Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=csje20
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/csje20
Pre-Migration Status, Social Capital, and the
Educational Aspirations of Children of Immigrants
in Disadvantaged Swedish Schools
Olav Nygård
To cite this article: Olav Nygård (2021): Pre-Migration Status, Social Capital, and the Educational
Aspirations of Children of Immigrants in Disadvantaged Swedish Schools, Scandinavian Journal of Educational Research, DOI: 10.1080/00313831.2021.1897878
To link to this article: https://doi.org/10.1080/00313831.2021.1897878
© 2021 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
Published online: 17 Mar 2021.
Submit your article to this journal
Article views: 204
View related articles
Pre-Migration Status, Social Capital, and the Educational
Aspirations of Children of Immigrants in Disadvantaged Swedish
Schools
Olav Nygård
Department of Culture and Society, Institute for Research on Migration, Ethnicity and Society, Linköping University, Norrköping, Sweden
ABSTRACT
High aspirations can be an important factor for educational attainment, especially for youth in disadvantaged schools who are otherwise more likely to leave school early. In this article, I study the relationships between pre-migration status, social capital, and educational aspirations among youth in disadvantaged Swedish schools, using data on 960 students collected in 2014. Regression results showed that access to social capital was related to pre-migration status, and that both factors contributed to high university aspirations among children of immigrants, partly through high expectations from parents. Thefindings consequently show how post-migration resources and outcomes relate to pre-migration factors, challenging the destination country bias that is often present in studies on immigrants and their children.
ARTICLE HISTORY
Received 27 August 2020 Accepted 10 February 2021
KEYWORDS
Children of immigrants; social capital; migrant selectivity; aspirations; education
Introduction
Children of immigrants tend to have higher educational aspirations compared to children of non-immigrant parents (Feliciano & Lanuza,2016). This leads to a higher likelihood of positive tran-sitions at key educational choice points, for example into academic programmes at upper secondary level (Jackson et al.,2012). High aspirations are therefore an important asset, especially for youth in disadvantaged schools who are otherwise more likely to leave school early. In this article, I focus on the educational aspirations among children of immigrants in Swedish schools with low grade-point averages, and the factors associated with these aspirations.
Explanations for the high levels of educational aspirations among children of immigrants typically draw on either of two factors: selectivity in the migration process, and resources, norms and expec-tations in the immigrant families’ networks. From the former perspective, educational aspirations are high in immigrant families because their pre-migration status was high (Engzell, 2019; Feliciano & Lanuza,2017; Ichou,2014). These families retain attitudes and beliefs aligned with their pre-migration status, even when facing downward mobility through migration. Immigrant parents therefore maintain strong educational expectations for their children who, in turn, internalize them as aspirations.
From the latter perspective, immigrant families have networks that are rich in information, sup-port and other resources (Behtoui,2017; Shah et al.,2010). The better access to resources embedded
© 2021 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http:// creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way.
CONTACT Olav Nygård olav.nygard@liu.se Department of Culture and Society, Institute for Research on Migration, Ethnicity and Society, Linköping University, 601 74 Norrköping, Sweden
in networks—or social capital (Bourdieu,1986)—results in high expectations on the part of parents and significant others for children to succeed, to fulfil the dreams of the migrating generation, to counteract anticipated discrimination, or to live up to high group standards.
While these two explanations have mostly developed separately from each other, scholars have recently begun discussing them in combination (Khachikian & Bandelj,2019; Nygård & Behtoui,
2020). High socioeconomic status is usually associated with better access to resources, and more resource-rich networks (Lin, 2000). Positive selection therefore implies that many immigrant families had above-average access to migration resources and more resource-rich pre-migration networks. Such resources and networks could be carried over—for example in the form of embodied cultural capital and transnational networks—and thereby facilitate the accumu-lation of post-migration resources. Strong identification with others in a similar situation can also produce a bounded solidarity that increases resource sharing to attain common goals or overcome shared adversities (Portes & Sensenbrenner,1993).
Despite this, with a few exceptions (most notably Lee & Zhou, 2015) the relationship between pre-migration status and post-migration social capital has seldom been explored. To my knowledge, there is currently no study that tests the extent to which the two factors are related, or the extent to which they contribute to higher educational aspirations among children of immigrants in otherwise disadvantaged contexts. With this study, I take afirst step to filling this research gap. The aim is to study the relationship between pre-migration status and post-migration social capital, and the extent to which they contribute to educational aspirations for children of immigrants in disadvantaged schools. In concrete terms, I answer the following three questions:
(1) What is the association between pre-migration status and post-migration social capital through family contacts among children of immigrants in disadvantaged Swedish schools?
(2) How are pre-migration status and post-migration social capital related to educational aspirations?
(3) To what extent can these factors explain the difference in aspirations between children of immi-grants and children of non-immiimmi-grants?
Theoretical Frame and Previous Research
Young people’s aspirations for university are generally high (Berrington et al., 2016). However, differences in available resources and embodied experiences result in different estimations of the costs, benefits and likelihood of success. Working-class youth therefore tend to have lower edu-cational aspirations (Seghers et al.,2019).
By analogy, it could be suspected that children of immigrants would also have lower aspirations, since they often grow up under harsher economic conditions than children of non-immigrant families (Gustafsson & Österberg,2018). This, however, is not the case. Instead, aspirations tend to be higher among children of immigrants than among children of non-immigrant families given the same class backgrounds, leading one observer to note that while most immigrant families are working class in their labour market positions, they are middle class in their views on education (Tomlinson,1989). To explain this, previous research has often drawn on two broad categories of explanations: migrant selectivity and immigrant or ethnic networks.
Migrant Selectivity
It has long been recognized that migrants are not a random sample of the origin country’s popu-lation (Lee,1966). The factors that influence emigration—such as economic hardship, persecution, and prospects for the future—affect different parts of a population unequally. Obstacles to
migration are also surmountable to differing degrees depending on the resources available. Migration is therefore neither equally desirable nor equally feasible for everyone in a country.
