Languages Reference Guide, Census of Population, 2016Languages Reference Guide, Census of Population, 2016

Definitions and concepts

The census collects a wealth of information on the languages of people living in Canada. Census data on language are used to measure the size, evolution and composition of language groups. These data are used most notably in implementing and administering a number of federal and provincial statutes, including the:

• Canadian Charter of Rights and Freedoms (1982) – Canada
• Official Languages Act (1988) – Canada
• Official Languages Act – New Brunswick
• Charter of the French Language – Quebec.

Census data on languages serves equally to study linguistic practices, knowledge as well as use of official and non-official language in Canada. Language data are also used by researchers, community groups, health-care providers, businesses and other organizations throughout the country to ensure equal opportunity for everyone. Data on language knowledge and use are important to consider when dealing with issues related to human resources policies, local education and training, social and economic integration of newcomers, health promotion and community programs and services.

The following variables, as defined in the Dictionary, Census of Population, 2016, Catalogue no. 98-301-X, have been created with language data collected during the census on May 10, 2016:

• Knowledge of official languages
• Home language
• Mother tongue
• First official language spoken
• Knowledge of non-official languages
• Language of work

In Canada, 'official languages' refer to English and French. 'Non-official languages' refer to all other languages.

Classifications

Data from language questions in the census are used to derive summary and detailed variables which provide a linguistic portrait of the population living in Canada. Information is provided on English- and French-speaking communities as well as other language groups, including those who speak Aboriginal languages.

The classification of languages collected in 2016 is available in the List of figures (Figures 3.3, 3.3A to 3.3D) in the Dictionary, Census of Population, 2016, Catalogue no. 98-301-X. A comparison of languages collected in the 2016, 2011 and 2006 censuses is also available in the Dictionary, Census of Population, 2016, Catalogue no. 98-301-X, List of appendices. Appendix 3.0 presents the classification of mother tongue and home language, Appendix 5.6 knowledge of non-official languages and Appendix 6.3 language of work.

The language classification used to disseminate data for the 2016 Census of Population has been developed over time and is based on the ISO 639-3 standard inventory of language identifiers administered by SIL International, as included in the Ethnologue: Languages of the World.

Questions

For the second consecutive census, four language questions were asked to the entire population. The knowledge of official languages question (Question 7) was followed by a two-part question on home language. The first part (Question 8. a)), asked respondents to indicate the language spoken most often at home, and the second part (Question 8. b)) asked respondents to report any other languages spoken on a regular basis at home. The mother tongue question (Question 9) was the last language-related question in the 2016 short-form census (2A) questionnaire.

The wording of Question 7 on English and French knowledge was the following:

7. Can this person speak English or French well enough to conduct a conversation?

The home language question (Question 8) was divided into two parts:

8. a) What language does this person speak most often at home?
   b) Does this person speak any other languages on a regular basis at home?

The wording of Question 9 on mother tongue was the following:

9. What is the language that this person first learned at home in childhood and still understands?

A quarter of Canadian households received the 2016 long-form census (2A-L) questionnaire. The Census long-form questionnaire consisted of the same four language questions that are on the 2016 short-form census (2A) questionnaire. Additional data on languages were collected in the sociocultural information section with Question 16 and in the labour market activities section with Questions 45. a) and 45. b) on the long-form census questionnaire.

Respondents living on Indian reserves, in Indian settlements and in other remote areas received the 2016 long-form census (2A-R) questionnaire. The wording of the 2A-R language questions was the same as the 2A-L language questions. However, for Question 16 on the knowledge of non-official languages, several examples of Aboriginal languages were provided to assist respondents with the write-in portion of the question.

The wording of Question 16 on the knowledge of non-official languages was the following:

16. What language(s), other than English or French, can this person speak well enough to conduct a conversation?

The language of work question was divided into two parts:

45. a) In this job, what language did this person use most often?
    b) Did this person use any other languages on a regular basis in this job?

On the English version of all census questionnaires, the mark-in circles for 'English' appear first, while on the French version, the mark-in circles for 'français' appear first. Similarly, the wording of the knowledge of official languages question (Question 7), and of the knowledge of non-official languages question (Question 16) also changes depending on the language of the questionnaire with 'English' appearing first on the English questionnaires, and 'français' appearing first on the French questionnaires.

To assist people whose first language was neither English nor French, the census questions were translated into 22 other languages, including 11 Aboriginal languages.

