Jump to ContentJump to Main Navigation
Unequal AttainmentsEthnic educational inequalities in ten Western countries$

Anthony Heath and Yaël Brinbaum

Print publication date: 2014

Print ISBN-13: 9780197265741

Published to University Press Scholarship Online: May 2015

DOI: 10.5871/bacad/9780197265741.001.0001

Show Summary Details
Page of
date: 27 May 2020

University Completion among the Children of Immigrants

University Completion among the Children of Immigrants

Chapter:
(p.167) 7 University Completion among the Children of Immigrants
Source:
Unequal Attainments
Author(s):

Amy C. Lutz

Publisher:
British Academy
DOI:10.5871/bacad/9780197265741.003.0007

Abstract and Keywords

This chapter examines the success of the second generation in completing tertiary education in Belgium, Canada, England and Wales, France, the Netherlands and the USA (the only countries for which appropriate data were available). The ethnic inequalities found in higher education largely reflect patterns seen earlier in the educational career but there is at the same time clear evidence of progress relative to majority populations. Thus, if we compare the second-generation groups and countries which are covered in both Chapter 3 and Chapter 7, we find that there are thirteen significant ethnic penalties (that is, negative estimates after controls for socioeconomic background) in Chapter 3 (out of a total of twenty-seven that we estimated), but only three in Chapter 7. This is an important and novel finding, suggesting that higher education may give valuable ‘second chances’ to disadvantaged ethnic minorities. This pattern applies in all six countries.

Keywords:   ethnic inequality, second generation, educational career, tertiary education, ethnic penalties, ethnic premia, primary effects, secondary effects

Introduction

THIS CHAPTER EXAMINES DIFFERENCES IN UNIVERSITY DEGREE completion among second-generation ethnic groups. Entry into higher education is one aspect of second-generation integration that has not received a great deal of scholarly attention in Western countries and yet it is the culmination of the educational career, and in most Western countries graduates earn considerably more than non-graduates. Furthermore, it is of very considerable interest to see whether the ambitious choices that many second-generation minorities made when choosing the academic track rather than the vocational track in upper secondary education (as described in Chapter 5) have been matched by subsequent success in completing a university degree. In this chapter we look at second-generation university completion in Belgium, Canada, England and Wales, France, the Netherlands and the USA.

As described in earlier chapters of this volume, a good deal of research has documented the educational disadvantages experienced by the second generation at previous stages of the educational career, such as test scores at the end of compulsory schooling described in Chapter 3. Thus for some groups, although not all, there seem to be negative ‘primary effects’ of ethnicity on educational attainment. However, Chapters 4 and 5 (and other research) have shown that many children of immigrants have relatively high levels of continuation at school and, among those who stay on, many have noticeably high levels of entry to the academic tracks in upper secondary education. In other words, there appear to be positive ‘secondary effects’ of ethnicity (Heath et al. 2008). The primary effects of social background are usually conceptualised to include the effects of the home environment, (p.168) particularly parental stimulation, encouragement and other aspects of socialisation, which contribute to success on achievement tests during the period of compulsory schooling. The secondary effects of social background are those which influence whether, among children of equal achievement, some choose to continue or not. Here it has been customary to think of the decisions that young people and their families make whether to drop out or to continue; whether to enter the academic track and to prepare for university entrance.

Sociologists of education have typically found that a working-class background has both negative primary and negative secondary effects, leading to a cumulation of disadvantage over the course of the academic career (Boudon 1974). In contrast, there is growing evidence that, for at least some ethnic minorities in some countries, negative primary effects of ethnic background are followed by positive secondary effects, thus leading to a mitigation of disadvantage over the educational career (Heath & Brinbaum 2007). These positive secondary effects of ethnicity may well be related, as was argued in Chapter 5, to the finding that immigrant parents are generally positively selected (Feliciano 2006; Bennett & Lutz 2009), and as a consequence may be especially ambitious for their children (Kao & Tienda 1995).

Of particular interest in this chapter is whether the positive secondary effects which Lessard-Phillips and her colleagues reported in Chapter 5 with respect to entry into the academic track are replicated in higher education. Lessard-Phillips and colleagues, it will be remembered, showed that many minorities who experienced ‘ethnic penalties’ with respect to test scores at around the age of 15 or 16 nevertheless had ‘ethnic premia’ with respect to choice of academic track in upper secondary. That is to say, they were significantly more likely to gain access to academic rather than vocational tracks than were members of the majority group from similar socioeconomic backgrounds and with similar test scores.

It will also be of considerable interest to see whether the progress which members of the second generation had made in the comprehensive ‘open-to-all’ upper secondary systems of Canada and the USA (as reported in Chapter 6) are replicated at university level. In short, do we see, either generally or in particular educational systems, a gradual move from ethnic penalties at the end of compulsory schooling to ethnic premia in university completion?

We proceed as follows. In the next section we briefly describe the systems of higher education in our six countries, together with summaries of previous findings in the literature on ethnic inequalities in higher education in each country. In the following section we provide descriptive statistics on the proportion of each second-generation group who have completed tertiary education (p.169) before turning to a discussion of the net coefficients after controlling for parental background. We finally consider how these net coefficients compare with those found at earlier stages of the educational career. (Table A7.1 in the appendix provides details of the datasets used.)

Systems of higher education and research across countries

Canada

The Canadian higher education system is one where universities and colleges are under the domain of the provincial governments, but funding also comes from the federal government (Unesco 1982). No exams are required for university entry. In addition to universities, which grant bachelor's degrees, typically after four years of study (or three in Quebec), Canadian tertiary education also includes colleges that offer 1–4-year programmes ranging from vocational training to programmes that allow students to transfer to a university (Unesco 1982; Lessard-Phillips 2009).

In Canada, the points system and geographic isolation mean that children of immigrants tend to have higher human capital than in other Western immigrant-receiving countries (for details, see Chapter 2 in this volume). Changes in refugee policy over time to restrict refugees who have passed through another qualified country en route to asylum in Canada (Garcia 2006) mean that even refugees in Canada may also be more selected than in some other Western countries. Indeed, Yu & Heath (2007) find that the first generation in Canada tends to be highly educated relative to the second and third generations. Reitz & Zhang noted that, relative to cities in the USA for example, there are ‘fewer immigrants with very low skills’ (2011: 211) in Canadian cities.

