This survey pertains to the vitality of Canada's official-language minorities, namely anglophones in Quebec and francophones outside of Quebec. The information collected allow for a more in-depth understanding of the current situation of individuals who belong to these groups on subjects as diverse as instruction in the language of the minority or access to different services in the language of the minority (i.e., health care), as well as language practices both at home and outside of the home.
Data release – December 11, 2007
In the spring of 2003, the Canadian government unveiled its Action Plan for Official Languages which will be reviewed in 2007. In order to obtain an overview of the current situation of francophone and anglophone minorities in areas such as family life, education, health, early childhood and language use in the public sphere, the Official Languages Branch of the Privy Council Office approached Statistics Canada about conducting a post-censal survey on the vitality of official language minorities.
The survey has two main objectives. First, it will collect information about areas that are top priorities for official language minority communities such as education, health and justice. Second, it will produce information that will assist various departments and agencies in policy development and program implementation. Moreover, the database produced will offer possible answers to issues identified by government, university and private researchers which are of concern to official-language minorities.
The information collected by the survey will not only shed light on the situation of official language minorities relative to their demographic, social, economic and cultural capital, but also offer a better understanding of their practices and their linguistic life. It will improve our understanding of the linguistic trajectory of members of official language minority communities from early childhood to adulthood, the language dynamics in exogamous families, the motivations behind parents' transmission of their mother tongue to their children and their choice of school system. In addition, the survey will collect statistics on the various subject areas related to language use in the public sphere, such as minority language access to health care, government services, retail and wholesale businesses, professional and non-professional associations, and the workplace. These statistics will provide information about whether members of official language minority communities are able to live their lives in the language of the minority.
The survey questionnaire covers the following topics: demographic, linguistic and cultural information about the respondent, demographic characteristics of household members, linguistic characteristics of the respondent's partner and selected children, demographic, linguistic and cultural information about the respondent's parents, linguistic, cultural and social information about the parents of selected children, respondent's language skills, education and linguistic trajectory from childhood to adulthood, sense of belonging and subjective vitality, experience of the child: daycare and school attendance, access to health care services in the minority language, civic participation, volunteering and social support, language use in the public sphere, linguistic practices in leisure activities, geographic mobility, economic activity, language practices at work and income.
Because these statistics will be used to assess the goals of the Action Plan for Official Languages, they will be used by different federal departments and agencies. Also, numerous university researchers and community organizations will use these statistics to obtain an overview of the current situations of minorities and to shed more light on certain situations.
The survey's target population consists of two groups: persons under age 18 in households where at least one parent belongs to the official language minority, and persons aged 18 and over who belong to the official language minority in the ten provinces and in the three territories together. Persons living in collective dwellings and on Indian reserves are excluded.
Respondents to the Survey on the Vitality of Official-language Minorities (SVOLM) are selected from the sample of the population that completed the long questionnaire in the 2006 Census, which was held in May 2006. The selection was based on answers to the questions on mother tongue, knowledge of official languages and language spoken most often at home. This ensures that the survey covers all the people belonging to the official-language minority.
Even if the survey is made up of two collection universes, that is (1) adults aged 18 years old and over and (2) children aged less then 18 years old for whom the parent (who is the respondent) belongs to the province's official-language minority, only one questionnaire has been developed. It contains three parts:
- One common part for both samples (modules ID, HLD, EPX, KOL, EDU, TRJ, SEN, INC);
- One part for the child sample (modules ENF, SPO, FAM, LPM, CLO, FRA to FRE, LPA, LEC, ACT, UTI);
- One part for the adult sample (modules PAR, HLT, COM, VOL, SOC, PUB, LEI, MOB, ECO, WRK).
Indeed, only one questionnaire is administered to a sample of adults (aged 18 and over) and to parents belonging to the province's official-language minority for a sample of children (aged under 18). The adult sample respondents have to provide information on the common part of the questionnaire and on the adult part, and the child sample respondents respond to the common part and the child part. This approach allows an approximately 40 minute interview, at the same time reducing the respondent burden. Wherever possible, to facilitate comparison, the questions that make up the various modules have been taken from other Statistics Canada surveys or other surveys on official languages.