The resultant selectivity has been increasingly highlighted in research, using education in the country of origin as an indicator of pre-migration status (Ichou,2014). However, pre-migration sta-tus does not always translate into outcomes after migration. Both context-specific and supposedly transferrable skills are often devalued in migration (Weiss, 2005). One reason is discrimination, while another is uncertainty about how to value foreign credentials. This means that pre-migration merits lose some of their signalling value (Guo,2009; Tibajev & Hellgren,2019). Policies can also institutionalize informal or formal barriers to labour market entry for newly arrived immigrants, resulting in lasting consequences for economic incorporation (Marbach et al.,2018). These and other inequalities are also further exacerbated by ethnic discrimination among both employers and employees (Bursell & Jansson,2018; Oreopoulos,2011).
Many migrants consequently have a lower post-migration than pre-migration status. However, parents with a high pre-migration status often retain high ambitions for their children, even when experiencing downward mobility. This results in high expectations for children’s educational suc-cess, making pre-migration status an important factor in the high educational aspirations among children of immigrants (Engzell,2019; Ichou,2014).
Immigrant Networks
The second major explanation for high educational aspirations among children of immigrants is that they have comparatively resource-rich networks. This might take the form of economic invest-ments in neighbourhoods, businesses or institutions, information and ideas circulated through minority-language channels, collective mobilization for shared goals, loans, and other forms of emotional and practical support that both increases the chances of educational attainments and produces norms and expectations of educational attainment (Lee & Zhou,2015).
The terms used to denote these resources typically allude to ethnicity, for example ethnic capital (Lee & Zhou,2017), ethnicity as social capital (Zhou2005), or ethnicized social capital (Nygård,
2017). However, immigrant communities are often ethnically mixed, and“ethnic businesses” typi-cally cater to multiple ethnicities from the same country or region of origin. Thus, Nederveen Pie-terse (2003) argues that the social capital that is invested in and arises from immigrant communities and institutions is typically cross-cultural, making it rainbow social capital rather than ethnic capi-tal. In Sweden, immigrant families are also likely to live in suburbs containing many different countries of origin (Andersson,2000), and to maintain both local and transnational ties (Andersson et al.,2018). Social capital might therefore also be embedded in a diaspora network (see Ferguson et al.,2016). Networks also vary from family to family. Empiricalfindings are consequently mixed with regard to whether immigrant families have more or less resource-rich social networks than non-immigrant families (for overview, see Nygård & Behtoui, 2020). This further suggests that the association between ethnicity and network resources is neither simple nor straightforward. I will therefore simply use“social capital” to denote resources available through membership of net-works or groups (Bourdieu,1986), regardless of the basis for these networks.
Immigrant families are not always found to have better access to social capital than non-immi-grant families with similar class positions. However, the fact that some are—despite facing more adverse conditions than non-immigrant families—is typically attributed to bounded solidarity. Portes and Sensenbrenner (1993, p. 1325) describe this process as a “situational reaction of a class of people faced with common adversities. If sufficiently strong, this emergent sentiment will lead to the observance of norms of mutual support, appropriable by individuals as a resource in their own pursuits”. However, better access to social capital in immigrant families’ networks can also reflect transnational ties and resources from the country of origin (Lee & Zhou,2017). Positive selectivity can consequently result in networks that are permeated by middle-class culture and ideals, including the valuation of education, irrespective of the network members’ current class
positions. The expectations of parents, peers and others in the social network in turn affect the aspirations of youth (Behtoui,2017).
Context
Due to selectivity, social capital, or both, children of immigrants often exhibit strong educational aspirations. However, whether or not these aspirations can be realized depends not only on the individual’s ambitions and efforts, but also on factors relating to segregation in society, diaspora organization, and the circumstances of reception and integration (Portes et al., 2005). Another major factor is the education system.
The Swedish education system is comprehensive. Children enter primary school at the age of seven for thefirst stage of education (grundskola), spanning grades one through nine. A non-com-pulsory “pre-school year” (förskoleklass) was introduced in 1998 (and made compulsory from 2018), effectively extending basic education to 10 years.
Upper secondary education is optional, but in practice is a requirement for most jobs. The vast majority of youth therefore continue to upper secondary education after completing compulsory education (Statistics Sweden,2017b). Many also pursue further education.
Both compulsory and post-compulsory education is free and subsidized through grants and loans. There is also a system of second-chance education for students who struggle or drop out of education. Reaching university is therefore relatively attainable for Swedish youth.
At all levels, parents are expected to choose schools on a pseudo-market where different edu-cation providers—including for-profit companies—compete for student enrolment (see Hennerdal et al.,2020). The effects of this system are debated, but evidence suggests that school choice has resulted in increased differences between schools, even though residential segregation remains the main driver of school segregation (Brandén & Bygren,2018). Currently, between-school vari-ation in test scores is larger in Sweden than in neighbouring countries (OECD,2016).
A central problem is the concentration of facilitating factors in and around some schools, and the concentration of impeding factors in and around others (Holmlund et al.,2019). This results in some schools having the advantages of qualified teachers, active parents, motivated peers, and rich extracurricular programmes. Other schools are disadvantaged as a result of temporary or unqualified teachers, unemployment and family poverty, youth delinquency, and a lack of services and activities outside school. Due to school and neighbourhood segregation, children of immigrants are typically overrepresented in the disadvantaged schools, to the point where the two concepts are sometimes perceived as synonymous (Bunar & Ambrose,2016; Voyer,2019). This stigmatization adds to the degree of disadvantage, leading resource-rich families to opt out and furthering the seg-regation between schools.
Migrant Selectivity in Sweden
Sweden has a long history of migration. However, it was only after World War II that Sweden chan-ged decisively from a country of origin to a country of destination. The early immigrants were lar-gely drawn by demand for labour. Most came from southern Europe and Finland, and many had working class backgrounds (OECD & EU,2015). In the last decades of the twentieth century, fol-lowing restrictions on labour migration, the character of immigration changed to include a larger proportion of refugees and family reunification migrants. Through the EU agreements on free movement and the 2008 labour migration legislation (Calleman, 2015), Sweden is again seeing an increase in labour migrants and growing international student migration.
