve was underrepresented in the sample.

4. There was a proxy response for some individuals concerning information on supplementary work experience data (e.g. weeks worked, weeks unemployed in the reference year).

D. DATA COLLECTION AND ACQUISITION back

1. Method of Data Collection

Households in the monthly Labour Force Survey for April 1988 were sent income questionnaires by mail two weeks prior to the April Labour force Survey. Respondents then completed the forms and gave their responses to interviewers by telephone.

Participation in the survey was required by law. The privacy of respondents is assured by the Statistics Act Chapter 15, Statutes of Canada 1970-71-72. This act prohibits the disclosure by Statistics Canada of any particulars which could reveal the identity of any individual.

2. Survey Instruments

One income questionnaire was provided to each person 15 years of age and over.

3. Survey Respondent Rules

Each individual 15 years and over should complete his or her questionnaire. However, another individual (the household contact for the Labor Force Survey) may relate the completed information to the interviewer during the telephone interview.

Respondents were asked, when applicable, to consult their income tax forms. It is not known what proportion of the respondents did this.

E. WEIGHTING PROCEDURES back

Weights have been assigned to each case in the Survey of Consumer Finances. These weights have been assigned in order to insure that the sum of the survey weights are representative of independent population estimates by Province, age and sex categories.

The assigned weights do not help adjust for missing income data or nonsampling errors related to income. (Missing income data were imputed through a hot-deck procedure.)

F. DETERMINATION OF SURVEY UNIT MEMBERSHIP back

A household member is a person who, during the survey week:

a. regards the dwelling to be their usual place of residence; or
b. is staying in the dwelling and has no usual place of residence elsewhere.

A household is defined as any person or group of persons living in a dwelling. It may consist of one person living alone, a group of people who are not related but who share the same dwelling, or it may be a family.

A dwelling is defined as any set of living quarters which:

a. is structurally separate from the living quarters of other dwellings; and
b. has a private entrance outside the building or a private entrance from a common hall or stairway inside the building. The entrance must be one that can be used without passing through the living quarters of another dwelling.

In the case of families, the following individuals and their relationship to the head of the family can be identified in the Survey of Consumer Finances:

1. Head of family
2. Spouse
3. Son or daughter (natural, adopted or step)
4. Grandchild
5. Son-in-law or daughter-in-law
6. Foster child (less than 18 years of age)
7. Parent
8. Parent-in-law
9. Brother or sister
10. Other relatives.

The head of household is always the head of the primary economic family.

The basic units of aggregation which are used in this survey are: Household, Economic Family, Census Family, and Individuals.

Within households, Economic Family units and individual relationships within that family, and Census Family Units and individual relationships within that family, can be identified.

G. CHILDREN AND SPOUSES back

In this survey natural children, adopted children, guardianship children and foster children can be identified, but it is not possible to distinguish between the children of the head versus children of the spouse.

Spouses are defined as legally married, and common-law couples.

H. AVAILABILITY OF BASIC SOCIAL AND DEMOGRAPHIC INFORMATION back

In Table CN87.1 are summarized the basic social and demographic information which is available in the Survey of Consumer Finances.

Table CN87.1
Basic Social and Demographic Information

I. AVAILABILITY OF LABOUR MARKET INFORMATION back

Labor market information is available for the Survey of Consumer Finances. This information is summarized in Table CN87.2.

Table CN87.2
Labor Market Information

J. AVAILABILITY OF GEOGRAPHIC INFORMATION back

Information on the Canadian Province of the respondent is available. In addition, information on the size of the area of residence is available. This has been coded into the following categories:

1. 500,000+ persons
2. 100,000-499,999 persons
3. 30,000-99,999 persons
4. urban areas less than 30,000 persons
5. rural areas.

K. SOURCES AND AMOUNTS OF CASH INCOME back

Sources and amounts of cash income are recorded for the previous calendar year (1987). Amounts recorded are actual amounts received.

L. TAXES back

Income tax information is available for the Survey of Consumer Finances. See variable V11 for more information on taxation.

M. BIBLIOGRAPHY OF MAIN PUBLICATIONS BASED ON THE CSP PANEL back

Statistics Canada. 1987. "Earnings of Men and Women." Catalogue 13-217.

Statistics Canada. 1987. "Family Incomes." Catalogue 13-208.

Statistics Canada. 1987. "Income After Tax, Distributions by Size in Canada." Catalogue 13-210.

Statistics Canada. 1987. "Income Distributions by Size in Canada." Catalogue 13-207. Statistics Canada. 1988. Survey of Consumer Finances. Interviewer's Manual.

Statistics Canada. Households Surveys Division. No date. SCF/National Accounts Reconciliation.