As we saw in Chapter 3, many second-generation groups in Canada experience ethnic premia even with respect to test scores at age 15, although ethnic penalties are present for both South European and Caribbean minorities. Several studies have found that children of immigrants and natives in Canada do not experience significant differences in educational attainment in either secondary or tertiary education (see e.g. Davies & Guppy 1998; Boyd 2002; Worswick 2004). We might therefore anticipate that children of immigrants in Canada would have equal or better likelihood to complete higher education relative to the Canadian majority population. Dinovitzer et al. (2003), for example, finds that students with English as a second language are more likely to enroll in university than their peers for whom English is a mother tongue. However, Abada et al. (2009) find that, at the university level, ethnic patterns among children of immigrants have become (p.170) apparent with most Asian groups having high levels of university degree attainment, while black and Filipino students have lower levels of attainment. They note that the categories by which parents enter through Canada's points system may be relevant for understanding youth attainment as those who enter through the Domestic Worker and Live-in Caregiver Programmes, coming primarily from the Philippines and the Caribbean, may witness a devaluation of educational credentials that also creates barriers for their children, while those who enter through the Business Immigrant Programme and other skilled categories may be more advantaged in society and able to form strong social networks and supportive communities of co-ethnics offering better occupational opportunities for their children (see also Abada & Tenkorang 2009).

Given the high net rates of completion of upper secondary education which we saw in Chapter 6, we therefore expect to see ethnic premia among all second-generation groups in Canada, albeit of varying magnitudes, and perhaps with the exception of Caribbeans and Filipinos.

The USA

The USA has a system of both 2-year and 4-year public and private colleges and universities with highly selective elite and non-selective 4-year institutions existing in both the public and private systems. The public system is mostly a state-based system and is partially subsidised directly at the state level. Individual students are subsidised by the federal government usually in the form of low-interest loans and sometimes grants. In addition, a highly controversial for-profit higher education system is emerging which includes institutions that are publicly traded stocks. The cost of higher education to the individual student is very high, even at public universities, and the question of how much debt students should incur to receive a university education is an increasingly important one.

In the USA, children of immigrants born in the USA have an advantage in entry to tertiary education over their foreign-born siblings who may have spent most of their lives in the USA, in that as citizens by birth they are eligible for federal financial aid. This is no small matter because higher education in the United States is exceedingly expensive with 4-year public colleges and universities costing an average of nearly $9000 per year for students from the state where they attend school and $22,000 per year for out-of-state students and 4-year private institutions costing an average of over $30,000 per year not including room and board which adds roughly an additional $10,000 (College Board 2013). To pay for the cost of education, students typically take out loans mostly from the federal government, but also through private lenders. For example, those students who borrowed money for college and completed (p.171) their bachelor's degrees in 2012 carried an average of $29,400 in student loan debt (Institute for College Access and Success 2013). Undocumented students are not only not eligible for federal loans, but (with the exception of 12 states, only three of which offer state aid) as ‘international students’ are also charged the higher out-of-state tuition fee.

Affirmative action programmes at selective institutions of higher education do impact some (but not all) children of immigrants. Espenshade & Radford (2009) find that affirmative action increases the chances of acceptance of black and Latino high school students into selective institutions relative to white and Asian students, but that the same level of enrolment could be attained if achievement gaps could be eliminated between these groups and more advantaged groups. As such, affirmative action programmes at selective colleges may make up for inferior educational opportunity at the lower-level stage. Despite relatively high levels of enrolment in higher education among Asians and at selective institutions, an Asian penalty appears to exist in access to selective institutions in the USA in the sense that, in order to be accepted, Asian American students need to have higher qualifications than students of other ethnic groups (Espenshade & Radford 2009).

Some states—California (1996), Texas (1996), Washington (1998), Florida (2000) and Michigan (2006)—have banned race-based affirmative action in higher education. Texas has substituted a programme that guarantees admission to the top 10 per cent of students in every school to attend the elite state institution of their choice. Since 2003, following the Supreme Court decision in the cases Grutter v Bollinger and Gratz v Bollinger, higher education institutions have by law shifted to affirmative action policies that require that a student's racial-ethnic background be considered only within the context of holistic reviews of the student's application materials rather than points systems that incorporate race directly.

In terms of educational outcomes, we saw considerable diversity between second-generation groups in test scores at age 15 with ethnic premia for the Asian groups but penalties for the second-generation black, Mexican and other Latino groups. But, as in Canada, Chapter 6 showed a trend towards the reduction or elimination of ethnic penalties by the end of upper secondary education. With respect to tertiary education, previous research has suggested that several groups of children of immigrants are also faring better than the native white population in terms of higher education outcomes. For example, black children of immigrants who are high school graduates are more likely to go into selective colleges and universities compared to both their native black and native white peers (Massey et al. 2003; Massey et al. 2007; Bennett & Lutz 2009; Charles et al. 2009). In other cases they appear to be faring much worse.

Latinos in the USA have been less likely to enroll in 4-year bachelor's degree programmes than their native white peers and disproportionately (p.172) attend 2-year community colleges (Swail et al. 2004). Oseguera & Malagon (2011) note that Latinos are also disproportionately enrolling in for-profit institutions, which they find is linked to the suggestions of guidance counselors at the secondary level. In their study of children of immigrants in New York, Kasinitz et al. (2008) find striking ethnic differences in terms of enrolling in 2-year community college versus 4-year bachelor degree programmes and elite versus non-elite 4-year schools.

We are unable in this chapter to pursue this important distinction between elite and non-elite institutions (because of lack of suitable data). We will focus instead on completion of the 4-year bachelor degree programmes, the key question being whether the progress that the second generation demonstrated in upper secondary education is maintained at tertiary level.

England and Wales

Like the American system, the England and Wales system of higher education includes both highly selective, elite universities (such as Oxford and Cambridge) as well as less selective universities. Entrance requirements vary by institution. Universities typically use scores on the A-level exams as the main criteria for entrance with more selective institutions requiring higher scores. As funding for higher education provided by the central government has fallen, universities in England and Wales have increasingly relied on student fees and other funds. Fees in England are now about £9000 for most universities. While in the late 1960s the central government provided 70 per cent of university funds, by the mid-1990s nearly 70 per cent of funds came from other sources (Cheung & Egerton 2007). The level of student fees has become increasingly controversial over time and there are concerns that it might discourage students from less affluent backgrounds from applying.