The census variable 'First official language spoken' is not a census question. As directed within the framework of the application of the Official Languages Act, this variable is derived from the questions on knowledge of official languages (Question 7), language spoken most often at home (Question 8. a)) and mother tongue (Question 9). For more information on the method used to derive it, see First official language spoken of person.

More information on the wording and format of the 2016 Census language questions and the instructions which were provided to respondents for those questions, can be found in the 2016 Census long-form and short-form questionnaires and the Dictionary, Census of Population, 2016, Catalogue no. 98-301-X.

Data quality

The 2016 Census of Population underwent a thorough data quality assessment, similar to what was done for past censuses. A number of data quality indicators (briefly described below) were produced and used to evaluate the quality of the data.

The data quality assessment was done in addition to the regular quality checks completed at key stages of the census. For example, during data collection and processing, the consistency of the responses provided was checked and the non-response rates for each question were analysed. As well, the quality of imputed responses was assessed as part of the data editing and imputation steps. Finally, resulting census counts were compared with other data sources, and certified for final release.

The 2016 Census language data for mother tongue, home language and knowledge of official languages was originally disseminated on August 2, 2017. After the dissemination of this data, Statistics Canada corrected anomalies in the original results, primarily for various regions in Quebec. The language data were then re-released. A description of how the error was identified and what the error was is available in an Update of the 2016 Census language data published on the Statistics Canada website.

As a result of the corrections to these variables, the knowledge of non-official language data and the language of work data were reprocessed to ensure that any changes made to the mother tongue and home language variables were applied for the processing of the knowledge of non-official language and language of work data. For more information refer to Appendix 1.8 of the Guide to the Census of Population, 2016, Catalogue no. 98-304-X on the evaluation of the impact of updating 2016 Census language data.

For information about data quality for the census subdivision of Wood Buffalo, the data collection methodology and the use of administrative data sources, please refer to Appendix 1.4 of the Guide to the Census of Population, 2016, Catalogue no. 98-304-X.

The main highlights of this assessment for the Language data are presented below.

Data quality indicators

A number of quality indicators were produced and analysed during the data quality assessment of the Census of Population. Two of these are presented to users: the global non-response rate (GNR) and the imputation rate by question.

The GNR combines non-response at the household level (or total non-response) and non-response at the question level (partial non-response). It is calculated for each geographic area. The GNR is the key criterion that determines whether or not the census counts are released for a given geographic area – data are suppressed for geographic areas with a GNR equal to or greater than 50%. More information on the GNR is available in the Guide to the Census of Population, 2016, Catalogue no. 98-304-X.

The imputation rate by question, including imputation for total non-response, is a quality measurement specific to each question in the census. It measures the percentage of persons who did not answer the question, the percentage of persons whose response was invalid, and the percentage of persons who did not return a questionnaire, for which a valid value was assigned. Imputation eliminates gaps in the data and, when done appropriately, reduces bias introduced by non-response. This is done by identifying persons or households that have characteristics similar to the incomplete record and by copying their values to fill in the missing or erroneous responses. The imputation rates by question are presented in Table 2, below.

When the census data analysis involves crossing data from the short-form census questionnaire and the long-form census questionnaire (e.g., analysing an education variable with an age variable or analysing an education variable with an income variable), users must take into consideration certain aspects of the quality, such as the non-response bias and the variability due to sampling and total non-response.

Variability due to sampling and total non-response for the long-form estimates

The objective of the long-form census questionnaire is to produce estimates on various topics for a wide variety of geographies, ranging from very large areas (such as provinces and census metropolitan areas) to very small areas (such as neighbourhoods and municipalities), and for various subpopulations (such as Aboriginal peoples and immigrants) that are generally referred to as 'domains of interest.' In order to reduce response burden, the long-form census questionnaire is administered to a random sample of households.

This sampling approach and the total non-response introduce variability in the estimates that needs to be accounted for. This variability also depends on the population size and the variability of the characteristics being measured. Furthermore, the precision of estimates may vary appreciably depending on the domain or geography of interest, in particular because of the variation in response rates. For more information on the variability due to sampling and total non-response in long-form census questionnaire estimates, please refer to the Guide to the Census of Population, 2016, Catalogue no. 98-304-X.

Non-response bias for the long-form estimates

Non-response bias is a potential source of error for all surveys, including the long-form census questionnaire. Non-response bias arises when the characteristics of persons who participate in a survey are different from those who do not.

