Which indicators are most useful for comparing citizenship policies? - Anita Manatschal: On the relevance of comprehensive comparative analyses at the subnational level
- Which indicators are most useful for comparing citizenship policies?
- David Reichel: We need different indicators for different research questions.
- Maarten Vink: Naturalisation rates and rejection rates measure different phenomena, and have different problems
- Jan Willem Duyvendak, Rogier van Reekum, Peter Scholten, Christophe Bertossi: The of/for distinction
- Thomas Janoski: What we need citizenship indicators for depends on who are “we”
- Thomas Huddleston: From politics to impact: How citizenship really works
- Anita Manatschal: On the relevance of comprehensive comparative analyses at the subnational level
- Concluding remarks by Marc Helbling
- All Pages
Anita Manatschal: On the relevance of comprehensive comparative analyses at the subnational level
March 31, 2011
The citizenship forum kicked off by Marc Helbling brings up very interesting and important aspects which should be considered by current analyses on citizenship and related policies. Focusing on subnational variations of integration policy, I will pick up some of the aspects raised in the preceding comments. Essentially, I will defend two arguments: first, I claim that we should apply a constant conceptual framework (e.g. policy indicators) when analyzing the same aspects at different analytical levels (i.e. national, regional, local). Second, while I consider Duyvendak et al.’s distinction between sociology of/for policy necessary and important, I tend to agree with Huddleston who points to the relevance of analyzing both, as I think that occasionally such combined research approaches are more revealing.
In his comment Janoski suggests to extend the analytical scope from the national to the subnational level. I strongly agree with this suggestion for three reasons: empirically, subnational variations of integration policy remain largely neglected, although they are particularly pronounced in federal immigrant states such as the United States, Canada, Austria, Germany, or Switzerland (cf. Akgün and Tränhardt, 2001). Methodologically, the subnational comparative level provides clear analytical benefits, as it offers the best approximation to a most similar cases setting, whereas the same degree of comparability is barely reachable at the international comparative level. Third, from an epistemological perspective leading comparativist Lijphard (1971) has demanded for a long time that empirical results from the international level should be validated at the subnational comparative level. It becomes clear that if we are interested in validating research results at the subnational level, we would need to keep the conceptual framework constant. Thus, when shifting the analytical level while following the same research purpose, the question is not primarily which indicator of citizenship or integration policy to choose, but that this indicator should be kept constant at different analytical levels.
As this is the purpose of my present research project, I transferred an internationally established instrument of integration policy to Switzerland’s subnational, cantonal level. I drew on the conceptual framework elaborated by Koopmans et al. (Koopmans, 2010, Koopmans, et al., 2005), which covers in large parts similar components as the MIPEX. The components measured at the cantonal level include access to the labour market, access to nationality, political participation, anti-discrimination, family reunion, the cultural requirements for naturalisation and religious rights. For a detailed and critical discussion of this transfer see Manatschal (forthcoming). The figure below illustrates the aggregated values (additive index based on components mentioned above) for the 26 cantonal integration policies, whereas values below zero stand for restrictive policies, and values above zero represent liberal integration policies. A look at the figure clarifies that the common characterization of Switzerland as an assimilationist country neglects a considerable subnational variety which is hidden behind this uniform national classification.
In a second step, I will explain why I consider it potentially more illuminating to develop comprehensive research projects accounting for both, “sociology for and of policy”, to use the terminology of Duyvendak et al., as opposed to isolated research projects focusing on only one aspect. I agree with Duyvendak et al. that the distinction between sociology of policy versus sociology for policy constitutes an important contribution to the theoretisation of two distinct overall research programmes. I furthermore agree with the point made by the authors (also raised by Janoski) that researchers should be careful not to mix the two research programmes unconsciously.