As in the USA, there was considerable diversity among minorities with respect to net test scores at age 15. However, Chapters 4 and 5 have shown that there are positive ‘secondary effects’ of ethnicity both in terms of continuation into upper secondary education and with respect to track choice, with children of immigrants disproportionately preferring the academic to the vocational tracks (an issue which we were unable to address in the American context). Do we then find that these ambitious choices in upper secondary education are followed by success at university in England and Wales?

Modood (2011) found that during the 1990s, which was an important period of higher education expansion in England and Wales, ethnic minorities increased their enrolments at greater rates than whites. By 2004, ethnic minorities represented about 18 per cent of higher education students in England and Wales, nearly double their proportion of the population (Modood 2011: 186). Shiner & Modood (2002) and Modood (2011) note (p.173) that, although ethnic minorities are over-represented in English and Welsh universities relative to their group size, they tend to be more concentrated in the newer, less prestigious universities which are more welcoming to ethnic minority students than the older, elite universities. In part, this is due to application patterns, but they also find significant ethnic penalties for ethnic minority students in the admissions process at older universities with some groups faring worse than others and with implications for further ethnic disparities in the labour market (see also Boliver 2004). Again, as with the USA, we are unable to pursue issues of elite versus non-elite universities in this chapter (although see Waters et al. (2013) for a comparison of England and Wales with the USA in this respect). Our focus in this chapter will be on completion of full three-year programmes, irrespective of prestige. One issue is whether the ‘Americanisation’ of English and Welsh higher education has led to a similar profile of ethnic penalties and premia as in the USA.

The Netherlands

The Dutch system of higher education has both vocational and university options at the tertiary level. University education since 2002 has included a bachelor-master degree programme whereby the bachelor's degree is attained after 3 years and the master's degree is attained after 1 or 2 more years of study (Rijken et al. 2007). In addition to university education, there are two vocational tertiary degree options including higher vocational colleges (HBOs) and senior vocational education (MBO). Rijken et al. note that the HBOs offer a wide diversity of programmes, including degree programmes that in many other countries are offered within university systems ‘such as journalism, teacher training… and management programs’ (2007: 268). The MBO programmes are more traditional vocational programmes ranging from one to four years to train ‘clerical workers, mechanics, salespersons, and assistant nurses’ (2007: 269). Entry into one of the tertiary options follows a tracking system that begins around age 12, although there is an option to shift from an HBO into a university after one year (2007: 269). The government funds all universities, nearly all HBOs and half of MBOs and students in all three tertiary programmes receive a study allowance from the Dutch government (Rijken et al. 2007).

As we saw in earlier chapters, the main second-generation groups in the Netherlands experience ethnic penalties both with respect to test scores and in continuation into upper secondary education, although there appears to be some diminution of the penalties at completion of upper secondary. The Surinamese and Antillean groups also showed positive secondary effects of ethnicity in track choice: they were more likely to enter the academic track than their peers from the majority group from similar socioeconomic backgrounds. (p.174) It will therefore be important to see whether these positive secondary effects are maintained at tertiary level.

De Valk notes that in the Netherlands, as in much of Europe, ‘migrant children's pathways into adulthood are still a rather unexplored field’ (2006: 13). However, she notes that there are reasons to suspect that children of immigrants may have different transitions to adulthood, particularly related to timing of family formation, with implications for higher education attainment, with native Dutch young adults postponing family formation longer than children of immigrants in the Netherlands. She finds that children of Turkish and Moroccan immigrants would like to marry earlier than native Dutch while Surinamese and Antillean children of immigrants prefer earlier childbearing than the native Dutch.

Crul (2011) notes that second-generation Turks in the Netherlands go on to higher education in greater numbers than in other parts of Europe, with 23 per cent of college-age Turks enrolled in 2001. Thus the educational picture for second-generation Turks in the Netherlands may be a bifurcated one where young people are more likely to drop out prior to completing secondary education, yet are also more likely to complete tertiary education. However, Van de Werfhorst & Van Tubergen (2007) note that within higher education, children of immigrants tend to be represented more in the lower tertiary options than in university.

Belgium

Belgium has both public and private higher education institutions. The universities offer two types of bachelor's degrees: the professional bachelor's degree and the academic bachelor's degree (Huys et al. 2009). Belgium also offers higher education outside of the university (HOBU), typically including three-year programmes that provide professional training to work in a variety of areas ranging from marketing to agricultural studies, teaching, technical education and nursing (Unesco 1982; Duchesne & Nonneman 2000). Belgium has high rates of participation in higher education (relative to the Netherlands, Germany and France) and demand has increased over time for both university and HOBU programmes, although more so in HOBU (Duchesne & Nonneman 2000). More recently, university degrees have been converted to a Bachelor-Master system.

As we saw in earlier chapters, the main second-generation groups experience ethnic penalties at most stages of the secondary school career, although this is somewhat less true of track choice than of continuation or completion. Phalet et al. (2007) find that gaps in educational attainment by ethnicity are greatest at the university level and that Turks, Moroccans and Italians have substantially lower percentages of students who have completed (p.175) a university degree compared to those of native origins. They find that differences in university degree attainment cannot fully be explained by social origins for Turks and Moroccans. Phalet also notes that educational attainment among Moroccans in Belgium is highly uneven, which she relates to their preference for the ‘risky’ strategy of pursuing the higher educational tracks leading towards university, but also potentially towards drop out (2003: 11). In addition, Phalet & Heath (2011) find that second-generation Turkish men and women are both significantly and dramatically less likely to complete higher education relative to their native-origin peers. This, they find, is related to their propensity to move into vocational tracks early on rather than the academic track which would prepare them for tertiary education.2

France

The French higher education system is stratified and includes a relatively open university system as well as highly selective grandes écoles. In addition, France has a system of vocational tertiary programmes with competitive entry. The elite grandes écoles offer five-seven-year programmes, oriented towards science, engineering and business, and are accessible only after being accepted into a two-to three-year preparatory course and receiving a high score on the entrance exams (Givord & Goux 2007; Duru-Bellat et al. 2008). Other universities typically admit any student who has received the bacca-laureat (the full secondary qualification). Students may receive the higher-level tertiary degree, the licence, or a shorter lower tertiary degree, such as DEUG (diplôme d'études universitaires générales), through the universities. Vocational post-secondary options include two-year degrees such as the BTS (p.176) (brevet de technicien supérieur) and DUT (diplôme universitaire de technologies), which are very relevant on the job market. Students from elite backgrounds have greatest access to the grandes écoles, but there is not a strong social class differentiation between university and the short-term vocational programmes, perhaps because even the short-term programmes offer degree options in fields such as engineering and accounting (Duru-Bellat & Kieffer 2008). Tertiary education is almost wholly financed by the French government with the exception of the relatively small but growing private institutions.