Today, approximately one in four Swedish children have two foreign-born parents (Statistics Sweden,2017a). Internationally, migrants are generally reported to be positively selected relative to the sedentary population (van Tubergen et al.,2004). However, unlike many countries, Sweden gives no formal priority to the well-educated. Sweden also has a high number of recent immigrants
per capita, and a comparatively large proportion of refugees and family reunification immigrants (OECD, 2017). Such factors are sometimes argued to result in negative selection (Razin & Wahba,2015). Still, contemporary research has found considerable degrees of selection to Sweden, with immigrants from many countries being positively selected (Polavieja et al.,2018). On balance, the general level of selectivity therefore seems to be on par with that of the UK and the USA (van de Werfhorst & Heath,2019).
Data and Measurements
The data for this study was collected by Statistics Sweden in the spring of 2014, through a survey of students at the end of compulsory school or in early upper secondary education in disadvantaged schools (Behtoui,2017). The schools in the sample were selected on having below median grade-point averages from the preceding years, and were located in the three largest cities in Sweden: Stockholm, Gothenburg and Malmö. With consent from guardians and students, all completed sur-veys were later supplemented with registry data on grades, parents’ level of education, and parents’ region of origin. After excluding cases with missing data, thefinal sample contained 960 cases, com-ing from 41 schools. To account for compositional differences, I controlled for gender, school municipality, and whether the student was surveyed before or after the transition to upper second-ary education.
The main outcome variable was university aspirations. I based this variable on survey responses to the question“Which level of education would you like to attain before you start working?” Since upper secondary education is the de facto norm in Sweden, I coded the variable 1 for university education and 0 otherwise. In the survey, respondents were also asked about the expectations they perceived from their parents. Based on the question“What is the highest level of education your parents expect you to attain?”, I created the variable parents’ expectations, coded 1 when respondents believed their parents expected them to complete university education and 0 otherwise. Grades were converted into percentages, giving a continuous variable ranging from 0 to 100 where a higherfigure indicates a higher grade.
To distinguish between students with two, one, or no foreign-born parents, I created the categ-orical variable immigrant background. Students with only one parent listed were coded as having either Swedish or foreign background depending on that parent.
Parents’ Education
I used two measures of parents’ education: years of education and education rank. Years of edu-cation is an indicator of eduedu-cational merits. Eduedu-cation rank is in turn an indicator of status or pres-tige associated with that level of education, relative to the context of attainment. For non-immigrant families, the two are strongly correlated. For immigrant families, however, the two differ substan-tially due to the differences in educational systems and incomplete recognition of foreign merits. Education rank can therefore be a better indicator of subjective social status when comparing families with very different access to education.
Years of education was coded as the number of years normally required to reach the relevant level of education, ranging from 5 for“some compulsory education” to 20 for “PhD or above”.
Parents’ education rank was calculated using the most recent information in the Barro and Lee (2013) data. I followed the procedure used in previous research (Ichou,2014), comparing each parent’s level of education to the proportion of adults (25+) of the same gender with equal or lower level of education in the country of origin. Using this information, I calculated the parent’s education rank as the percentage of people in the comparison group with a lower level of education, plus half of the percentage with the same level of education. The resultant measure ranges from just above 0 to just under 100, and can be read as the percentile in the education distribution corre-sponding to the parent’s level of education. I averaged results for both parents whenever possible,
since this does not introduce a gender bias in the two measures and is generally suggested to pro-duce better estimates (Thaning & Hällsten,2020).
To approximate the country of origin, I combined answers from an open survey question on the languages spoken at home with registry data on the parents’ region of origin. Ambiguous cases were excluded, as were cases where parents were born in different regions outside Sweden.
I coded languages based on data from the CIA World Factbook (CIA,2019). A country was identified if the indicated language was only listed under one country in the region of origin. Fol-lowing previous research, I used population-weighted averages from neighbouring countries to proxy for countries missing from the Barro-Lee data set (Engzell & Ichou,2020). This procedure was also used for cases that could only be localized to a region of neighbouring countries speaking the same language.“Spanish” was consequently coded as Spain if the parents were born in Europe, but“Spanish-speaking South America” if they were from South America. Estimates were not sub-stantially affected by the country or country group included (see Figure A1, in supplement). I there-fore only report results including all countries.
Family Contacts
Social capital is the aggregate resources available to a person through their membership of a group or network (Bourdieu,1986). For school-age youth this could mean several things, including peer group attitudes and support from the extended family (Behtoui,2017). However, network resources are commonly measured through the family’s contacts and relations (Lin & Erickson,2008).
To measure social capital in immigrant networks, studies often use proxies such as the average level of education among immigrants from the same country. However, it was recognized early on that real-life groups are not neatly bound by social categories, and that de facto networks often differ substantially from national averages in aspects such as general level of education (Borjas,1995). To account for this, most researchers now use proxies for social capital based on smaller geographical areas, such as neighbourhoods, or self-reported contact networks.
Of these options, contact networks has the advantage of starting from the actual contacts reported by each person instead of constructed ethnic“ingroups”. For this reason, I use family con-tacts as an indicator of social capital, measured using a position generator (Lin & Dumin,1986). The assumption is that ethnicity is one of several structuring factors in social relations (Cederberg,
2012), and that the social position of contacts in the network is a better proxy for network resources than alternatives such as neighbourhood averages.
In the survey, respondents were presented with a list of common occupations and asked to indi-cate which occupations the friends of their family hold. These occupations were assigned a prestige score from Treiman’s SIOPS scores (Ganzeboom & Treiman, 1996). Following the procedure suggested in previous research (Behtoui,2017; Lin,2000), I calculated four indicators of network contacts: contact heterogeneity, meaning the number of different positions among contacts; highest prestige among contacts; range in prestige; and the average contact prestige. These indicators were combined into a single index using factor analysis, to rule out multidimensionality and weigh each indicator according to their contribution. This is the variable called family contacts. The variable is standardized (mean = 0, sd = 1), and positive numbers consequently indicate better than average access to valuable contacts. All calculations were done using Stata.