However, similar to Huddleston, I assume that the two research programmes can be very well combined in one and the same comprehensive research design, provided the combination is done consciously and with enough analytical and methodological rigour, meaning that the two research programmes are clearly classified as distinct analytical parts of an overall research project. Whereas Helbling, Reichel and Vink use the notion of “output” and “outcome” to refer to this distinction, Huddleston suggests to refer to policy as “outcome” and “input” respectively. As for my quantitative comparative analyses, I apply an even simpler terminology and speak of integration policy as “dependent” (sociology of policy) or “independent variable” (sociology for policy) respectively. While I think it is important that researchers take seriously Duyvendak et al.’s warning (also raised by Favell (2001)) that research on citizenship should not be dictated by governments or policy advocates, I do agree with Huddleston that research on citizenship is potentially most revealing if we consider all the aspects of the policy cycle, including the evolution of policies, politics, discourse, implementation, outcomes, and impact.
I will illustrate this point by coming back to my cantonal data on integration policy. Comparative cross-sectional regression analyses revealed that the attitude of the cantonal population towards immigrants, which is influenced by the cultural-linguistic French- or German-speaking heritage of a canton, is clearly related to cantonal integration policies (Manatschal, forthcoming). This pattern is already discernible in the univariate figure above: five of overall six predominantly French-speaking cantons (JU, NE, GE, VD, FR) account for the most liberal cantonal integration policy indices. One might now assume that cantonal public attitudes towards immigrants, as they are for instance expressed in national or cantonal votes on citizenship attribution, influence cantonal formulations of integration policy. However, one could argue that the causality works also in the other direction, i.e. that cantonal policies shape public attitudes on immigrants. While these are not policy outcomes regarding the immigrant population, at which such policies are addressed (and which seem to be in the focus of Duyvendak et al.’s concept of sociology for policy), they are nevertheless policy outcomes of concern when interested in the dynamics of policies. Even more plausibly, it might be argued that the relationship between attitudes and citizenship or integration regimes is mutually reinforcing (cf. Weldon, 2006). It becomes clear that the possibility of such a mutual relationship cannot be accounted for adequately if we would consider integration policy only as an outcome, or as an input. A study of such complex causalities requires a comprehensive analysis which keeps the whole policy cycle in mind.
Abbreviations of Swiss cantons
AG= Argovia; AI = Appenzell Inner Rhodes; AR = Appenzell Outer Rhodes, BE = Berne; BL = Basel-Country; BS = Basel-City; FR = Fribourg; GE = Geneva; GL = Glarus; GR = Grisons; JU = Jura; LU = Lucerne; NE = Neuchâtel; NW = Nidwald; OW Obwald; SG = St. Gall; SH = Schaffhausen; SO = Solothurn; SZ = Schwyz; TG = Thurgovia; TI = Ticino; UR = Uri; VD = Vaud; VS = Valais; ZG = Zug; ZH = Zürich
Akgün, L. and D. Tränhardt (2001). Integrationspolitik in föderalistischen Systemen. Münster: Lit.
Favell, A. (2001). Integration Policy and Integration Research in Europe: A Review and Critique. In Citizenship today: global perspectives and practices. Washington D.C.: Carnegie Endowment for International Peace (349-399).
Koopmans, R. (2010). Tradeoffs between Equality and Difference: Immigrant Integration, Multiculturalism and the Welfare State in Cross-National Perspective. Journal of Ethnic and Migration Studies 36(1): 1-26.
Koopmans, R., P. Statham, M. Giugni and F. Passy (2005). Contested citizenship. Immigration and Cultural Diversity in Europe. Minneapolis; London: University of Minnesota Press.
Lijphart, A. (1971). Comparative Politics and the Comparative Method. American Political Science Review 65(3): 682-693.
Manatschal, A. (forthcoming). Path Dependent or Dynamic? Cantonal Integration Policies between Regional Citizenship Traditions and Right Populist Party Politics. Ethnic and Racial Studies.
Weldon, S. A. (2006). The Institutional Context of Tolerance for Ethnic Minorities: A Comparative Multilevel Analysis of Western Europe. American Journal of Political Science 50(2): 331-349.