In general, the risk of non-response bias increases as the response rate declines. For the 2016 long-form census questionnaire, Statistics Canada adapted its collection and estimation procedures in order to mitigate, to the extent possible, the effect of non-response bias. For more information on these mitigation strategies, please refer to the Guide to the Census of Population, 2016, Catalogue no. 98-304-X.

Certification of final counts

Once data processing, editing and imputation were completed, the data were tabulated to represent the total Canadian population. Certification of the final counts was the last step in the validation process leading to recommendation for release of the data for each geography and domain of interest. Based on the analysis of data quality indicators and the comparison of the census counts with other data sources, the recommendation is for unconditional release, conditional release or non-release for quality reasons. In the case of conditional release or non-release, appropriate notes and warnings are included in this guide. Several data sources were used to evaluate the census counts. However, since the risk of error often increases for lower levels of geography and for smaller populations, and the data sources used to evaluate these counts are less reliable (or not available) at these lower levels, it can be difficult to certify the counts at these levels.

Census counts are also subject to confidentiality rules that ensure non-disclosure of individual respondent identity and characteristics. For more information on census confidentiality rules, please refer to the Guide to the Census of Population, 2016, Catalogue no. 98-304-X.

Comparability

Comparability of concepts over time

The 2016 Census of Population questionnaire contained the same four language questions asked of the entire Canadian population as the 2011 Census: on the ability to conduct a conversation in English or French (Question 7), on the language spoken most often at home (Question 8a), on other languages spoken on a regular basis at home (Question 8b), and on the language first learned at home in childhood and still understood (mother tongue) (Question 9). These questions were asked in the same order as they had been asked in 2011.

In addition, the same three language questions were asked to a sample of the Canadian population in 2016 as in 2011: on knowledge of non-official languages (Question 16), on the language used most often at work (Question 45a) and on other languages used regularly at work (Question 45b). These questions were asked to one in four households in 2016, compared with one in three households in 2011. There were also minor changes in the numbering of these three questions since 2011 (Table 1).

A change in the order of the language questions between 2006 and 2011 had created some observable changes in response patterns between the two periods, namely, an increase in the reporting of multiple languages spoken. In contrast, 2016 Census language data are largely consistent over time with 2011 Census language data. Readers can refer to a complete analysis of the factors affecting comparability of language results in the Methodological Document on the 2011 Census Language Data, Catalogue no. 98-314-XWE2011051.

A change did occur in the 2016 Census questionnaire response options for the questions on the mother tongue, language spoken most often at home and other language(s) spoken regularly at home on the short- and long-form questionnaires. The same change occurred for the questions on the knowledge of non-official languages, language used most often at work and other language(s) used regularly at work on the long-form questionnaire. For these questions, the wording of the "Other – Specify" response choice was changed to "Other language – specify." This change did not have an impact on the data quality or the distribution of answers to the language questions.

Another change for the 2016 Census was the introduction of prompts into the electronic questionnaire for the language questions. When a respondent initially provided a broad response of "Aboriginal," "Chinese," "Cree," "Indian," "Native," "Slavey" or "Tutchone," a prompt appeared on the respondent’s screen asking for a more specific response and offering examples. These prompts resulted in an increase in the precision and quality of data related to several specific language categories, such as Mandarin, Cantonese and Swampy Cree. As a result, when comparing 2016 Census data on the language categories that are associated with these broad responses with past censuses, data users are encouraged to compare these categories at the language family level. Please see Appendix 3.0 in the Dictionary, Census of Population, 2016, Catalogue no. 98-301-X, for the concordance of language families between the 2016, 2011 and 2006 censuses.

All trend analyses presented for the 2016 Census language release and its accompanying products compare 2016 Census short-form data with 2011 Census short-form data. Prior to 2011, language data and analysis published using census data were based almost exclusively on responses from the long-form census questionnaires administered to a sample of the population.

When Census Program data are compared over time—for example, comparisons between the 2016 Census and the 2011 National Household Survey (NHS) or the 2006 Census—changes in coverage and measurement need to be taken into account. Changes to coverage include, for example, changes in employment patterns over time. Changes in measurement can include changes to the questionnaire and legal changes.

Differences in methodology for the 2011 National Household Survey

The 2011 National Household Survey (NHS) employed a different methodology from that used for the 2016 Census, the 2006 Census and prior censuses. These differences can affect comparability between 2016 Census estimates and 2011 NHS estimates. For more information on the comparability between the 2016 Census and the 2011 NHS, please refer to the Guide to the Census of Population, 2016, Catalogue no. 98-304-X.