Results
After excluding cases with missing data, primarily due to missing data from the questions about language spoken at home, the resultant sample contained 960 students from 41 disadvantaged schools. Fifty-three countries or country groups of origin were represented in the sample. After Sweden, the two most frequent were Turkey and Afghanistan. A full list of countries of origin, based on self-reported language and registry data on parents’ region of origin, is presented in
Table A1 (in supplement). To protect the anonymity of respondents, however, distributions are only reported by region of origin.
As seen from the descriptive statistics inTable 1, there were substantial differences in region of origin between families with one and two foreign-born parents. Backgrounds in Europe outside the Nordic countries and in Asia (mostly West Asia considering migration to Sweden) together accounted for almost 95% of backgrounds for families with two foreign-born parents. Meanwhile, a background in the Nordic countries was much more common in families with one foreign-born parent in the sample, and a background in North America, South America or Oceania was exclu-sively—or almost exclusively—reported in families with one foreign-born parent. A background in Africa was equally common in families with both one and two foreign-born parents in the sample. This strong correlation between number of foreign-born parents and their country of birth means a risk of conflating the two factors. However, the separation between students with two, one, or no foreign-born parents is of analytical interest for this study only because of how years of education and education rank are related in the three categories. The uneven dis-tribution of birth regions is therefore not a problem for the purposes of this study, especially con-sidering that the procedure of calculating education rank accounts for each parent’s country or country group of origin.
As expected, university aspirations were most common among students with two foreign-born parents, and least common among students with both parents born in Sweden. The difference was substantial: in the former category more than 70% of youth in the sample replied that they aspired to university, compared to just over half in the other two categories. For comparison, grades were lowest among children with two foreign-born parents.
Grades in the sample were generally above 60%. This is close to the 2014 national average and thus considerably high given the sampling of low GPA schools. The main reason is a higher degree of missing responses among those with the lowest grades.
Parents’ average education was about a year shorter in families with two foreign-born parents than in the other two categories. This corresponds to barely completing upper secondary education, compared to some post-secondary education—a substantial difference in qualifications for the aver-age parent. Seen in relation to the country of origin, the education rank was substantially higher for families with two foreign-born parents. As noted in previous research (Ichou,2014), one measure is not very indicative of the other.
Table 1.Descriptive statistics by foreign and Sweden-born parents.
Two foreign-born parents One foreign-born parent No foreign-born parents University aspirations 71.67% 54.64% 51.69% Grades (percent of max.) 62.23 (18.88) 69.22 (15.03) 65.52 (15.87) Parents’ years of education 11.96 (2.71) 13.30 (2.58) 13.36 (2.51) Parents’ education rank 62.54 (24.42) 53.18 (27.21) 52.52 (26.90) Average contact prestige 14.14 (8.22) 13.82 (8.16) 12.35 (7.97) Family contacts (standardized) 0.21 (1.00) 0.13 (0.98) −0.08 (1.00) Parents expect university 69.33% 49.48% 40.14%
Boys 44.67% 45.36% 56.66%
In secondary education 20.33% 25.77% 38.90% Family region of origin
Sweden – – 100.00%
Nordic excl. Sweden 1.33% 41.24% Europe excl. Nordic 50.00% 26.80% N America 0.00% 6.19% S America 0.67% 12.37% Oceania 0.00% 1.03% Asia 44.33% 8.25% Africa 3.67% 4.12% Observations 300 97 563
Access to high-status family contacts was highest in families with two foreign-born parents and lowest in families with no foreign-born parent. Due to standardization, this is seen as a positive average score for family contacts in the two first categories, and a negative in the lat-ter. The same pattern, even more pronounced, was seen for parents’ expectations. Students with two foreign-born parents reported substantially higher levels of parents’ expectations than students with one foreign-born parent, who reported somewhat higher expectations than children with two Swedish-born parents did. However, boys—who more often favour working over further education—were underrepresented in the sampled families with two foreign-born parents and underrepresented in those with no foreign-born parents. Students who had already transitioned to secondary education were even more unequally distributed. To account for these compositional differences, I controlled for gender and stage of education in all models in the analysis.
The Relationship Between Pre-Migration Status and Social Capital
To explore the association between pre-migration status and ethnic capital, Ifitted OLS (ordinary least squares) regressions between family contacts and the two measures of parents’ education, con-trolling for gender, current stage of education, and school municipality. Since parents’ years of edu-cation and eduedu-cation rank are highly correlated in the non-immigrant families, both measures cannot be included together with immigrant background without risking problems with multicol-linearity. Because of this, Ifitted three separate models. In the first, the two measures of education were included to test their respective strengths as predictors of family contacts. In the second and third models, immigrant background was included together with the separate measures of parents’ education. By comparing the estimates across the latter columns, it is possible to see if there was indeed evidence of better access to family contacts among children of immigrants, and if so, which of the measures of parents’ education could better account for this difference. The results are summarized inTable 2.
As expected from the descriptive statistics, the regression showed that parents’ education rank was a better predictor of access to family contacts than years of education. This indicates that cur-rent status is less important for access to social capital than past status. Like previous research (Nygård & Behtoui,2020), the results showed a substantial positive association between family con-tacts and having two foreign-born parents. However, the magnitude of this association was substan-tially lower (0.16 as opposed to 0.30 for two foreign-born parents) when controlling for parents’
Table 2.OLS regression of family contacts on immigrant background and parents’ education.
(1) (2) (3)
Parents’ years of education −0.01 0.06*** (0.02) (0.01)
Parents’ education rank/10 0.07** 0.06***
(0.02) (0.01)
Swedish background (ref.)