As we saw in earlier chapters, in France there were negative primary effects of ethnicity on test scores at age 15 but positive (or neutral) secondary effects with respect to continuation, track choice, and completion of upper secondary education (see also Vallet & Caille 1995; Brinbaum & Werquin 1999; Brinbaum & Kieffer 2009).

Significant efforts have been made in France to democratise the system of secondary education so that greater numbers of students now graduate with school-leaving qualifications (Beaud 2002). This democratisation at the secondary level was accomplished in part by adding lower-status technological and vocational baccalauréat programmes (Givord & Goux 2007). Tracking of children of immigrants within the system of secondary education may mean that educational stratification may be viewed more clearly at tertiary levels of education. Indeed, Duru-Bellat et al. find that access to different higher education options in France is ‘mediated through the different types of baccalauréat’ (2008: 358). Brinbaum & Guégnard argue that ‘While increased access to university and the diversification of tertiary courses have created new opportunities for young people from immigrant families, the inequalities in secondary education could have shifted to higher education’ (2013: 1). They also find social inequalities in access are particularly notable in the case of the elite and highly selective grandes écoles, where few children of immigrants even apply. Givord & Goux (2007) likewise find that over time the grandes écoles have become even more selective by social origin.

Ethnic differences in the completion of higher education

The results in this chapter relate to completion of a full university degree, which in the International Standard Classification of Education (ISCED) would be referred to as category 5A (which involves at least three years' fulltime study) rather than the lower category 5B.3 Within category 5A, however, (p.177) we are unable to distinguish elite from non-elite universities, distinctions which are particularly important in England and Wales, the USA and France. Table 7.1 shows the gross figures (i.e. unadjusted for social background or other variables) for each of the main second-generation groups in each country. In order to ensure that the figures are broadly comparable in each dataset we select respondents in their mid-to-late 20s (who will generally have finished their bachelor's degrees). Note that the denominator includes the whole of the age group, not just those who entered higher education. The percentages thus tell us what proportion of the second-generation age groups secured university degrees (at level 5A).

Table 7.1 shows ethnic differences which closely parallel those from earlier chapters (and from the previous research cited above). Children of East and South Asian immigrants, for the most part, have rates of higher education completion similar to or greater than those of the majority groups across countries. The sole exception, the Bangladeshi second generation in England and Wales, has a rate of tertiary completion nearly seven percentage points less than that of the majority population. East Asians, in particular, have very high levels of university completion, with Chinese in England and Wales, Asians in the USA and Chinese in Canada having levels of higher education completion nearly twice that of the majority populations in those countries. In contrast the West Asian minorities with a background in Turkey have in all cases lower rates of completion.

Second-generation members of European-origin groups show a mixed picture, with some groups exhibiting lower levels of completion than the majority population, such as second-generation ‘other whites’ in England and Wales and Portuguese in France, while some other groups have high levels of completion, such as second-generation ‘other whites’ in the USA (although the latter likely also includes Arabs who usually identify as white).

Likewise, second-generation members of Latin American, Caribbean and African-origin groups also show a mixed picture across countries. The black Caribbean second generation in England and Wales and other Latino groups in the USA have higher rates of university completion than members of the majority groups in those countries, while second-generation Puerto Ricans and Mexicans in the USA exhibit much lower university completion than the majority group. In Canada, university completion rate of the Latin American and Caribbean second generations is similar to that of the majority and the Surinamese and Antillean second generations in the Netherlands have similar (p.178)

Table 7.1. Percentage completing university education

Native-born majority

England and Wales

3rd or higher generation majority group

28.8

USA

3rd or higher generation white

34.1

Canada

3rd or higher generation majority group

27.0

Germany

3rd or higher generation majority group

26.5

Belgium

3rd or higher generation majority group

18.5

Netherlands

Native

42.5

France

Majority population

24.0

Second-generation Asian

England and Wales

Chinese

64.3

USA

Asian

62.1

Canada

Chinese

65.8

England and Wales

Indian

54.7

England and Wales

Pakistani

32.0

England and Wales

Bangladeshi

22.0

Canada

South Asian

51.0

Canada

East and southeast Asian

37.8

Germany

Turkish

6.9

Belgium

Turkish

4.3

Netherlands

Turkish

19.7

Second-generation European

England and Wales

White

17.5

USA

White

49.0

Canada

Western Europe

34.4

Southern Europe

32.3

Germany

Polish

27.3

Former Soviet Union

66.7

(Former) Yugoslav

20.0

Belgium

Italian

5.9

France

Portuguese

18.0

Second-generation Latin American and Caribbean

England and Wales

Black Caribbean

41.7

USA

Mexican

10.6

USA

Puerto Rican

13.0

USA

Other Hispanic

40.8

Canada

Central and Latin American

24.5

Canada

Caribbean

26.8

Netherlands

Surinamese/Antillian

40.3

Second-generation African

England and Wales

Black African

41.4

Canada

Africa

52.2

France

Sub-Saharan African

8.0

France

North African (Magreba)

12.0

Belgium

Moroccan

5.1

Netherlands

Moroccan

25.9

Note. The French percentages come from a cohort of school leavers at all education levels, while the other percentages come from birth or educational cohorts. Thus the percentage of degree completers from France is not directly comparable to those in other countries.

(p.179) university completion rates to the Dutch majority group. However, North Africans in France, the Netherlands and Belgium exhibit lower university completion relative to majorities, while Africans in England and Wales and Canada exhibit very high levels of higher education completion.4

Again, as we saw in Chapter 6, there are apparent differences between countries of destination. In Canada and England and Wales most second-generation groups have higher proportions than their respective majority groups completing university. In Belgium and France, on the other hand, all second-generation groups, including the minority with Italian background in Belgium, have markedly lower proportions than their respective majority groups, while there is a more mixed picture in the Netherlands and the USA.

