Mother-Tongue Maintenance in Canada: The Role of the Coethnic Community and Coethnic Resources

ABSTRACT

Much of our understanding of language attainment among immigrants and their descendants focuses on the US with substantial evidence of mother-tongue loss within a few generations. However, these patterns may differ in Canada given its unique linguistic context, such as continuous migration, multicultural policy, two official languages, and a skilled migration policy. Thus, this study focuses on several contexts under which immigrants and their children are received. Specifically, this article examines the role of the coethnic community and coethnic resources on unofficial and official language proficiency among the first, 1.5, and second generations in Canada. Using data from the 2002 Ethnic Diversity Study and 2001 Canadian microdata, this study finds that simultaneous proficiency in non-official and official languages is structured by the availability of coethnic resources and the average size and education of the coethnic community. Specifically, individuals with greater access to coethnic resources and many coethnic neighbors are more likely to maintain the mother tongue with the official language(s).

KEYWORDS:

• Mother tongue
• immigrant integration
• linguistic assimilation
• Canada

Acknowledgements

While the research and analysis are based on data from Statistics Canada, the opinions expressed do not represent the views of Statistics Canada. The analysis presented in this paper was conducted at the Quebec Interuniversity Centre for Social Statistics which is part of the Canadian Research Data Centre Network (CRDCN). The services and activities provided by the QICSS are made possible by the financial or in-kind support of the Social Sciences and Humanities Research Council (SSHRC), the Canadian Institutes of Health Research (CIHR), the Canada Foundation for Innovation (CFI), Statistics Canada, the Fonds de recherche du Québec and the Quebec universities. The views expressed in this paper are those of the authors, and not necessarily those of the CRDCN, the QICSS or their partners. I thank the editorial board and anonymous reviewers for their comments on previous versions of this manuscript and Alessandra Chinsen for her research assistance. The project was generously supported by grants from Quebec Inter-University Centre of Social Statistics, International Council for Canadian Studies, and Foreign Affairs and International Trade Canada. I thank the QICSS at the University of Montreal for their assistance.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes

1 Census tracts are small geographic areas ranging from 2500−8000 persons whereas metropolitan areas have a minimum population of 100,000 persons.

2 One limitation is that EDS does not include information on factors that may shape official language proficiency, such as exposure or consumption to mainstream media and language of instruction in schools.

3 I measured coethnic resources in a similar manner to Breton (1964), who measured the institutional capacity of organizations in ethnic communities based on completeness of information. I included a variety of search words in several Internet search engines to identify each resource. I counted most resources with websites in English. In rare instances, I included resources that did not have a website, but included ample information for me to decipher the purpose of the resource.

4 I measure coethnic resources at the metropolitan level. It is possible that individuals living in suburbs may have access to coethnic resources more locally. However, coethnic resources in suburbs are more difficult to capture as they are likely to be smaller, informal organizations that serve a narrower population or do not have an online presence.

5 In previous analyses, I tried grouping resources by size or type. I found that linear operationalizations and quartiles were not significant. I find that a non-linear operationalization of coethnic resources is better suited to my data, which may suggest that non-linear effect of coethnic resources. In separate analyses, I also found that a dichotomous variable (12-17 vs. 0-11) was significant. Ultimately, I included a three category operationalization as it is more informative and shows the distinction across access to medium and high number of resources versus low. While the grouping for the categories is asymmetrical, there is support that access to a higher number of resources is associated with a significant difference in language proficiency. I also tried coding coethnic resources as a series of scales and conducted a factor analysis to test which of these variables loaded highly together. However, only one scale was significant in the analyses—the presence of ethnic arts and the presence of a consulates-general in the city—which theoretically are not well-motivated.

6 In separate analyses, I examined the regression separately for English-speaking provinces and Quebec and obtained similar results.

7 Svy commands in Stata do not permit the use of weights and clustered robust standard errors simultaneously. ‘Common Errors and How to Avoid Them.’ UNC: Carolina Population Center. from http://www.cpc.unc.edu/research/tools/data_analysis/statatutorial/sample_surveys/svy_errors (accessed May 15, 2018.)

8 A significant F-test for Table 3, Model 1 indicates sufficient evidence that the regression model fits the data better than a model with no independent variables or that the independent variables improve the fit of the model.

9 Table 4 and Appendix Table 2 show that the F-test is significant for the full models for each group. Despite differences across groups, the full regression models fit the data better than a model with no independent variables.

Additional information

Funding

This work was supported by Quebec Inter-University Centre for Social Statistics (QICSS)/Fonds de recherche sur la société et la culture (FQRSC); Foreign Affairs and International Trade Canada (DFAIT); International Council for Canadian Studies (ICCS).

Notes on contributors

Rennie Lee

Rennie Lee, senior research fellow at Institute for Social Science Research at University of Queensland, is a sociologist specializing in racial, ethnic, and migrant stratification. Her current research examines immigrant, racial, and ethnic groups in North America and Australia with an emphasis on the intersection of gender and immigration policy on immigrant outcomes.