ABSTRACT
This paper explores the cross-national variation in the gap in university expectations between native and second-generation immigrant students and whether it is moderated by characteristics of the country of residence of children of immigrants. Three domains have been considered at the national level: the context of reception of immigration (legal status of non-national immigrants and views about immigration of the local population); labour market segmentation and education systems (external differentiation). The study tests the general hypothesis that settings that enhance individual autonomy and widen room for choice by removing formal and informal barriers that migrants and their descendants face in everyday life and in specific domains (education and labour market) are associated with a larger gap in university expectations that is favourable to children of immigrants (a larger ‘immigrant advantage’). Applying a two-stage analysis to PISA 2015 data, results provide overall support for this hypothesis. More welcoming contexts of reception are associated with a larger immigrant advantage. External differentiation is negatively associated with it, both for students attending general and vocational tracks, especially the latter, as compared to non-tracked students attending common tracks. There is not a robust association between labour market segmentation and the university expectations’ gap.
Acknowledgements
I am grateful to the funding institutions for providing financial support for this research; to Luis Ortiz Gervasi; the anonymous reviewers of the journal, and to those who commented earlier versions of this paper.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
The raw datasets containing the individual-level data analysed in this study are publicly available under the form of Public Use Files (PUF) in https://www.oecd.org/pisa/data/2015database/.
Notes
1 First-generation immigrants and inter-ethnic second-generation immigrants were included as categories of the variable, but results are only reported for second-generation migrants, as they are the focus of this work. I did this to have a larger sample to achieve more accurate estimates.
2 See Jerrim et al. (Citation2017) and OECD (Citation2009) for details on analysing plausible values.
3 Migrant Integration Policy Index: https://www.mipex.eu/what-is-mipex.
4 Estonia is a country where second-generation immigrants are disadvantaged also in the transition to general upper secondary education (instead of professional or vocational upper secondary education) and the transition to tertiary education (Lindemann and Saar Citation2012, Citation2011). Second-generation immigrants in Estonia belong mainly to the Russian minority. Their educational disadvantage (in expectations and educational transitions) seems to be the result of a very constraining opportunity structure characterised by language and cost barriers to access university (tuition-free higher education is only provided in Estonian and fluency in this language among Russian-origin second generation is not widespread), worst labour market returns than ethnic Estonians and an undermined citizenship status (Lindemann and Saar Citation2012, Citation2011; Saar, Krusell, and Helemae Citation2017).
5 Sweden is an outlier due to its comparatively high immigrant advantage and MIPEX score.
6 Great Britain is an outlier given its high conversion rate and large immigrant advantage.
7 In Models 5a and 5b, the MIPEX and MAI scores are not included simultaneously because they have a medium-to-high correlation (r = 0.57) that could cause multicollinearity and, thus, estimation problems.
8 Jackknife estimates are obtained by excluding one country at a time from calculations.