We next turn to analyses which control for socioeconomic background. Figures 7.17.4 show the results of probit equations. The horizontal label indicates the ethnic origin/generation for the group while the vertical bar indicates the country of residence. The top bars (in black) indicate the significant (p < 0.05) results of the gross effects without controls. Those which are unshaded are non-significant coefficients. The lower grey bars show the significant results (p < 0.05) of probit equations that model the relationship between ethnic group and higher education completion controlling for gender, family composition, parental education and parental occupation (henceforth referred to as ‘social origins’) for various host countries.5

Figure 7.1 presents the coefficients for Asian second-generation students relative to native majorities in a variety of Western destination countries. In two cases, Turks in Belgium and the Netherlands, the second generation faces a gross disadvantage in university completion, which extends net of social origins in the Netherlands but not in Belgium. In the cases of East and Southeast Asians in Canada, and Bangladeshis and Pakistanis in England and Wales, there are no significant gross differences in university completion, yet both Bangladeshis and Pakistanis in England and Wales experience a net advantage relative to the majority population. In the cases of Chinese in Canada and England and Wales, and Indians in England and Wales, the second generation evidences both a gross and a net advantage over their majority groups. (p.180)

University Completion among the Children of Immigrants

Figure 7.1. Probit coefficients, Completion of university education by second-generation Asians relative to native majority population.

(p.181)
University Completion among the Children of Immigrants

Figure 7.2. Probit coefficients, Completion of university education by second-generation Europeans relative to native majority population.

(p.182)

Figure 7.2 presents the coefficients for European second-generation students relative to the majority groups. Only in the case of the Italians in Belgium does a second-generation group evidence both a gross and net disadvantage in university completion. In France, the Portuguese second generation is at a gross disadvantage but net advantage in university completion relative to the native majority population. Both the Southern and Western European second generation in Canada are not significantly different from the white majority in Canada overall, but net of social origins Southern Europeans have an advantage. In England and Wales, the white second generation similarly is overall not significantly different from the native white population, yet net of social origins has an advantage. In the USA, the white second generation has a net advantage in university completion relative to the native white majority.

Figure 7.3 presents the coefficients for African second-generation students relative to majority groups in a variety of Western destination countries. In a few cases, those of Moroccans in the Netherlands and Sub-Saharan Africans in France, the second generation is at both a gross and net disadvantage in university completion. In France, North Africans (from Morocco, Algeria and Tunisia) are at a gross disadvantage, but are not significantly different from native French when social origins are taken into consideration. In Canada, the African second generation is not significantly different from the native majority population in terms of their university completion. Finally, in England and Wales, the black African second generation has a gross advantage over the native majority, but this disappears when social origins are taken into account, highlighting the selectivity of this migration stream.

Figure 7.4 presents the coefficients for the university completion of Latin American and Caribbean second generations in the USA, Canada and England and Wales, the only countries for which this data was available. In the USA, both Mexicans and Puerto Ricans evidence a gross disadvantage that disappears when social origins are considered, indicating that social origins explain the disadvantage of these groups relative to the majority populations. However, for other Hispanic members of the second generation in the USA, there is no gross difference, but a net advantage in university completion compared to native-born majority students. In Canada, Central and Latin American and Caribbean second-generation students are not significantly different from native majorities in their likelihood of college completion. In England and Wales, the black Caribbean second generation has a gross advantage in university completion that becomes non-significant (although of basically the same magnitude) when social origins are controlled.

Across all ethnic groups the models with gross effects indicate that the second-generation groups that have a significantly higher likelihood to complete tertiary education relative to the majority group are Chinese, Indians, black Caribbeans and black Africans in England and Wales, and Chinese and (p.183)

University Completion among the Children of Immigrants

Figure 7.3. Probit coefficients, Completion of university education by second-generation Africans relative to native majority population.

(p.184)
University Completion among the Children of Immigrants

Figure 7.4. Probit coefficients, Completion of university education by second-generation Latin Americans and Caribbeans relative to native majority population.

(p.185) South Asians in Canada. Those groups that fare significantly worse than the native majority in higher education are Mexicans and Puerto Ricans in the USA, and Moroccans and Turks in Belgium and the Netherlands.

The results from the models that control for social origins indicate that in a few countries social origins explain second-generation disadvantage. In the USA, the higher education disadvantage of second-generation Mexicans and Puerto Ricans disappears when controlling for social origins as it does for North Africans in France and Turks in Belgium. But significant net disadvantages remain for second-generation Turks and Moroccans in the Netherlands, for Italians in Belgium and Sub-Saharan Africans in France. In several cases we see a net advantage emerge when controlling for social origins. This is the case for the Pakistani, Bangladeshi and white second-generation groups in England and Wales, for whites in the USA, and for the second-generation Portuguese in France. This indicates that those with similar socioeconomic circumstances to the native majority are more likely to complete higher education than members of the native majority group.

Interpreting differences across the educational career

At the beginning of this chapter we asked whether the positive ‘secondary effects’ of ethnicity on track choice which we had observed for many minorities in Chapter 5 are replicated at university level. We can make these comparisons only for Belgium, England and Wales, France and the Netherlands, since track choice was not distinguished in Canada or the USA. This gives us a total of 16 coefficients which we can compare, as shown in Table 7.2.

In the case of track choice we find nine significant positive coefficients (indicative of ethnic premia in comparison with the respective majority groups) and only one significant negative coefficient (the remainder being non-significant). In the case of university completion, we find three significant negative coefficients and five significant positive ones, so the story on balance is slightly positive although much less favourable than it was for track choice.

We need to remember, of course, that the analysis of track choice excluded respondents who had already dropped out, and the drop outs could in principle account for the less positive story at university completion compared with the story for track choice. This further implies that we may be seeing some degree of bifurcation or polarisation within each ethnic members of the second generation, some individuals going on to university while others leave early with little in the way of formal qualifications. Another possibility is that members of the second generation are somewhat more likely to drop out of university, or take longer to complete their degrees. (p.186)

Table 7.2. Net university completion coefficients compared with net coefficients at earlier stages of the educational career.