One parent foreign-born 0.16 0.15 (0.10) (0.10) Both parents foreign-born 0.30*** 0.16* (0.08) (0.07)
Constant −0.09 −0.18** −0.14*
(0.06) (0.07) (0.06)
Observations 960 960 960
Adjusted R2 0.048 0.048 0.052 Note: Models control for gender, current education and municipality. Standard errors in parentheses, clustered by school. *p < 0.05, **p < 0.01, ***p < 0.001.
education rank instead of their years of education. The difference in family contacts between families with two foreign-born parents and those with no foreign-born parents could consequently to a large extent be attributed to differences in education rank. This supports the notion that pre-migration status is a major factor for immigrant families’ access to social capital, and thus that selec-tivity and social capital are related.
Predictors of Aspirations
To study the role of selectivity and social capital for educational aspirations, Ifitted LPM (linear probability model) regressions (Mood, 2010) for university aspirations on education rank and family contacts, again controlling for gender, current stage of education, and school municipality. I also tried logit models, including ordered logit models for university aspirations and confidence in completing. This produced substantively similar outcomes (Table A2, in supplement). In the inter-est of brevity, I discuss only the mainfindings here.
In thefirst model, I regressed university aspirations on the two measures of education and on family contacts to test their relative strengths as predictors. In the second model, I regressed uni-versity aspirations on immigrant background and parents’ years of education. The third model instead used parents’ education rank. The first two models should therefore be interpreted as repre-senting the traditional measure of aspirations among children of immigrants (model 2), against the explanation proposed by the selectivity hypothesis (model 3). Finally, in the fourth model, I included both parents’ education rank and family contacts, to test the relative strengths of the two explanations.
The results, summarized inTable 3, showed that education rank was a better predictor of uni-versity aspirations than parents’ years of education. Once education rank was controlled for, the estimate for years of education even turned slightly negative. This indicates that aspirations are more associated with parents’ educational status, than with the length of education per se. The results also showed an association between university aspirations and family contacts, net of parents’ education. In sum, the results map neatly to findings from previous research (Behtoui,
2017; Behtoui & Neergaard,2016; Engzell,2019).
University aspirations were generally higher among children of immigrants. The estimate for how much higher, however, depended strongly on which measures of parents’ education I used. Comparing models 2 and 3 shows that the estimate for having two foreign-born parents was
Table 3.LPM regression of university aspirations on immigrant background, parents’ education and family contacts.
(1) (2) (3) (4)
Parents’ years of education −0.02* 0.04*** (0.01) (0.01)
Parents’ education rank/10 0.06*** 0.04*** 0.04*** (0.01) (0.01) (0.01) Family contacts (standardized) 0.04* 0.03*
(0.02) (0.02)
Swedish background (ref.)
One parent foreign-born 0.00 −0.00 −0.01 (0.04) (0.04) (0.04) Both parents foreign-born 0.21*** 0.11** 0.10* (0.04) (0.04) (0.04) Constant 0.50*** 0.44*** 0.47*** 0.48***
(0.04) (0.05) (0.04) (0.04)
Observations 960 960 960 960
Adjusted R2 0.120 0.107 0.121 0.125 Note: Models control for gender, current education and municipality. Standard errors in parentheses, clustered by school. *p < 0.05, **p < 0.01, ***p < 0.001.
again almost halved (0.11 compared to 0.21 for two foreign-born parents) when parents’ education rank—rather than years of education—was used.
Neither selectivity nor family contacts could fully account for the high educational aspirations among children of immigrants. Still, standardized coefficients show that parents’ education rank was the most important predictor of educational aspirations (β = 0.23) by a large margin, while family contacts (β = 0.07) and foreign-born parents (β = 0.09) had similar magnitudes.
Parents’ University Expectations
As afinal step, I analysed the association between parents’ education, family contacts and parents’ expectations. Dating back to the Wisconsin model (Sewell et al.,1969), expectations from parents and others have been proposed as a key factor in shaping educational aspirations. A wealth of research has since substantiated this claim with empiricalfindings (Haller & Portes,2019). Expec-tations have therefore been proposed as the mechanism linking both selectivity and social capital to high educational aspirations among children of immigrants (Behtoui,2017; Ichou,2014).
To see if I couldfind support for this explanation, I fitted linear probability model regressions of parents’ university expectations on their education, and family contacts. As in the previous analyses, I started by excluding immigrant background to test the other predictors’ relative strength in model 1. In models 2 and 3, I included one measure of parents’ education at a time to see if expectations were higher among immigrant parents and, if so, to what extent this could be explained by their education rank relative to their countries of origin. Finally, in model 4, I added family contacts. As in the previous analyses, I controlled for gender, current stage of education, and school munici-pality in all models. The results are summarized inTable 4.
As seen in column 1, parents’ education rank was a better predictor of parents’ expectations than their years of education. Furthermore, there was a positive association between family contacts and parents’ expectations, net of their education. This gives support to key assumptions in both the selectivity and the social capital hypotheses, namely that parents’ expectations are reflective of both their (past) status and their contact network.
Expectations were markedly higher in families where both parents were foreign born. Compar-ing columns 2 and 3, however, shows that usCompar-ing years of education, instead of education rank, again substantially inflated the difference (0.32 compared to 0.22, for having two foreign-born parents). The results thus suggest that expectations of children’s academic success are high among immigrant parents, but that the large gap is partly a measurement artefact. In model 4, I also included family
Table 4.LPM regression of parents’ university expectations on background, parents’ education and family contacts.
(1) (2) (3) (4)
Parents’ years of education −0.03** 0.04*** (0.01) (0.01)
Parents’ education rank/10 0.07*** 0.04*** 0.04*** (0.01) (0.01) (0.01) Family contacts (standardized) 0.05*** 0.05**
(0.01) (0.01)
Swedish background (ref.)
One parent foreign-born 0.08* 0.07 0.07 (0.04) (0.04) (0.04) Both parents foreign-born 0.32*** 0.22*** 0.21***
(0.04) (0.04) (0.04) Constant 0.44*** 0.34*** 0.37*** 0.38***
(0.04) (0.04) (0.04) (0.04)
Observations 960 960 960 960
Adjusted R2 0.115 0.120 0.130 0.138 Note: Standard errors in parenthesis, clustered by school.
contacts. The model showed a slightly betterfit, but very small changes in the estimates. Contacts and education rank thus appear to be separately associated with parents’ expectations, so that they cannot be reduced to each other. This further supports the interpretation that the two explanations are complementary.
