General Social Survey - Social Engagement (GSS)

Status:
Active
Frequency:
Quinquennial (5 year)
Record number:
5024

The two primary objectives of the General Social Survey (GSS) are: to gather data on social trends in order to monitor changes in the living conditions and well being of Canadians over time; and to provide information on specific social policy issues of current or emerging interest.

This survey collects data on dimensions of social engagement, including social participation, civic participation, trust and reciprocity.

Detailed information for 2003 (Cycle 17)

Data release - July 6, 2004

Description

The two primary objectives of the General Social Survey (GSS) are: to gather data on social trends in order to monitor changes in the living conditions and well being of Canadians over time; and to provide information on specific social policy issues of current or emerging interest.

This survey collects data on dimensions of social engagement, including social participation, civic participation, trust and reciprocity.

The data collected will be used by government and researchers to better understand the role that social engagement and social networks play in the well-being of individuals and society.

Statistical activity

This record is part of the General Social Survey (GSS) program. The GSS, originating in 1985, conducts telephone surveys. Each survey contains a core topic, focus or exploratory questions and a standard set of socio-demographic questions used for classification. More recent cycles have also included some qualitative questions, which explore opinions and perceptions.

Until 1998, the target sample of respondents was approximately 10,000 persons. This was increased in 1999 to 25,000. With a sample of respondents of 25,000, results are available at both the national and provincial levels and possibly for some special population groups such as disabled persons and seniors.

Subjects

  • Social networks and civic participation
  • Society and community

Data sources and methodology

Target population

The target population is non-institutionalized persons 15 years of age or older, living in the ten provinces.

The samples for most GSS cycles are selected using random digit dialing telephone methods and the interviews are conducted by telephone. Thus persons in households without telephones cannot be interviewed. However, persons living in such households represent less than 2% of the target population. Interviews are not conducted by cellular telephone so persons with only cellular telephone service are also excluded; again, this group makes up a very small proportion of the population, less than 3%.

Instrument design

The questionnaire was designed based on qualitative testing (focus groups), a pilot test and interviewer debriefing.

Sampling

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

Data for Cycle 17 of the GSS were collected in 7 independent samples (waves) from February to December 2003. The target sample sizes for each wave were initially the same but were adjusted slightly during the year to try to achieve a final overall sample size of 25,000 respondents. These samples were all selected using the random digit dialling (RDD) technique known at Statistics Canada as the Elimination of Non-Working Banks (ENWB). This sampling technique is a method in which an attempt is made to identify all working banks for an area (i.e., to identify all sets of 100 telephone numbers with the same first eight digits containing at least one number that belongs to a household). Thus, all telephone numbers within non working banks are eliminated from the sampling frame.

In order to carry out sampling, each of the ten provinces was divided into strata and separate samples were selected from each stratum. These strata were defined geographically. Many of the Census Metropolitan Areas (CMAs) were each considered separate strata. This was the case for St. John's, Halifax, Saint John, Montreal, Quebec City, Toronto, Ottawa, Hamilton, Winnipeg, Regina, Saskatoon, Calgary, Edmonton, Vancouver and Victoria. CMAs not on this list are located in Quebec and Ontario. Two more strata were formed by grouping the remaining CMAs in each of these two provinces. Finally, the non-CMA areas of each of the ten provinces were also grouped to form ten more strata. This resulted in 27 strata in all. This is the same stratification used for Cycles 13 and 15 but is different from that used for Cycle 14 and many previous cycles of the GSS, when there were 21 strata in all.

In each stratum, a simple random sample without replacement of telephone numbers was selected by choosing a simple random sample with replacement of banks from the frame, and then randomly generating the last two digits for each bank to obtain the telephone number. Each sample of telephone numbers was produced before the first day of interviewing for the wave.

Data sources

Data collection for this reference period: 2003-02-10 to 2003-12-14

Responding to this survey is voluntary.

Data are collected directly from survey respondents.

Computer assisted telephone interviewing (CATI) was used to collect data for the GSS. Households were selected through Random Digit Dialling methods. When a private household was reached, interviewers enumerated all household members, collecting basic demographic information including age, sex and martial status. An algorithm was then used to randomly select one household member (age 15 and older) to participate in the survey. Respondents were interviewed in the official language of their choice. Interviews by proxy were not allowed. The overall response rate during collection for Cycle 17 was 78%.

All interviewing took place using centralized telephone facilities in four Statistics Canada regional offices, with calls being made from approximately 9:00 a.m. until 9:00 p.m., Monday to Saturday inclusive. The four regional offices were: Halifax, Montreal, Winnipeg and Vancouver. Statistics Canada staff trained interviewers in survey concepts and procedures as well as telephone interviewing techniques using CATI. The majority of interviewers had previous experience interviewing for the GSS.