Test scores or grades

Continuation

Academic v vocational

Secondary completion

Tertiary completion

France

Portuguese

−0.03

0.22

0.27

0.20

0.08

Maghrebian

−0.12

0.38

0.38

0.32

−0.01

Sub-Saharan African

−0.27

0.03

0.22

−0.04

−0.42

Canada

West European

0.03

−0.07

0.05

South European

−0.17

0.46

0.49

Caribbean

−0.25

0.54

0.07

South Asian

0.13

1.47

0.38

Chinese

0.23

0.02

1.19

East and Southeast Asian

0.24

0.71

0.08

Other

−0.14

0.71

0.44

USA

Other white

−0.20

−0.35

0.60

Mexican

−0.49

−0.08

0.12

Other Latino

−0.41

−0.22

0.56

East Asian

0.39

0.41

0.41

Belgium

North African

−0.54

0.05

−0.28

−0.13

Turkish

−0.77

−0.23

−0.36

−0.23

Other

−0.52

0.29

−0.13

0.08

Mixed

0.25

0.06

−0.18

−0.03

Netherlands

Moroccan

−0.35

−0.26

0.00

−0.16

−0.33

Surinamese & Antillean

−0.16

−0.33

0.27

−0.07

−0.08

Turkish

−0.20

−0.17

0.06

−0.44

−0.34

England and Wales

Black Caribbean

−0.22

0.20

0.10

−0.07

0.35

Black African

−0.01

0.78

0.33

−0.14

−0.03

Indian

0.29

0.69

0.59

0.36

1.05

Pakistani

0.11

0.33

0.32

−0.11

0.58

Bangladeshi

0.39

0.70

0.35

0.50

0.64

Chinese

0.65

0.79

0.86

0.50

1.26

In the cases of Canada and the USA, we cannot compare university completion with track choice. What we can do, however, is to compare the ethnic penalties and premia with respect to test scores at around age 15 with penalties and premia at university completion. This also gives us an overview of the primary and secondary effects of ethnicity. By comparing the first and last stages in this way, we increase the number of comparisons to 27. We find that there were 13 cases of significant ethnic penalties with respect to test scores (p.187) and these are reduced to only three at completion of tertiary education. There is also a slight increase in the number of ethnic premia—up from eight for test scores to 11 for completion. There is, then, considerable evidence of progress vis à vis the majority group.

Finally, do we find greater progress in the more open educational systems of Canada and the USA? There certainly is considerable evidence of progress in both Canada and the USA, mostly of a modest kind with second-generation groups moving from ethnic penalties with respect to test scores to non-significant differences from the majority group at university completion, or from non-significant differences to ethnic premia. But the same pattern is apparent in Belgium, while there are some hints of progress in France and England and Wales, with the Netherlands showing the least evidence of change.

Discussion of findings

The results of these analyses are consistent with the hypothesis that negative primary effects of ethnicity are followed by positive secondary effects, which extend as far as university completion. These positive secondary effects appear to have largely mitigated the ethnic disadvantage seen for test scores. We need to remember, however, the major differentiation within higher education between elite and non-elite institutions, especially in the USA, England and Wales and France, and so major inequalities between minorities and the majority group almost certainly persist among university graduates. Nevertheless, our evidence does indicate that second-generation minorities make considerable progress over the educational career relative to their peers from the majority group. These results echo those of Kasinitz et al. (2009) in highlighting the tremendous educational strides made by children of immigrants across Western countries, while also highlighting the continuing differences between second-generation groups.

We interpret these positive ‘secondary effects’ to be indicative of the high levels of motivation, drive and aspirations of immigrant parents and their children, and perhaps also what might be termed defensive over-qualification in the anticipation that they will need to be more qualified than their majority peers in order to achieve similar labour market success as a result of discrimination in the hiring process (Beaud & Pialoux 1999; Beaud & Beaud-Deschamps 2003; Lessard-Phillips 2009; Modood 2011; Heath et al. 2013). As noted in previous chapters, these high levels of motivation and drive are readily interpretable in the light of the positive selection of the migrant (parental) generation (see Chapter 9 for a more detailed investigation of the effect of selectivity).

(p.188) It is also of considerable interest that these positive secondary effects appear to be present in several of our countries. They are not restricted to the countries with more open, comprehensive systems and mass elite education. This is consistent with the argument advanced by Gibson et al. (2013) that very different educational systems can be compatible with rather similar outcomes. One interpretation would be that highly motivated and ambitious families (whether from the minority or the majority group) will find ways to navigate through the educational system so that they can achieve their objectives. This was earlier suggested as the reason for the failure of educational reforms to empower the working class: highly motivated middle-class parents will find ways to adapt to any new set of rules (Halsey et al. 1980). But it could equally well apply to the second generation too.

References

Bibliography references:

Abada, T. & Tenkorang, E. (2009), 'Pursuit of University Education among the Children of Immigrants in Canada: The Roles of Parental Human Capital and Social Capital, Journal of Youth Studies, 12(2): 185–207.

Abada, T., Feng, H. & Ram, B. (2009), ‘Ethnic Differences in Educational Attainment among the Children of Canadian Immigrants’, Canadian Journal of Sociology, 34(1): 1–28.

Beaud, S. (2002), 80% au bac … et après? Les enfants de la démocratisation scolaire (Paris, La Découverte).

Beaud, S. & Beaud-Deschamps, S. (2003), ‘Regards Croisès sur la Démocratisation Scolaire’, Mouvements, 30: 124–31.

Beaud, S. & Pialoux, M. (1999), Retour sur la Condition Ouvrière: Enquête aux usines Peugeot de Sochaux-Montbeliard (Paris, Fayard).

Bennett, P. & Lutz, A.C. (2009), ‘How African American is the Net Black Advantage? Differences in College Enrollment among Immigrant Blacks, Native Blacks and Whites’, Sociology of Education, 82: 70–100.

Boliver, V. (2004), ‘Widening Participation and Fair Access at the University of Oxford’, Sociology Working Paper 2004–02, http://www.sociology.ox.ac.uk/documents/working-papers/2004/2004-02.pdf

Boudon, R. (1974), Education, Opportunity and Social Inequality: Changing Prospects in Western Society (New York, Wiley).

Boyd, M. (2002), ‘Educational Attainments of Immigrant Offspring: Success or Segmented Assimilation?’ International Migration Review, 36: 1037–60.

Brinbaum, Y. & Guégnard, C. (2013), ‘Choices and Enrolments in French Secondary and Higher Education: Repercussions for Second-generation Immigrants’, Comparative Education Review, 57(3): 481–502.

Brinbaum, Y. & Kieffer, A. (2009), ‘Trajectories of Immigrants' Children in Secondary Education in France: Differentiation and Polarization’, Population-E, 64: 507–54.