Still, none of the models could fully account for the high expectations among foreign-born parents. This could indicate that other factors, such as the experience and varying anticipation of discrimination or the values assigned to education, must be factored in to fully understand the observed differences. It could also indicate that mobility trajectories must be included more fully, for example through measuring status differences or relevant network resources more directly. Comparing standardized coefficients, however, shows that both parents’ foreign-born status (β = 0.21) and education rank (β = 0.19) were substantially more important predictors of parents’ expec-tations than family contacts (β = 0.09).
Summary and Discussion
In this study, I have studied the relationship between pre-migration status and post-migration social capital, and the extent to which these factors contribute to the high educational aspirations among children of immigrants in disadvantaged schools.
The results showed substantial positive educational selectivity among foreign-born parents, despite comparatively short education in that group. High pre-migration status predicted better access to social capital. This suggests that selectivity is a factor in why immigrant families some-times have more resource-rich networks than non-immigrant families in otherwise similar con-texts, indicating that post-migration resources are influenced by pre-migration resources (Lee & Zhou, 2015).
The results also showed that both pre-migration status and social capital contributed to the higher educational aspirations among children of immigrants than among children of non-immi-grants in disadvantaged Swedish schools, but that pre-migration status was the more important fac-tor. Correcting for the destination country bias by contextualizing educational attainment (Feliciano & Lanuza,2017) thus recasts the problem of educational attainment from a question of why children of immigrants aim high, to why they struggle to realize these aspirations. This refor-mulation forefronts the shortcomings of the Swedish educational system to promote equality in outcomes, urging more research into the processes within and between schools that maintain inequality and those that might counter it. This includes the redistribution of resources, but also the experiences and anticipations of discrimination, and on the availability or lack of compensatory education and“institutional agents” (Stanton-Salazar,2011).
Both selectivity and social capital have been suggested to affect children’s aspirations by increas-ing parents’ expectations. Consistent with theory, I found that both parents’ education rank and family contacts were positively associated with parents’ expectations. However, only education rank could explain some of the higher expectations in families with two foreign-born parents. Parents’ expectations therefore generally seem to come more from their subjective status position than their exposure to aspirational or resource-rich contacts.
In the study, I used family contacts as an indicator for social capital. Since this indicator is focused on occupations it might miss aspects of social capital that are more relevant to school-age youth—including peer groups (Lorenz et al.,2020). Further research is needed to explore extent to which such aspects are also associated with migrant selectivity, for example through classed forms of concerted cultivation (Carolan,2016). Researchers have also noted that migrants exercise agency by creating new forms of migration-specific capital (Erel & Ryan, 2019). Immigrant net-works that are rich in social capital can potentially be understood through this lens, as something that emerges from the interplay between factors relating to the country of origin, the available resources, and the social and institutional contexts in the country of destination. Future research should therefore seek to tap network resources more directly.
Another limitation in this study is the reliance on educational attainments to proxy both pre-migration and current social positions. This leaves many questions unanswered, including a detailed analysis of to what extent the factors associated with access to social capital differ between immigrant and non-immigrant families. Future research should therefore attempt to develop indi-cators of pre-migration status that are not reliant on education, for example through occupation in the country of origin.
By design, the study only covered students in disadvantaged schools. This made it possible to highlight the status ambiguity for immigrant families, where parents on average were less well-edu-cated than in non-immigrant families considering years of education, but more well-eduwell-edu-cated con-sidering education rank. Still, associations might look different in schools that are not disadvantaged, where parents are established in the labour market and university aspiration is a more general norm. Research on representative samples is therefore needed, to study the interplay between pre- and post-migration status and resources more generally, in shaping educational aspirations.
Acknowledgements
Many thanks to the RESL.eu project for kindly providing access to the survey data. Thanks to Alireza Behtoui, Anders Neergaard, Andrey Tibajev and Danuta Lindekrantz for commenting on drafts throughout the research process. Your insights and support have been extremely valuable.
Disclosure Statement
No potential conflict of interest was reported by the author(s).
ORCID
Olav Nygård http://orcid.org/0000-0003-2272-8150
References
Andersson, A. B., Edling, C., & Rydgren, J. (2018). The intersection of class origin and immigration background in structuring social capital: The role of transnational ties. British Journal of Sociology, 69(1), 99–123.https://doi.org/ 10.1111/1468-4446.12289
Andersson, R. (2000). Etnisk och socioekonomisk segregation i Sverige 1990–1998. In J. Palme, Å. Bergmark, E. Näsman, J. Fritzell, L. Sommestad, O. Lundberg & M. Szebehely (Eds.), Arbetsmarknad, Demografi och Segregation, SOU 2000:37 (pp. 223–266). Fritzes Offentliga Publikationer.
Barro, R., & Lee, J.-W. (2013). A new data set of educational attainment in the world, 1950–2010. Journal of Development Economics, 104, 184–198.https://doi.org/10.1016/j.jdeveco.2012.10.001
Behtoui, A. (2017). Social capital and the educational expectations of young people. European Educational Research Journal, 16(4), 487–503.https://doi.org/10.1177/1474904116682248
Behtoui, A., & Neergaard, A. (2016). Social capital and the educational achievement of young people in Sweden. British Journal of Sociology of Education, 37(7), 947–969.https://doi.org/10.1080/01425692.2015.1013086
Berrington, A., Roberts, S., & Tammes, P. (2016). Educational aspirations among UK Young teenagers: Exploring the role of gender, class and ethnicity. British Educational Research Journal, 42(5), 729–755.https://doi.org/10.1002/ berj.3235
Borjas, G. J. (1995). Ethnicity, neighborhoods and human capital externalities. American Economic Review, 85(3), 365–390.https://doi.org/10.3386/w4912
Bourdieu, P. (1986). The forms of capital. In J. G. Richardson (Ed.), Handbook of theory and research for the sociology of education (pp. 241–258). Greenwood.