The questionnaire content reflects the information needs of a certain number of federal government partners. The final content was determined based on results of a series of qualitative tests and from a pilot test. The pilot test was used to evaluate the questions, the consistency among them, the format of the questionnaire, as well as collection, interviewing and computer-assisted telephone interviewing procedures.
This is a sample survey with a cross-sectional design.
Respondents to the Survey on the Vitality of Official-language Minorities (SVOLM) are selected from the sample of the population that completed the long questionnaire in the 2006 Census, which was held in May 2006. The strata are generally defined by the cross-classification of the provinces and the age as defined by the Census. In New Brunswick, Quebec and Ontario, sub-provincial regions were also used to create strata. The age groups considered are 0-4, 5-11, 12-17, 18-24, 25-44, 45-64 and 65 years of age and over. The three territories will be regrouped to form two strata: one of people aged 0-17 years and the other of those 18 years and over.
In order to determine the sample size in each stratum, a targeted minimum proportion and coefficient of variation (CV) were fixed. Some parameters were estimated using the 2005 pilot test, such as the response rate and the percentage of persons who report a change in their mother tongue, official language knowledge or home language.
The sampling design used for the SVOLM is a two-phase stratified design. In the first phase, 2006 Census long form questionnaires are systematically distributed to approximately every fifth household across Canada. In two provinces (Newfoundland and Prince Edward Island), the 2006 Census short form is also used. In the second phase, respondents from the first phase who are official language minorities are divided into their strata and a systematic sample of people is chosen from each stratum. In addition, in order to do an optimal allocation of the sample, the concentration of official language minorities in each region is included in the stratification.
The total sample size is approximately 52,000 potential respondents. Of these individuals, approximately 30,000 are selected for the adult sample and 22,000 for the child sample.
Data collection for this reference period: 2006-10-10 – 2007-01-15
Responding to this survey is voluntary.
Data are collected directly from survey respondents.
Data were collected by computer-assisted telephone interviewing (CATI) in the fall of 2006. Respondents were sent an introductory letter to their home address and this was followed-up by a telephone call to collect the data, with regional office interviewers conducting the 40 minute interviews. Proxy interviews were not permitted for the adult sample. For the child sample, a respondent for the child was chosen a priori from the sampling frame. This was usually one of the child's parents or, on rare occasions, one of the child's grandparents if the child did not live with their parents but their grandparents. Since the child's belonging to the target population depends on the parents' (or grandparents') belonging to the official-language minority, it was important to contact the selected parent for the interview and not just any adult in the household. If the selected parent was absent for the duration of the survey, it was possible to conduct the interview with the other parent but only if, and only if, the other parent was also part of the official-language minority. The questionnaire allowed for these situations to be identified.
The computer system used by the interviewers to collect respondent data allowed for the prevention of a number of errors. When an impossible, improbable or incoherent answer was entered into the system by the interviewer, the system displayed a message which allowed the interviewer to correct typing errors or verify information with the respondent, without permitting them to continue until the error was fixed. Checking of some interviews was done by the interviewer supervisors and feedback was provided in order to avoid the repetition of errors.
Once collection was finished, a data processing system was implemented. This included data validation, checking the consistency between sections, coding of write-in responses, checking the consistency of the relationships among members of the household, derivation of a response status, and validating the flow of the questionnaire.
In a sample survey, each respondent represents not only himself/herself, but also other persons that were not sampled. Consequently, a weight is associated to each respondent to indicate the number of persons that this respondent represents. This weight must be used for all estimations.
After data processing, the next step consists of attributing a weight to each record in the sample. The weight calculation consists of three main steps: (1) calculation of the initial weight, (2) adjustment for non-response, and (3) post-stratification.
(1) In the first step, the inverse of the probability of selection was attributed as the initial weight to each record in the sample. Therefore, this weight reflects the sampling design that was used.