Brinbaum, Y. & Werquin, P. (1999), Transition into the Labour Market of Young People from Immigrant Families: Differential Pathways', in D. Raffe, R. Van der Welden & P. Werquin (eds) Education, the Labour Market and Transitions in Youth: Cross-national Perspective, Proceedings of the 1998 European Workshop, (Edinburgh, CES), 147–64.

(p.189) Charles, C.Z., Fischer, M.J. & Mooney, M. (2009), Taming the River: Negotiating the Academic, Financial, and Social Currents in Selective Colleges and Universities (Princeton, NJ, Princeton University Press)

Cheung, S.-Y, & Egerton M. (2007), ‘Great Britain: Higher Education Expansion and Reform—Changing Educational Inequalities’, in Y. Shavit, R. Arum & A. Gamoran (eds), Stratification in Higher Education: A Comparative Study (Stanford, CA, Stanford University Press), 195–219.

College Board Advocacy and Policy Center (2013), Trends in College Pricing 2013, http://trends.collegeboard.org/college-pricing/figures-tables/average-published-undergraduate-charges-sector-2013-14/

Crul, M. (2011), ‘How do Educational Systems Integrate? Integration of Second-generation Turks in Germany, France, the Netherlands and Austria’, in R. Alba & M.C. Waters (eds), The Next Generation: Immigrant Youth in a Comparative Perspective (New York, New York University Press), 269–82.

Davies, S. & Guppy, N. (1998), ‘Race and Canadian Education’, in V. Satzewich (ed.), Racism & Social Inequality in Canada (Toronto, Thompson Educational Publishing Inc.), 131–56.

de Valk, H. (2006), ‘Pathways into Adulthood: A Comparative Study on Family Life Transitions among Migrant and Dutch Youth’, PhD thesis, University of Utrecht.

Dinovitzer, R., Hagan, J. & Parker, P. (2003), ‘Choice and Circumstance: Social Capital and Planful Competence in the Attainments of Immigrant Youth’, Canadian Journal of Sociology, 28: 463–88.

Duchesne, I. & Nonneman, W. (2000), ‘The Demand for Higher Education in Belgium’, Economics of Education Review, 17: 211–18.

Duru-Bellat, M. & Kieffer, A. (2008), ‘From the Baccalaureat to Higher Education in France: Shifting Inequalities’, Population-E, 63: 119–54.

Duru-Bellat, M., Kieffer, A. & Reimer, D. (2008), ‘Patterns of Social Inequalities in Access to Higher Education in France and Germany’, International Journal of Comparative Sociology, 49: 347–68.

Espenshade, T.J. & Radford W.A. (2009), No Longer Separate, Not Yet Equal: Race and Class in Elite College Admission and Campus Life (Princeton, NJ, Princeton University Press).

Feliciano, C. (2006), Unequal Origins: Immigrant Selection and the Education of the Second Generation (New York, LFB Scholarly Publishing).

Garcia, M.C. (2006), Seeking Refuge: Central American Migration to Mexico, the United States, and Canada (Berkeley, CA, University of California Press).

Gibson, M., Carrasco, S., Pàmies, J., Ponferrada, M. & Ríos-Rojas, A. (2013), ‘Different Systems, Similar Results: Youth of Immigrant Origin at School in California and Catalonia’ in R. Alba & J. Holdaway (eds), The Children of Immigrants in School (New York, New York University Press), 84–119.

Givord, P. & Goux D. (2007), ‘France: Mass and Class—Persisting Inequalities in Postsecondary Education’, in Y. Shavit, R. Arum & A. Gamoran (eds). Stratification in Higher Education: A Comparative Study (Stanford, CA, Stanford University Press), 220–39.

Halsey, A.H., Heath, A.F. & Ridge, J.M. (1980), Origins and Destinations: Family, Class and Education in Modern Britain (Oxford, Oxford University Press).

(p.190) Heath, A., & Brinbaum, Y. (2007), ‘Explaining Ethnic Inequalities in Educational Attainment’, Ethnicities, 7: 291–304.

Heath, A.F., Rothon, C. & Kilpi, E. (2008), ‘The Second Generation in Western Europe’, Annual Review of Sociology, 34: 211–35.

Heath, A., Liebig, T. & Simon, P. (2013), ‘Discrimination against Immigrants—Measurement, Incidence and Policy Instruments’, Chapter 4 of OECD International Migration Outlook 2013 (Paris, OECD), 191–230.

Huys, I., Debackere, K. & De Kock, L. (2009), Higher Education in the Flemish Community of Belgium, the French Community of Belgium, Luxembourg and the Netherlands, Expertisecentrum O&O Monitoring van de Vlaamse Gemeenschap in collaboration with the Flemish Ministry of Education and Training.

Institute for College Access and Success (2013), ‘The Project on Student Debt’, http://projectonstudentdebt.org/files/pub/classof2012.pdf/.

Kao, G. & Tienda, M. (1995), ‘Optimism and Achievement: The Educational Performance of Immigrant Youth’, Social Science Quarterly, 76: 1–19.

Kasinitz, P., Mollenkopf, J., Waters, M. & Holdaway, J. (2008), Inheriting the City: The Second Generation Comes of Age (Cambridge, MA, Harvard University Press & Russell Sage Foundation).

Lessard-Phillips, L. (2009) ‘Degrees of Success: The Education of the Second Generation in Canada and Britain’, PhD dissertation, Oxford University.

Massey, D., Charles, C.Z., Lundy, G.F. & Fischer M.J. (2003), The Source of the River: The Social Origins of Freshmen at America's Selective Colleges and Universities (Princeton, NJ, Princeton University Press).

Massey, D., Mooney, M., Charles, C. & Torres, K. (2007), ‘Black Immigrants and Black Natives Attending Selective Colleges and Universities in the United States’, American Journal of Education, 113: 243–71.

Modood, T. (2011), ‘Capitals, Ethnic Identity, and Educational Qualifications’, in R. Alba & M.C. Waters (eds), The Next Generation: Immigrant Youth in a Comparative Perspective (New York, New York University Press), 185–203.

Oseguera, L. & Malagon, M.C. (2011), ‘For-profit Colleges and Universities and the Latino Students who Enroll in Them’, Journal of Hispanic Higher Education, 10: 66–91.

Phalet, K. (2003), ‘Measuring Immigrant Integration: The Case of Belgium’, Studi Emigrazione/Migration Studies, XL: 773–803.