Brandén, M., & Bygren, M. (2018). School choice and school segregation: Lessons from Sweden’s school voucher sys-tem. In The IAS Working Paper Series. The Institute for Analytical Sociology.
Bunar, N., & Ambrose, A. (2016). Schools, choice and reputation: Local school markets and the distribution of sym-bolic capital in segregated cities. Research in Comparative and International Education, 11(1), 34–51.https://doi. org/10.1177/1745499916631064
Bursell, M., & Jansson, F. (2018). Diversity preferences among employees and ethnoracial workplace segregation. Social Science Research, 74, 62–76.https://doi.org/10.1016/j.ssresearch.2018.03.009
Calleman, C. (2015).“The most open system among OECD countries”: Swedish regulation of labour migration. Nordic Journal of Migration Research, 5(1), 28–35.https://doi.org/10.1515/njmr-2015-0001
Carolan, B. V. (2016). Unequal academic achievement in high school: The mediating roles of concerted cultivation and close friends. British Journal of Sociology of Education, 37(7), 1034–1055.https://doi.org/10.1080/01425692. 2015.1013085
Cederberg, M. (2012). Migrant networks and beyond: Exploring the value of the notion of social capital for making sense of ethnic inequalities. Acta Sociologica, 55(1), 59–72.https://doi.org/10.1177/0001699311427746
CIA. (2019). The world factbook 2019.
Engzell, P. (2019). Aspiration squeeze: The struggle of children to positively selected immigrants. Sociology of Education, 92(1), 83–103.https://doi.org/10.1177/0038040718822573
Engzell, P., & Ichou, M. (2020). Status loss: The burden of positively selected immigrants. International Migration Review, 54(2), 471–495.https://doi.org/10.1177/0197918319850756
Erel, U., & Ryan, L. (2019). Migrant capitals: Proposing a multi-level spatio-temporal analytical framework. Sociology, 53(2), 246–263.https://doi.org/10.1177/0038038518785298
Feliciano, C., & Lanuza, Y. R. (2017). An immigrant paradox? Contextual attainment and intergenerational edu-cational mobility. American Sociological Review, 82(1), 211–241.https://doi.org/10.1177/0003122416684777
Feliciano, C., & Lanuza, Y. R. (2016). The immigrant advantage in adolescent educational expectations. International Migration Review, 50(3), 758–792.https://doi.org/10.1111/imre.12183
Ferguson, J. E., Salominaite, E., & Boersma, K. (2016). Past, present and future: How the Lithuanian diaspora in the Netherlands accumulates human capital from social capital. Journal of Ethnic and Migration Studies, 42(13), 2205– 2225.https://doi.org/10.1080/1369183X.2016.1164590
Ganzeboom, H. B., & Treiman, D. J. (1996). Internationally comparable measures of occupational status for the 1988 International Standard Classification of Occupations. Social Science Research, 25(3), 201–239.https://doi.org/10. 1006/ssre.1996.0010
Guo, S. (2009). Difference, deficiency, and devaluation: Tracing the roots of non-recognition of foreign credentials for immigrant professionals in Canada. Canadian Journal for the Study of Adult Education, 22(1), 37–52. Retrieved fromhttps://cjsae.library.dal.ca/index.php/cjsae/article/view/1002
Gustafsson, B., & Österberg, T. (2018). How are immigrant children in Sweden faring? Mean income, affluence and poverty since the 1980s. Child Indicators Research, 11(1), 329–353.https://doi.org/10.1007/s12187-016-9416-9
Haller, W., & Portes, A. (2019). Class and ambition in the status attainment process: A Spanish replication. The British Journal of Sociology, 70(5), 1825–1849.https://doi.org/10.1111/1468-4446.12695
Hennerdal, P., Malmberg, B., & Andersson, E. K. (2020). Competition and school performance: Swedish school lea-vers from 1991–2012. Scandinavian Journal of Educational Research, 64(1), 70–86. https://doi.org/10.1080/ 00313831.2018.1490814
Holmlund, H., Sjögren, A., & Öckert, B. (2019). Jämlikhet i möjligheter och utfall i den svenska skolan. Bilaga 7. SOU 2019:40 [Equality in opportunity and outcomes in Swedish schools. Appendix 7. SOU 2019:40]. Norstedts Juridik. Ichou, M. (2014). Who they were there: Immigrants’ educational selectivity and their children’s educational
attain-ment. European Sociological Review, 30(6), 750–765.https://doi.org/10.1093/esr/jcu071
Jackson, M., Jonsson, J. O., & Rudolphi, F. (2012). Ethnic inequality in choice-driven education systems: A longitudi-nal study of performance and choice in England and Sweden. Sociology of Education, 85(2), 158–178.https://doi. org/10.1177/0038040711427311
Khachikian, O., & Bandelj, N. (2019). Immigrant socioeconomic mobility in the United States: Mechanisms of inequality and the role of ethnic capital. Sociology Compass, 13(8), 8.https://doi.org/10.1111/soc4.12723
Lee, E. S. (1966). A theory of migration. Demography, 3(1), 47–57.https://doi.org/10.2307/2060063
Lee, J., & Zhou, M. (2015). The Asian American achievement paradox. Russel Sage Foundation.
Lee, J., & Zhou, M. (2017). Why class matters less for Asian-American academic achievement. Journal of Ethnic and Migration Studies, 43(14), 2316–2330.https://doi.org/10.1080/1369183X.2017.1315851
Lin, N. (2000). Social capital: A theory of structure and action. Cambridge University Press.
Lin, N., & Dumin, M. (1986). Access to occupations through social ties. Social Networks, 8(4), 365–385.https://doi. org/10.1016/0378-8733(86)90003-1
Lin, N., & Erickson, B. H. (2008). Theory, measurement, and the research enterprise on social capital. In N. Lin & H. Bonnie (Eds.), Social capital: An international research program (pp. 1–24). Oxford University Press.
