![Sample our Social Sciences journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110822/TOC-Subject-Sample-2021-Social-Sciences.jpg"})
ABSTRACT
The concept of national boundaries has been employed extensively in the social sciences, with most research focusing on the host society's perspective. This study innovates by measuring immigrants’ perceptions of how boundaries of their host nation are drawn and examining how such perceptions impact their national identifications in a multinational context, more specifically in the province of Quebec in Canada. It relies on a stratified sample of first- (n = 1129) and second-generation immigrants (n = 1286) as well as non-immigrants (n = 1472). We show that boundary perceptions impact the availability and attractiveness of different identity options, including identification with Quebec, the country of origin, and Canada. First, while perceptions of ascriptive boundaries to Quebec push first- and second-generation immigrants away from Quebec identity, the reactive effect of strengthening identification with the country of origin and Canada is limited. Second, although the second generation perceives Quebec boundaries as more ascriptive than the first generation, it is not more strongly impacted by their boundary perceptions in their identifications. Third, perceptions of attainable boundaries promote all three national identifications, but only for the first generation. Finally, we demonstrate that attention to boundary perceptions adds important explanatory leverage beyond the impact that discrimination has on national identifications.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Correction Statement
This article has been corrected with minor changes. These changes do not impact the academic content of the article.
Notes
1 The definition of interculturalism can vary across national contexts. In the context of Quebec, Bouchard defines interculturalism as ‘a model based on the search for equilibrium, which aims to chart a path between assimilation and segmentation, and which, to this end, emphasizes integration, interaction and the promotion of a common culture while respecting rights and diversity’ (Citation2012: 51) (our translation).
2 It is worth noting here that their study is limited to adolescents and that the main finding is the greater predominance of dual identifiers among immigrants in comparison to non-immigrants.
3 The authors are solely responsible for the analysis and interpretation of the data.
4 Respondents received an invitation email that contained a unique invitation link (url) that could only be used once. Respondents were all members of the Léger web panel of more than 400,000 people currently living in Canada who were recruited randomly over the phone (61 percent) or through various other means. The annual recruitment rate for the panel is approximately 15,000 new members a year, while about 10,000–12,000 panelists are removed from the panel or opt out each year. Panelists are rewarded for their participation over time with a series of financial incentives. We evaluate the response rate to be about 17%. However, no exact response rate can be calculated for an online survey because, unlike telephone surveys, it is not possible to evaluate whether people refused to participate or did not read or receive the invitation.
5 We measure boundary perceptions in relation to the Quebec national community. We focus on boundary perceptions in Quebec (as opposed to those of Canada) because Quebec is a more proximate national entity (than Canada) and the one primarily engaged in immigrant integration and service delivery to citizens.
6 ‘Being Catholic’ was included because Quebecers have been historically in vast majority Roman Catholics and the Catholic Church has played a critical influence in management of education and social welfare in the province (Meunier et al. Citation2011). Moreover, we distinguish between being able to speak French and speaking it with the majority accent. Whereas some see language as a neutral mean of communication – attainable marker – , others associate language usage with assimilation into a deeply rooted cultural identity – ascriptive marker – (Hjerm Citation1998, 336, 340).
7 We presume that the causality goes from boundary perception to strength of identification, but this is only presumed and not demonstrated. It possible that the causal relation runs the other way around.
8 Alpha score for first- and second-generation immigrants combined.
9 When second generation respondents have two parents born in different countries, we used the mother's country of birth (except if the mother was born in Canada).
10 It is difficult to assess the representativity of the samples for first- and second-generation respondents as the benchmark are not easily accessible: for first generation, the sample is limited to immigrants arrived post 1977, and there is hardly a reliable benchmark for second generation immigrants. We recognize though that the sample might not be completely representative.
11 When controlling for the sociodemographic profile of respondents, first-generation immigrants report slightly more attainable boundaries than non-immigrants, but no difference is reported for the second generation. Results not presented.
12 Differences between immigrants and non-immigrants are statistically significant even when controlling for the sociodemographic profile of respondents. Results not presented.
13 The results are the same if we remove discrimination from the model.
14 The results are the same if we remove discrimination from the model.
15 Detailed results in Appendix 2 indicate that nonwhite immigrants express greater identification with Quebec than white immigrants (B coefficient = −.05), which might seem counterintuitive. Further analyses indicate that this finding holds only when control variables are included in the analysis. Bivariate analysis indicates that nonwhite immigrants actually express a weaker sense of identification with Quebec than white immigrants. No single control variable in the model can account for this finding.