Phalet, K. & Heath, A. (2011), ‘Ethnic Community, Urban Economy, and Second-generation Attainment: Turkish Disadvantage in Belgium’, in R. Alba & M.C. Waters (eds) The Next Generation: Immigrant Youth in a Comparative Perspective (New York, New York University Press), 135–65.

Phalet, K., Deboosere, P. & Bastiaenssen, V. (2007), ‘Old and New Inequalities in Educational Attainment’, Ethnicities, 7: 390–415.

Reitz, J.G. & Zhang, Y. (2011), ‘National and Urban Contexts for the Integration of the Second Generation in the United States and Canada’, in R. Alba & M.C. Waters (eds), The Next Generation: Immigrant Youth in a Comparative Perspective (New York, New York University Press), 207–28

Rijken, S., Maas, I. & Ganzeboom, H.B.G. (2007), ‘The Netherlands: Access to Higher Education—Institutional Arrangements and Inequality of Opportunity’, in Y. Shavit, R. Arum & A. Gamoran (eds) Stratification in Higher Education: A Comparative Study (Stanford, CA, Stanford University Press), 266–93.

(p.191) Shiner, M. & Modood, T. (2002), ‘Help or Hindrance? Higher Education and the Route to Ethnic Equality’, British Journal of Sociology of Education, 23(2): 209–32.

Swail, W.S., Cabrera A.F. & Lee, C. (2004), Latino Youth and the Pathway to College (Stafford, VA, Educational Policy Institute).

Unesco (1982), World Guide to Higher Education: A Comparative Survey of Systems, Degrees and Qualifications (New York, Unipub)

Vallet, L.-A. & Caille, J.-P. (1995). ‘Les carrières scolaires au collège des élèves étrangers ou issus de l'immigration’, Éducation et Formations, 40: 5–14.

Van de Werfhorst, H.G. & Van Tubergen, F. (2007), ‘Ethnicity, Schooling, and Merit in the Netherlands’, Ethnicities, 7: 416–44.

Waters, M.C., Heath, A., Van Tran, C. & Boliver, V (2013), ‘Second-generation Attainment and Inequality: Primary and Secondary Effects on Educational Outcomes in Britain and the U.S.’ in R. Alba, & J. Holdaway (eds), The Children of Immigrants in School (New York, New York University Press), 120–59.

Worswick, C. (2004), ‘Adaptation and Inequality: Children of Immigrants in Canadian Schools’, Canadian Journal of Economics, 37(1): 53–77.

Yu, S. & Heath A. (2007), ‘Inclusion for all but Aboriginals in Canada’ in A.F. Heath & S.-Y. Cheung (eds), Unequal Chances: Ethnic Minorities in Western Labour Markets. Proceedings of the British Academy 137 (Oxford, Oxford University Press), 181–220.

Table A7.1. Data sources for Chapter 7.

Data source

Year of data collection

Population

Sample size for Chapter 7 analyses

Belgium

Belgium linked census data

2001

Residents aged 25–28

39,527

Canada

Ethnic Diversity Survey

2002

Residents aged 22–28

3,960

England and Wales

Longitudinal Survey (linked data from successive censuses)*

1991 (parents) 2001 (students)

Resident population aged 24 to 28

24,285

France

Generation 2004

2007

Representative samle of students who left education in 2004

9,909

Netherlands

Social Integration of Minorities Survey

2006

Persons born 1969 or later

17,420

USA

National Education Longitudinal Study

2000(4th follow up)

Respondents aged25–28

10094

(*) The permission of the Office for National Statistics to use the Longitudinal Study is gratefully acknowledged, as is the help provided by staff of the Centre for Longitudinal Study Information & User Support (CeLSIUS). CeLSIUS is supported by the ESRC Census of Population Programme (Award Ref: ES/K000365/1). The authors alone are responsible for the interpretation of the data.

(p.192)

Notes:

Proceedings of the British Academy, 196, 167–191. © The British Academy 2014

(1) Raya Muttarak kindly prepared the data and ran the analyses for England and Wales. She is a research scholar at the Wittgenstein Centre, Vienna Institute of Demography.

(2) Regional differences also seem to play a role in second-generation Turks' likelihood of continuing on through tertiary education. In Brussels, second-generation Turks are more likely to drop out before completing secondary education than in other parts of Belgium, but also more likely (together with those in the south) to complete tertiary education (Phalet et al. 2007). Phalet & Heath (2011) believe that the greater likelihood to pursue tertiary education in Brussels may be related to the availability of local universities and professional career opportunities. Phalet (2003) also notes that different regions in Belgium attract migrants with different educational qualifications; the most highly educated immigrants settle in Belgium, which may have implications for the educational attainment of the second generation. Phalet et al. (2007) also find that while sharing the same educational system, different parts of Belgium have different traditions within secondary schooling that may relate to these regional differences in entry into tertiary education. In Dutch-speaking schools in Flanders, children of immigrants who are struggling are more likely to be oriented towards vocational training, while in the French-speaking schools of Wallonia, students are more likely to be held back when they are struggling. The researchers find that these practices have different implications for tertiary education. Children of immigrants coming from French-speaking schools are more likely to go on to tertiary education than those in Dutch-speaking schools who are more likely to be in vocational programmes, but they are also more likely to drop out of secondary without a terminal degree.

(3) For Belgium, we also analysed all tertiary degree completion as an outcome including both university degree and vocational higher education. Belgium's extensive vocational higher education system includes some fields of study that exist in university systems of other countries such as the USA. The programmes, which took 3–4 years to complete using data from 2001 prior to the conversion to the BA-MA system, enrolled greater proportions of native majority students than universities. If we use completion of any tertiary degree, we do get some different outcomes. Second-generation Moroccans and Turks stand out as facing ethnic penalties not explained by social origins, while the disadvantage of Italians in higher education is explained by social origins.

(4) Although not shown here, as a whole, mixed-generational groups have percentages of university completion at similar to or higher than the majority groups.

(5) In all cases except France, these analyses are done with birth or educational cohort data. The French data is different in that it is a study of school leavers at all educational levels. Also, the French data does not allow for the inclusion of a control for family composition or parental education. Thus, the French coefficients should be interpreted with some caution in terms of comparability with other countries. In the case of England and Wales, the net coefficients for the other white, Bangladeshi, black Caribbean and black African groups are for men.