Lorenz, G., Boda, Z., Salikutluk, Z., & Jansen, M. (2020). Social influence or selection? Peer effects on the development of adolescents’ educational expectations in Germany. British Journal of Sociology of Education, 41(5), 643–669.
https://doi.org/10.1080/01425692.2020.1763163
Marbach, M., Hainmueller, J., & Hangartner, D. (2018). The long-term impact of employment bans on the economic integration of refugees. Science Advances, 4(9).https://doi.org/10.1126/sciadv.aap9519
Mood, C. (2010). Logistic regression: Why we cannot do what we think we can do, and what we can do about it. European Sociological Review, 26(1), 67–82.https://doi.org/10.1093/esr/jcp006
Nederveen Pieterse, J. (2003). Social capital and migration: Beyond ethnic economies. Ethnicities, 3(1), 29–58.https:// doi.org/10.1177/1468796803003001785
Nygård, O. (2017). Early tracking and immigrant optimism: A comparative study of educational aspirations among students in disadvantaged schools in Sweden and the Netherlands. Comparative Migration Studies, 5(1), 20.
https://doi.org/10.1186/s40878-017-0063-1
Nygård, O., & Behtoui, A. (2020). Access to social capital and educational returns for children of immigrants: Evidence from three Swedish studies. Nordic Journal of Migration Research, 10(2), 50–66.https://doi.org/10. 33134/njmr.248
OECD. (2016). PISA 2015 results: Excellence and equity in education (volume I), Pisa. OECD Publishing. OECD. (2017). International migration outlook 2017. OECD Publishing.
OECD, and EU. (2015). Indicators of immigrant integration 2015: Settling in. OECD Publishing.
Oreopoulos, P. (2011). Why do skilled immigrants struggle in the labor market? Afield experiment with thirteen thousand resumes. American Economic Journal: Economic Policy, 3(4), 148–171.https://doi.org/10.1257/pol.3.4. 148
Polavieja, J. G., Fernández-Reino, M., & Ramos, M. (2018). Are migrants selected on motivational orientations? Selectivity patterns amongst international migrants in Europe. European Sociological Review, 34(5), 570–588.
https://doi.org/10.1093/esr/jcy025
Portes, A., Fernández-Kelly, P., & Haller, W. (2005). Segmented assimilation on the ground: The new second gen-eration in early adulthood. Ethnic and Racial Studies, 28(6), 1000–1040. https://doi.org/10.1080/ 01419870500224117
Portes, A., & Sensenbrenner, J. (1993). Embeddedness and immigration: Notes on the social determinants of econ-omic action. American Journal of Sociology, 98(6), 1320–1350.https://doi.org/10.1086/230191
Razin, A., & Wahba, J. (2015). Welfare magnet hypothesis, fiscal burden, and immigration skill selectivity. Scandinavian Journal of Economics, 117(2), 369–402.https://doi.org/10.1111/sjoe.12092
Seghers, M., Boone, S., & Van Avermaet, P. (2019). Social class and educational decision-making in a choice-driven education system: A mixed-methods study. British Journal of Sociology of Education, 40(5), 696–714.https://doi. org/10.1080/01425692.2019.1581051
Sewell, W. H., Haller, A. O., & Portes, A. (1969). The educational and early occupational attainment process. American Sociological Review, 34(1), 82–92.https://doi.org/10.2307/2092789
Shah, B., Dwyer, C., & Modood, T. (2010). Explaining educational achievement and career aspirations among young British Pakistanis: Mobilizing “ethnic capital”? Sociology, 44(6), 1109–1127. https://doi.org/10.1177/ 0038038510381606
Stanton-Salazar, R. D. (2011). A social capital framework for the study of institutional agents and their role in the empowerment of low-status students and youth. Youth & Society, 43(3), 1066–1109. https://doi.org/10.1177/ 0044118X10382877
Statistics Sweden. (2017a). Olika levnadsvillkor för barn med svensk och utländsk bakgrund [Different living con-ditions for children with Swedish and foreign background].
Statistics Sweden. (2017b). Temarapport 2017:4. Unga utanför? Så har det gått på arbetsmarknaden för 90-talister utan fullföljd gymnasie utbildning [Theme report 2017:4. Young people left behind? The situation on the labour market for those born in the 1990s without completed upper secondary education].
Thaning, M., & Hällsten, M. (2020). The end of dominance? Evaluating measures of socio-economic background in stratification research. European Sociological Review, 36(4), 533–547.https://doi.org/10.1093/esr/jcaa009
Tibajev, A., & Hellgren, C. (2019). The effects of recognition of foreign education for newly arrived immigrants. European Sociological Review, 35(4), 506–521.https://doi.org/10.1093/esr/jcz011
Tomlinson, S. (1989). Ethnicity and educational achievement in Britain. In L. Eldering & J. Kloprogge (Eds.), Different cultures, same school (pp. 15–38). Swets & Zeitlinger.
van de Werfhorst, H. G., & Heath, A. (2019). Selectivity of migration and the educational disadvantages of second-generation immigrants in ten host societies. European Journal of Population, 35(2), 347–378.https://doi.org/10. 1007/s10680-018-9484-2
van Tubergen, F., Maas, I., & Flap, H. (2004). The economic incorporation of immigrants in 18 western societies: Origin, destination, and community effects. American Sociological Review, 69(5), 704–727.https://doi.org/10. 1177/000312240406900505
Voyer, A. (2019).‘If the students don’t come, or if they don’t finish, we don’t get the money.’ principals, immigration, and the organisational logic of school choice in Sweden. Ethnography and Education, 14(4), 448–464.https://doi. org/10.1080/17457823.2018.1445540
Weiss, A. (2005). The transnationalization of social inequality: Conceptualizing social positions on a world scale. Current Sociology, 53(4), 707–728.https://doi.org/10.1177/0011392105052722
Zhou, M. (2005). Ethnicity as social capital: Community-based institutions and embedded networks of social relations. In G. C. Loury, T. Modood, & S. M. Teles (Eds.), Ethnicity, social mobility and public policy: Comparing the US and UK (pp. 131–159). Cambridge University Press.