Comparability between the 2016 Census short-form and long-form language data

Both the short-form (100%) and long-form (25%) data from the 2016 Census of Population provide information on the Canadian population for various levels of geography and for numerous common topics (e.g., demography, marital status, family and language). For two main reasons, differences can exist between the 2016 Census long-form estimates and the 2016 Census short-form counts for the topics that are common to both.

First, the target population is different for each data source. The target population for the 2016 Census short-form data includes usual residents in collective dwellings and persons living abroad, whereas the target population for the 2016 Census long-form excludes them.

Second, the 2016 Census long-form estimates are derived from a sample survey and are therefore subject to sampling variability.

For more information on sampling and weighting for the long-form please see the Guide to the Census of Population, 2016, Catalogue no. 98-304-X.

Comparability with other data sources

Statistics Canada disseminates a wealth of data on languages. In addition to disseminating data on languages from the Census of Population, Statistics Canada publishes language data collected by the Aboriginal Peoples Survey, the General Social Survey, the Labour Force Survey and other household surveys.

Many factors affect comparisons of language data across these sources. Among other factors, comparability is affected by differences in survey target populations, reference periods, sampling and collection methods, question wording, questionnaire format, examples and instructions, approaches to data processing, and the social and political climate at the time of data collection.

Data quality notes

At the national level, the 2016 Census imputation rate for the mother tongue question is 4.3% (Table 2). The questions on the language spoken most often at home and knowledge of official languages have imputation rates of 3.9% and 4.0%, respectively. The imputation rate for language used most often at work is 3.1%.

At the provincial level, the imputation rate for the question on mother tongue ranges from 3.9% in Ontario to 7.8% in Nunavut. For the question on the language spoken most often at home, the rate ranges from 3.5% in Ontario to 7.7% in Nunavut. For the question on knowledge of official languages, the imputation rate ranges from 3.6% in Ontario to 7.7% in Nunavut. For the question on language used most often at work, the imputation rate ranges from 2.6% in Nunavut to 5.8% in Newfoundland and Labrador.

The imputation rates for the language data in collective dwellings are higher than the imputation rates for the population living in private households and overseas (Table 3). For example, in 2016, the mother-tongue imputation rate for the population living in private households and overseas is 3.2%, compared with 62.3% for the population living in collective dwellings. The high imputation rates in collective dwellings are an effect of the challenge of collecting data for this population. Throughout past census cycles, Statistics Canada has used various collection approaches to enumerate persons living in collective dwellings, including direct enumeration and administrative records. The procedures used for 2016 were similar to those used in 2011. The intent was to gather as much information as possible from administrative records. When the information required was not available in the administrative records, Statistics Canada worked with a representative from the facility to complete the census questionnaire or interviewed residents to collect their census information.

For the 2016 Census and previous censuses, language data and analysis have been disseminated for the population excluding institutional residents. The population living in non-institutional collective dwellings is generally included in language data and analysis disseminated by Statistics Canada. As a result of higher imputation rates, users are advised to exercise caution when analyzing language data for small geographic areas with large proportions of individuals living in collective dwellings.

Additional language questions are asked on the 2016 Census long-form questionnaire for the population in private households. These include the knowledge of non-official languages question, the language used most often at work question and the other language(s) used regularly at work question. Questionnaires in which respondents did not respond to the knowledge of non-official languages question, or responded with 'none', were examined to determine if respondents had reported a non-official language as their mother tongue or home language. If the knowledge of non-official languages question was not responded to and a non-official language was reported to the mother tongue or home language questions then the information for the knowledge of non-official languages question was updated accordingly. If a non-official language was not mentioned in the home language or mother tongue question, then a response of 'none' was assigned. Also, if someone responded 'none' to the non-official language question but had a non-official home language then this language replaced the 'none' response. Overall, this resulted in 2.4% of responses being edited for the knowledge of non-official languages question for Canada.

Validation of increase in the use of English at work in Quebec

An increase in the overall use of English at work was observed in Quebec from 40.4% in 2006 to 42.5% in 2016. This increase in the use of English at work was also observed for certain areas and cities throughout the province of Quebec including the metropolitan areas of Québec, Saguenay and Rimouski. This increase was driven by several factors including increases in certain jobs where individuals are more likely to use English at work (such as computer programmers) and the evolution in certain sectors (such as working more in English to deal with international stakeholders). At a local level, these increases are not only driven by the factors mentioned above but also by local industry. For example, the increase of English used at work in Saguenay is largely driven by an increase in English workers in defence services and computer systems design and related services.