(2) The correction of the weights for total non-response was done using a method that predicts the probability of response. The probability of response of the respondents and the non-respondents was estimated using a logistic regression model. Response classes based on the response probabilities predicted by the model were then formed with the help of a classification algorithm. Once the classes were formed, the mass of the weights of the non-respondents was transferred to the respondents within each class. The correction for non-response was done in three parts for each of the two samples: adjustment for "non-contact", adjustment for refusals and adjustment for out of scope records. Since the variables explaining each type of non-response are different, it was preferable to construct different models.
(3) Post-stratification consists of correcting the weights of respondent records in such a way that totals for certain variables such as province, region, age group and language group are consistent with the corresponding Census totals.
Since estimates are obtained from a sample as opposed to a census, estimates will vary from sample to sample (sampling error). In order to provide estimates of sampling error for statistics with SVOLM data, the bootstrap method is used. This method, which is a resampling method, consists of selecting M subsamples (with replacement) from the main sample. Each subsample is then weighted by calculating the initial weights and applying to them the same adjustments we applied to the main sample weights, i.e. adjustments for non-response and post-stratification. The sampling error is measured and estimated by the bootstrap variance which is the empirical variance of the desired statistic calculated from the main sample and the M bootstrap subsamples.
This was the first time Statistics Canada conducted a survey specifically on the vitality of official-language minorities. However, during the weighting process, a mechanism was put in place to ensure that estimates obtained from important demographic variables such as the knowledge of official languages, home language, mother tongue, place of birth, immigrant status, highest level of schooling and language at work were coherent with those of the 2006 Census.
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.
This is a brief overview. For more details, please refer to the methodology section (Appendix A) of the analytical report: Minorities Speak Up: Results of the Survey on the Vitality of the Official-Language Minorities, Catalogue no. 91-548-X (available through the Publications link on the sidebar above).
The errors that occur in surveys can be separated into two categories, depending on whether they were caused by sampling or not.
Sampling errors are mainly due to the fact that only a sample, not the entire population, is being used for analysis. Therefore, they cannot be avoided completely, but it is possible to measure their magnitude. A measure is thus provided for each cell in all disseminated data tables. For a given estimate, this measure is presented as the coefficient of variation (CV) which is the ratio between the square root of the variance (standard error) of the estimate and the estimate itself. The CV indicates the proportion that the standard error represents compared to the estimate. So, the smaller the CV, the more the corresponding estimate is reliable. The CVs that accompany the estimates in the SVOLM tables (see "Minorities Speak Up: Results of the Survey on the Vitality of the Official-Language Minorities", Catalogue no. 91-548-X, Appendix E - Supporting Tables for the detailed values of the CVs) were calculated using the "bootstrap" method.
Non-sampling errors cannot be easily estimated, but they can be avoided. These errors can occur during any stage of a survey. They include coverage, non-response, measurement and processing errors.
Concerning potential coverage errors, it was possible to avoid most such cases by using the Census as the sampling frame, which provides a very good coverage of the Canadian population. However, because the Census is a self-completed survey and it allows proxy responses, it is possible that errors occurred in the responses to the language questions. Hence, it was possible for an adult in a given household to answer the three language questions on behalf of the other adults in the households without having sufficient knowledge to do so. Such a situation could cause the inclusion of an individual in the target population when they shouldn't be there (over coverage) or exclude individuals who should be in the target population (under coverage). Keeping in mind that the three language "filter" questions where asked of respondents during the survey, correction was done for part of the over coverage. Once it was determined that an individual was not part of the target population, the individual was excluded from the sample and as a consequence, the weighting was adjusted. On the other hand, it was impossible to correct for under coverage and this error is difficult to quantify.
The response rate for the survey is approximately 73% (for the adult and child samples combined). The units which are excluded from the target population, the out of scopes, are not included in the calculation of this rate since the sample size was already inflated to take these losses into account. If, however, one is interested in the proportion of individuals who completed the entire questionnaire from everyone selected for the survey, a rate of 67% is observed; in this case the out of scopes are considered to be non-respondents. From another point of view, the out of scopes could also be considered as respondents since they were contacted and interviewed. However, this option was not retained here.