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

Error detection

Error detection was done through edits programmed into the CATI system.

The CATI data capture program allowed a valid range of codes for each question and built-in edits, and automatically followed the flow of the questionnaire.

All survey records were subjected to computer edits throughout the course of the interview. The CATI system principally edited flow of the questionnaire and identified out of range values. As a result, such problems could be immediately resolved with the respondent. If the interviewer was unable to correctly resolve the detected errors, it was possible for the interviewer to bypass the edit and forward the data to head office for resolution. All interviewer comments were reviewed and taken into account in head office editing.

Head office edits performed the same checks as the CATI system as well as more detailed edits.

Imputation

The flow editing carried out by head office followed a 'top down' strategy, in that whether or not a given question was considered "on path" was based on the response codes to the previous questions. If the response codes to the previous questions indicated that the current question was "on path", the responses, if any, to the current question were retained, though "don't know" was recoded as 8 (98 or 998, etc.) and refusals were recoded as "Not Stated", i.e. 9 (99 or 999, etc.). If, however, a response was missing to the current question, it was coded as "Not Stated", i.e. 9 (99 or 999, etc.). If the response codes to the previous questions indicated that the current question was "off path" because the respondent was clearly identified as belonging to a sub-population for which the current question was inappropriate or not of interest, the current question was coded as "Not Applicable", i.e. 7 (97 or 997, etc.).

Due to the nature of the survey, imputation was not appropriate for most items so missing data were coded as 'Not Stated'.

However, non-response was not permitted for those items required for weighting. Values were imputed in the rare cases where either of the following was missing: sex or number of residential telephone.

Estimation

When a probability sample is used, as was the case for the GSS, the principle behind estimation is that each person selected in the sample represents (in addition to himself/herself) several other persons not in the sample. For example, in a simple random sample of 2% of the population, each person in the sample represents 50 persons in the population (himself/herself and 49 others). The number of persons represented by a given respondent is usually known as the weight or weighting factor.

The initial weight for the GSS is the inverse of the probability of the telephone number being selected. This weight is inflated to account for non-response, then adjusted for the number of telephone numbers in the household so that a non-response adjusted weight reflecting the household's probability of being selected for the survey is produced. This weight is then adjusted for the selection of only one respondent among the eligible household members to yield a non-response adjusted person weight. Finally, the weights have been adjusted using a raking ratio calibration (post-stratification) technique to match Census based population estimates for strata and for provincial age-sex groups.

WGHT_PER: This is the weighting factor for analysis at the person level, i.e. to calculate estimates of the number of persons (non-institutionalized and aged 15 and older) having one or several given characteristics. WGHT_PER should be used for all estimates.

GSS Cycle 17 was a survey of individuals and the analysis file contains questionnaire responses and associated information from 24,951 respondents.

GSS Cycle 17 was not designed to be a survey of households, so questions such as "In what type of dwelling are you now living?" should be used to estimate the number of persons who live in a particular type of dwelling (rather than the number of dwellings of a given type).

In addition to the estimation weights, two hundred bootstrap weights have been created for the purpose of design-based variance estimation.

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 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.

Data accuracy

The methodology of this survey has been designed to control errors and to reduce the potential effects of these. However, the results of the survey remain subject to error due to both sampling error and non-sampling error.

Sampling error:
As the data are based on a sample of persons they are subject to sampling error. That is, estimates based on a sample will vary from sample to sample, and typically they will be different from the results that would have been obtained from a complete census. The potential range of this difference has been estimated for key data and used to produce tables that can be used to estimate the sampling variability of many estimates. More precise estimates of the sampling variability of estimates can be produced with the bootstrap method using bootstrap weights that have been created for this survey. The bootstrap method was used to estimate the sampling variability for all of the estimates included in '2003 General Social Survey on Social Engagement, cycle 17: an overview of findings' and for the important comparisons made in the text. Estimates with high sampling variability are indicated in this article and all of the highlighted differences between subgroups of the population are significant at the 95% level.

Non-sampling error:
Even a census of the population of interest produces estimates subject to error. While these are called non-sampling errors, estimates from samples still contain errors of this type. Common sources of these errors are imperfect coverage, non-response, response errors, and processing errors.

Coverage of the GSS-17 targeted population by the RDD frame is estimated to be more than 98% complete; rates of telephone service are very high in Canada. These rates are high for virtually all socio-demographic groups, but are lowest among those households with the lowest incomes. As a result persons living in such households are slightly under-represented in the GSS-17 sample. In addition, while every effort was made to avoid non-response, the non-response rate for GSS-17 was 22%. Little or nothing is known about the non-responding cases, and so the results may be biased to the extent that the non-responding cases differ from those that provided responses

Data file